18. Qualitätsindikatoren – Lungenkarzinom

kein statistisch signifikanter Unterschied von Lorlatinib gegenüber Brigatinib

keine hinreichende Ähnlichkeit bei der Operationalisierung

keine hinreichende Ähnlichkeit bei der Operationalisierung

keine hinreichende Ähnlichkeit bei der Operationalisierung

Barlesi 2015

(22, 31)

08/2014 to 12/2017

Frankreich

RCT

efficacy of gemcitabine plus cisplatin with docetaxel plus cisplatin in completelly resected  NSCLC on quality of life

Inclusion Criteria:

· 18–75 yrs patients with completely resected (R0) Stage IB–III NSCLC, ECOG 0 or 1

· adequate bone marrow reserve and organ function, including calculated creatinine clearance of 45 ml/min

Exclusion Criteria:

· postoperative complications (acute respiratory distress syndrome, bronchial fistula or severe pneumonia)

· previous history of cancer within ≤ 5 years or clinically significant cardiac dysfunction, active infection or neurological/ psychiatric disorders

n=136

74% males

median age 57 (36-71)

Stage IB/II/III: 32/34/34%

bi(lobectomy): 85%

adenocarcinoma/Squamous cell carcinoma/others: 55/23/22%

minor postoperative complications: 15%

within 8 wks after surgery:

Cisplatin (75 mg/m², d1) plus gemcitabine (1250 mg/m², d1 and d8) (n=67)

vs.

Cisplatin (75 mg/m², d1) docetaxel (75 mg/m² d1) (n=69)

for three cycles (9 weeks)

Primary:

Quality of life (EORTC QLQ-C30, QLQ-LC12: raw scores from each functional and symptom scale were translated onto an overall scale from 0 to 100,.higher functional scores are indicative of global health scores better QoL, whereas lower symptom and single item scores reflected fewer symptoms)

Secondary:

overall survival, safety and cost

Quality of life:

·  (GHS) slightly improved from 63.5 vs. 62.7 at baseline to 64.5 vs. 65.4 after 3 months with no differences between groups (p=0.8)

· no significant differences for other functional scores between  groups

· Social function was not altered by the number of infusions (Days 1 and 8 for the GC arm)

· no differences for symptom scores, except for alopecia, where the CG showed a significantly higher score compared to IG

Safety:

· satisfactory compliance (80.6 vs. 85.5 % completed all the 3 planned cycles of CT)

· 32.8 vs.  21.7% experienced Grade 3/4 haematological toxicities with no significant difference between arms:

-    anaemia: 1.5 vs. 0%

-    neutropenia: 28.4 vs. 21.7 %

-    Thrombopenia: 9.0 vs. 1.4 %

· 32.8 vs. 26.1 % experienced non-haemotological toxicities:

-    asthenia: 7.5 vs. 7.2 %

-    nausea/vomiting: 10.4 vs. 7.2 %

-    infection: 3.0 vs. 1.4 %

-    neuropathy: 0 vs. 1.4 %

-    alopecia: fewer with IG: 3.0 vs. 13.0 % (p=0.03)

Overall survival:

· no differences between groups: median follow-up: 20.2 months, 1-year survival: 100 vs. 96.8% 2-year survival: 92.9 vs. 89.8%

· 15 (7 vs. 8) patients died in the course of study

2b

Abwertung aufgrund von Studienlimitationen, geringer Präzision und Indirektheit

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: „In conclusion, DC or GC adjuvant chemotherapy for completely resected NSCLC might be an acceptable alternative to currently recommended combinations and might be compared with the VC regimen. This study did not demonstrate any significant negative QoL impact of DC and GC in resected lung cancer patients, and might be useful when discussing treatment options with such patients.“

Schlussfolgerung der Begutachterin: Multizentrische, in Frankreich durchgeführte Studie unter Einschluss von 136 NSCLC Patienten (Grad 1B-III, daher Abwertung aufgrund von Indirektheit) zur Bestimmung des optimalen Regimes einer CT nach vollständiger Resektion. Es fehlt eine nähere Beschreibung der Randomisation und Verblindung (z.B. in der Erfassung der Lebensqualität). Es existiert aber ein Protokoll in französischer Sprache (kein Zugriff auf Volltext). Die Studie wurde auf Empfehlung des Steering-Kommites vorzeitig aufgrund neuer ASCO-Empfehlungen für 4 (statt der hier untersuchten 3) Zyklen CT gestoppt (daher Abwertung aufgrund geringer Präzision). Ergebnisse zur Lebensqualität nach 93 Monaten liegen für 71% der Patienten vor, bei Gesamtüberleben werden in der Grafik viele Drop-outs sichtbar, welche nicht beschrieben werden (Abwertung aufgrund von schwerwiegenden Studienlimitationen), die Studie wurde von Lilly Oncology, Sanofi-Aventin, Amgen und mit öffentlichen Mitteln unterstützt.

Burdett 2015

(21)

Search from 1995 to 12/2013

Systematic Review

Effects of CT following surgery, or following surgery plus RT in patients with histologically diagnosed early stage NSCLC on OS, time to locoregional recurrence, time to distant recurrence and recurrence-free survival

Investigation of predefined patient subgroups benefit from cisplatin-based CT

Inclusion Criteria:

·  RCTs that compared OS, time to locoregional recurrence, time to distant recurrence and recurrence-free survival of additional adjuvant CT in patients with histologically diagnosed early stage NSCLC

·  published and unpublished completed trials with proper randomised

·  patient had undergone a potentially curative resection and not received previous CT

Exclusion criteria:

·  trials using long-term alkylating agents for >1 year

Surgery trials: mostly men with a median age of 61 (18-84) yrs

tended to have good performance status and

tumours of stage I-II adenocarcinomas or

squamous cell carcinomas.

Surgery+RT trials: mostly men, good performance status, median age of 59 yrs (range 27 to 81), stage III, squamous carcinomas

stage IIIA, IIB and IV (combined as stage III)

Risk of bias: inclusion of trials with adequate methods of randomization, no blinding, but OS is unlikely to be influenced, low reporting bias due to analysis of individual patient data

Surgery plus adjuvant CT:

o platinum plus

-vinca alkaloid/ etoposide

-or vinorelbine

-or taxane

o platinum plus vinca alkaloid + tegafur and uracil/tegafur other platinum regimens

o tegafur and uracil/tegafur plus other agent

o tegafur and uracil/tegafur

vs.

Surgery

median follow-up: 5.5 yrs

Surgery (plus RT)

vs.

Surgery (plus RCT) plus adjuvant CT:

o platinum plus vinca alkaloid/ etoposide

o platinum plus vinorelbine

o other platinum regimens

o antimetabolic agent only

median follow-up: 6.4 yrs

Primary:

OS

Secondary:

time to locoregional recurrence

time to distant recurrence

recurrence-free survival

Overall survival:

Surgery+adjuvant CT vs. surgery (N=35, N= 26 with individual patient data; IPD, n=8447 with 3223 deaths):

·    benefit of adding CT after surgery (HR 0.86; 95% CI 0.81 to 0.92; p< 0.0001) with small heterogeneity between results of different studies (I²=4%)

·    no influence of age group, sex, histology, PS and stage of the disease on the effect, for stage III survival was improved

·    with differences depending on CT category:

-     Platinum plus vinca alkaloid/ etoposide (N=9, n=2404): no difference: HR 0.94; 95%CI 0.84 to 1.05)

-     Platinum+vinorelbine (N=4, n=1304): benefit: HR 0.82; 95%CI 0.70 to 0.97)

-     Platinum+taxane (N=1, n=244): no difference: HR 0.77; 95%CI 0.57 to 1.05)

-     Other platinum regimes (N=4; n=699): no difference: HR 0.90; 95%CI 0.72 to 1.13)

-     platinum plus vinca alkaloid + tegafur and uracil/tegafur (N=8, n=1375): benefit: HR 0.79; 95%CI 0.67 to 0.93)

-     Tegafur and uracil/tegafur+other agent (N=1, n=83): harm: HR 1.79; 95%CI 1.00 to 3.20)

-     Tegafur and uracil/tegafur (N=7; n=2390): benefit: HR 0.76; 95%CI 0.64 to 0.90

·    absolute increase in survival of 4% (95%CI 3 to 6) at five years (5-yrs OS increased from 60 to 64%)

·    with no influence of subgroup analyses by age, sex, histological cell type, tumour stage, and performance status

·    for stage III, 5-year survival was improved by 5 % (95 % CI 3 to 8) from 30 to 25 %

·     

Surgery plus RT+ adjuvant CT vs. surgery+RT (N=15, N= 12 with IPD, n=2660 with 1909 deaths):

·    benefit of adding CT after surgery+RT (HR 0.88; 95% CI 0.81 to 0.97; p= 0.0009) with small heterogeneity between results of different studies (I²=0%)

·    no evidence of a differential effect by CT category,  the extent of resection achieved or timing of CT in relation to RT

·    absolute increase in survival of 4% at five years (5-yrs OS increased from 29 to 33%)

·    no influence of subgroup analyses by age, sex, histological cell type, tumour stage, and performance status

Recurrence-free survival

Surgery+adjuvant CT vs. surgery (n=5379 with 2519 events):

benefit of adding CT after surgery (HR 0.83; 95% CI 0.77 to 0.90; p< 0.0001)

Surgery+RT+adjuvant CT vs. surgery+RT (n=2247 with 1673 events):

benefit of adding CT after surgery (HR 0.85; 95% CI 0.77 to 0.93; p=0.0006)

Time to locoregional recurrence

Surgery+adjuvant CT vs. surgery (n=5226 with 936 events):

benefit of adding CT after surgery (HR 0.75; 95% CI 0.66 to 0.94; p< 0.0001)

Surgery+RT+adjuvant CT vs. surgery+RT (n=2247 with 533 events):

benefit of adding CT after surgery (HR 0.79; 95% CI 0.67 to 0.85; p= 0.0008)

Time to distant recurrence

Surgery vs. surgery+adjuvant CT (n=5224 with 1267 events):

benefit of adding CT after surgery (HR 0.80; 95% CI 0.72 to 0.89; p= 0.0007)

Surgery vs. surgery+adjuvant CT (n=2247 with 533 events):

benefit of adding CT after surgery (HR 0.75; 95% CI 0.66 to 0.87; p< 0.0001)

1a

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Results from 47 trial comparisons and 11,107 patients demonstrate the clear benefit of adjuvant chemotherapy for these patients, irrespective of whether chemotherapy was given in addition to surgery or surgery plus radiotherapy.

These systematic reviews and meta-analyses use individual participant data, which is considered the gold standard of this type of review. We included all eligible trials if possible, no matter what language they were published in or whether they were published or not. The first meta-analysis (surgery versus surgery plus adjuvant chemotherapy) included 92% of all patients in eligible trials and the second meta-analysis (surgery plus radiotherapy versus surgery plus radiotherapy plus adjuvant chemotherapy) included 86% of all patients in eligible trials.

We are confident that further research is unlikely to change the findings. The studies were well designed and conducted, address the review question, and the effects are consistent across trials. The impact of any data we have not been able to include in our analyses is small.”

Schlussfolgerung der Begutachterin: Systematische Übersicht, geringfügige Einschränkung der Qualität (keine Untersuchungen zum Publikationsbias, aber Suche in Registern laufender Studien und Konferenzbänden und Kontaktierung von Autoren), basiert auf 47 Vergleichen in Studien unter Einschluss von 11 107 Patienten aller Stadien (geringe Anzahl im Stadium IIIB und IV) mit NSCLC mit einer Auswertung von individuellen Patientendaten und in der Regel geringem Verzerrungspotential (fehlende Verblindung beeinflusst den Hauptendpunkt Gesamtüberleben kaum).   Es wird die Wirksamkeit von adjuvanten Chemotherapien zusätzlich zur Operation oder zu Operation und Strahlentherapie berichtet und es werden Ergebnisse aus Subgruppenanalysen (auch für Stadien ) berichtet. Die Suche erfasste Studien mit einem Publikationsdatum bis Dezember 2013, die Autoren gehen davon aus, dass zukünftige Studien die Behandlungseffekte nicht beeinflussen werden.

Hata 2017

(23)

04/2005 to 01/2012

Japan

RCT

feasibility of the conventional treatment schedule of S-1 and the shorter treatment schedule of S-1 as adjuvant treatment of patients with completely resected NSCLC

Inclusion Criteria:

· histologically confirmed primary lung adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and adenosquamous carcinoma;

· R0 resection of the primary tumor; pathological stage IB to IIIA disease

· patients aged 20-74 yrs; ECOG 0 or 1;

· adequate organ function able to start within 9 weeks after surgery;

· no prior therapy

Exclusion criteria:

· history of previous CT, RT or surgery for lung cancer;

· pulmonary fibrosis; pleural effusion, ascites, or cardiac effusion that required drainage; concomitant malignancy; significant comorbidity; diarrhea; pregnancy;

· desiring to have children;

· drug allergy to S-1 or any of its components

n=80 randomized, 78 analysed

age: 63±8 yrs

males: 65%

ECOG 0/1: 94/6 %

Adeno/squamous/large cell carcinoma: 64/29/5 %

Stage IB/IIA/IIB/IIIA: 78/1/4/4 %

Lobectomy: 96 %

within 9 wks after surgery:

conventional treatment schedule of S-1 (n=39): administered for 4 weeks followed by a 2-week rest

vs.

shorter treatment schedule of S-1 (n=41): administered for 2 weeks followed by a 1-week rest

over 12 monthsunless there was any evidence of recurrence, other malignancies, or SAEs 

Primary:

rates of completing the planned administration schedule over 12 months

Secondary:

relative total administration dose of S-1

toxicity

3-year disease-free survival (DFS)

Median follow-up: 64 (range 6-113) months

Completion rates (over 12 months):

· higher in IG: 73.7% (95% CI 58.0%–85.0%) vs. 45.0% (95% CI: 30.7%–60.2%) (p=0.001)

· 73.7% vs. 45.0% received S-1 administration according the planned schedule

· S-1 administration was halted because of

-    AEs or refusal for 18% (n = 6 AEs, n= 1 refusal) vs. 38%) (n = 9 AEs, n= 6 refusal).

-    tumor recurrence or other non-S-1-related complications for 8% (n =1 tumor recurrence, n = 2 non-S-1-related complications) vs. 18% (n = 4 tumor recurrence, n =3 non-S1-related complications)

· higher averages of the relative dose intensity over 12 months in IG:  77.2 % vs.  58.4% (p = 0.01)

Adverse events:

· primary adverse events were hematological, gastrointestinal, and cutaneous signs and symptoms

· no difference in AEs: 38 (100%) (grade 1/2 in 89% and grade 3 in 11%) vs. 39 (98%) (grade 1/2 in 93% and grade 3 in 5%; p = 0.42)

· no difference in SAEs grade 3: 4 (11%) (elevated bilirubin, neutropenia, and rash) vs. 2 (5%) (anorexia and nausea, p = 0.43)

· Elevated bilirubin, AST, ALT, and alkaline phosphatase levels were more frequent in IG vs. CG (p = 0.01, <0.01, 0.01, <0.01, respectively).

· Two patients, 1 in each died during drug administration period, although the causes death were unknown and were not considered to be related to S-1 administration

Disease-free survival and recurrance:

· No difference in 3-year DFS rates: 79.0% vs. 79.3% (p=0.94)

· 9 (23.7%) vs. 8 (20.0%) relapsed within 3 years with predominant locoregional recurrence in both arms (6 vs. 5)

2b

Abwertung aufgrund von Studienlimitationen, geringer Präzision und Indirektheit

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: „The superiority of feasibility of the shorter schedule was not recognized in the present study. The conventional schedule showed higher completion rates over 12 months (p = 0.01) and relative dose intensity of S-1 (p = 0.01). Toxicity showed no significant difference among the shorter schedule and the conventional schedule, except for grade 1–3 elevation of bilirubin.“

Schlussfolgerung der Begutachterin: Multizentrische, in Japan durchgeführte Machbarkeitsstudie unter Einschluss von 80 NSCLC Patienten (Grad 1B-IIIA, daher Abwertung aufgrund von Indirektheit und fehlender Präzision) zum Vergleich von 2 Regimes zur Verabreichung einer Chemotherapie auf der Grundlage von S-1 mit zentraler Randomisierung, aber ohne Beschreibung der Randomisierungsordnung, ohne Verblindung oder verblindete Endpunkterfassung und Registrierung des Protokolls (daher Abwertung aufgrund von schwerwiegenden Studienlimitationen). Die Studie wurde von JSPS KAKENHI unterstützt (kein Einfluss auf Datensammlung und –analyse, Interpretation der Ergebnisse und Schreiben der Veröffentlichung.

Iwamoto 2015

(24)

Japan

 09/2007 to 12/2009 (Japan)

RCT

Efficacy and safety of S-1 versus cisplatin +S-1 in patients with completely resected stage II and IIIA non–small cell lung cancer

Inclusion criteria:

· adults (20-74 yrs)

· completely resected NSCLC (stage II or IIIA, metastasis to a single mediastinal lymph node only)

· ECOG PS 0 or 1

· Adequate bone marrow and organ function

Exclusion criteria:

· drug allergies, interstitial pneumonia, concomitant malignancies

· heart failure, uncontrolled diabetes mellitus, and active infections.

· Pneumonectomy

n=200

median age: 62 years

males: 76%

IIA/B/III: 23/34/43 %

Adeno/Squamous/other carcinoma: 67/23/10%

Lobectomy: 100%

ND0-1 /ND2 lymph node dissection: 5/95 %

within 8 weeks after surger:

Longterm S-1 (40 mg/m² twice per day) for 2 consecutive weeks repeated every 3 weeks for 1 year (n=100)

vs.

Cisplatin/S-1 (60 mg/m²) d1 plus S-1 (40 mg/m² twice per day) for 2 consecutive weeks repeated every 3 weeks for four cycles (n=100)

Primary:

relapse-free survival (RFS) at 2 years and identification of predictive biomarkers* Secondary:

OS

rate of AEs  (NCI-CTC 3.0)

medical

treatment completion rate

relapse

*Not extracted

Feasibility:

· 52.6 % received 15 cycles of adjuvant CT vs. 74.7% completed four cycles of adjuvant CT as planned

· High median relative dose intensities (85.2 vs. 86.4 and 94.1%

Survival:

· Comparable 2-year RFS rate: 65.6% (95% CI 55.3–74.0%) vs. 58.1% (95% CI 47.7–67.2); HR 0.90 (95%CI 0.59–1.37).

· comparable 5-year OS rates: 72.6% (95% CI, 64.3–82.0) vs. 72.2% (95% CI, 63.8–81.7)

Safety:

· anemia or neutropenia (grade 3 or 4): lower in IG (anemia: 1.0% vs. 8.4% (p=0.018); neutropenia: 13.4% vs. 27.4% (p=0.02)

· nonhematologic toxicities (grade 3 or 4): anorexia (anorexia: 2.1 vs. 9.5%; p=0.032) and nausea (0 vs. 6.3%; p=0.014) were lower in IG

· but: more grade 1 or 2 skin rash with IG (26.8% vs.9.5; p=0.002)

· 17 vs. 18 deaths (not treatment related)

2b

Abwertung aufgrund von geringer Präzision und Indirektheit

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: „In conclusion, this study suggests that adjuvant long-term S-1 monotherapy or CDDP+S-1 for completely resected stage II–III

NSCLC is a viable alternative to cisplatin doublet chemotherapy in terms of efficacy and toxicity. UMPS/OPRT expression may be a useful marker for identifying patients who would benefit most from adjuvant long-term S-1, although the fact that we did not adjust for multiple testing may limit the relevance of this finding. „

Schlussfolgerung der Begutachterin: Multizentrische, in Japan durchgeführte Phase II unter Einschluss von 200 Patienten mit NSCLC nach vollständiger Resektion (Grad IIA/B-IIIA) (daher Abwertung aufgrund von Indirektheit) mit adäquater Beschreibung der Randomisationgszuweisung, aber ohne verblindete Messpunkterhebung und ohne Zugriff auf das Protokoll (trotz Angabe der Registrierungsnummer) mit vollständiger Auswertung aller Patienten nach 2 Jahren zeigt Machbarkeit, Wirksamkeit und Sicherheit einer Langzeit-Monotherapie mit S-1 bei geringen Nebenwirkungen. Abwertung aufgrund von fehlender Präzision (geringe Fallzahl) und Indirektheit (43% der Patienten waren im Stadium III, Studie wurde vollständig in Japan durchgeführt).  Die Studie wurde von der West Japan Oncology Group unterstützt, mehrere Autoren berichten Interessenkonflikte (Forschungsförderung und Honorare durch pharmaceutische Firmen).

Kenmotsu 2017

(25)

03/2012-08/2016

Japan

RCT

efficacy of pemetrexed plus cisplatin versus vinorelbine plus cisplatin as postoperative adjuvant chemotherapy in patients with pathologic stage II-IIIA nonsquamous NSCLC

Inclusion criteria:

·       patients who underwent complete surgical resections by lobectomy or pneumonectomy with resection of any involved N2 lymph nodes within 3-8 weeks before enrollment.

·       patients were age 20-75 years with histologically confirmed pathologic stage II or IIIA nonsquamous NSCLC, proven results of the EGFR gene mutation test, ECOG 0 or 1, and adequate hematologic and organ function.

Exclusion criteria:

·  Severe postoperative complications, interstitial pneumonia on computed tomography of the chest, current pregnancy, and other severe comorbidity.

·  Patients who received neoadjuvant CT or had planned to receive EGFR tyrosine kinase inhibitors

n=804, 784 treated

males: 58.9 %

ECOG 0: 76.7 

IIA/IIB/IIIA: 33.9/13.8/52.3 %

Adenocarcinoma/others: 95.9/4.1 %

EGFR pos: 24.5 %

Lobectomy: 98.5 %

vinorelbine (25 mg/m², days 1 and 8) plus cisplatin (80 mg/m², day 1) (n=402, n=287 (71.4 %) completed treatment)

vs.

pemetrexed (500 mg/m², day 1) plus cisplatin (75 mg/m², day 1) (n=402 randomised, n=342 (85.0 %) completed treatment)

every 3 wks for 4 cycles

Primary: overall survival (changed to recurrence-free survival in an protocol amendment)

Secondary:  recurrence-free survival

rate of treatment completion

toxicity (NCI-CTC)

for up to 30 days beyond the last dose of any protocol treatment.

recurrence-free survival (RFS):

Disease recurrence or death in 53% vs. 51%

Median RFS: no difference: 37.3 months (95% CI 28.8 to 52.5) vs. 38.9 months (95% CI, 28.7 to 55.3 months), HR 0.98 (95% CI, 0.81 to 1.20),

With no differences depending on pathologic stge (stage IIIA:  1.05 (95%CI 0.82 to 1.34)

2-year RFS: 60.7% (95% CI, 55.7% to 65.3%) vs. 58.3% (95% CI, 53.2% to 63.0%)

3-year RFS: 50.2% (95% CI, 45.0% to 55.2%) vs. 51.1% (95% CI, 45.8% to 56.0%)

Overall survival (OS):

75 vs. 71 patients had died at time of analysis

Median OD: not reached, HR 0.98 (95% CI, 0.71 to 1.35)

3-year OS: 83.5% (95% CI, 79.2% to 87.0%) vs. 87.2% (95% CI,

83.8% to 90.2%)

Adverse events:

any grade AE: 100 vs. 99.7 %

grade 3-5 AE: 89.4% vs. 47.4%

Grade 3-4 febrile neutropenia: 11.6% vs. 0.3% (p<0.01)

Grade 3-4 WBC count decrease, neutrophil count decrease,

and anemia: 51.0%, 81.1 % and 9.3 % vs. 5.9%, 22.7% and 2.8% (p<0.01)

grade 4 WBC count decrease and neutrophil count decrease:  9.3% and 56.6% vs. 0.3% and 3.3%

One treatment-related death in each arm by sudden death and pneumonitis

1b-

Abwertung aufgrund von Indirektheit

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In conclusion, although the JIPANG study failed to show the superiority of pemetrexed plus cisplatin in terms of recurrence-free survival, toxicity profiles favored this regimen. Therefore, this combination can be an option for postoperative adjuvant chemotherapy in patients with completely resected nonsquamous NSCLC. The results of the JIPANG study indicate that the optimal platinum-based chemotherapy is still unclear in the adjuvant chemotherapy setting for resected nonsquamous NSCLC.”

Schlussfolgerung der Begutachterin: Multizentrische, in Japan durchgeführte Phase III unter Einschluss von 904 Patienten mit NSCLC (Grad II-IIA, >50 % im Grad IIIA) nach vollständiger Resektion mit Ergebnissen zum Grad IIIA für das krankheitsfreie Überleben mit adäquater Beschreibung der Randomisationgszuweisung, aber ohne verblindete Messpunkterhebung und ohne Zugriff auf das Protokoll mit vollständiger Auswertung aller behandelten Patienten (98%) zeigt keinen Unterschied in der Wirksamkeit, aber eine geringere Toxizität in der Gruppe mit Pemetrexed und Cisplatin, keine Abwertung der Evidenz aufgrund von Indirektheit (Studie wurde vollständig in Japan durchgeführt, Ergebnisse entsprechen denen der deutschen Studie).  Die Studie wurde von der West Japan Oncology Group, Pharma-Valley Center und der Japan Agency for Medical Research and Development unterstützt, mehrere Autoren berichten Interessenkonflikte (Forschungsförderung und Honorare durch pharmaceutische Firmen).

Kreuter 2014

(26-28)

(TREAT study)

NCT00349089

10/2006 to 12/2009

Deutschland, Belgien und Luxemburg

RCT

feasibility of a protocol with reduced toxicity might improve feasibility of postoperative delivery of adjuvant chemotherapy drugs in NSCLC, to improve compliance and survival

Inclusion criteria:

· adults (≥18,

·  histo-pathologically confirmed NSCLC (stage IB, IIA, IIB or T3N1) without need for further radiotherapy and R0 resection and removal of mediastinal lymph nodes > 1,5 cm on the preoperative CT sca

· full recovery after surgery

· Karnovsky performance ≥ 80%, ECOG 1

Exclusion criteria:

· Presence of a Pancoast tumor

· Involvement of N2/N3 lymph nodes

· Distant metastases

n=132

age: 59 (38-74) years

26% females

29% smoker

66% ex-smoker

83% lobectomy, 14% pneumenectomy, 3% complex resections

Stage IB/IIA/IIB/T3N1: 38/10/47/5 %

COPD (I/II/III): 23/13/2%

under therapy with bronchodilators: 28%

4-6 weeks postoperative:

Cisplatin(75 mg/m² d1; q d22) and Pemetrexed (500 mg/m² d1)

with vitamin-supplemen-tation

(folic acid, vitamin B 12 ) and dexamethasone (4 mg of oral or

equivalent, twice daily the day before, the day of and the day after each dose of pemetrexed for rash prophylaxis, unless medically

contraindicated)

(n=67)

vs.

Cisplatin (50 mg/m² d1, 8; q d29) and Vinorelbine (25 mg/m² d1, 8, 15, 22; q d29) (n=65)

for 4 cycles

Primary:

feasibility of 4 cycles of adjuvant CT (no death due to cancer, toxicity, or comorbidity; no non-acceptance by the patient leading

to premature withdrawal; and no DLT

Secondary:

drug delivery, time to treatment failure, distant metastases free survival, local RFS, OS, localization of relapse, dose delivery, RFS

safety: NCICTC version 3.0

Feasibility:

90 % vs. 66 % of the planned doses were given

Tumor relapse and relapse-free survival (RFS):

· Relapse in 31 % vs. 36 %

· RFS: medians not reached, very similar 3-year RFS rates (59 vs. 60%), no difference in RFS: HR 0.831 (95% CI: 0.317–2.177); p=0.707)

· More patients with local relapses in IG (18% vs. 5%), with most being mediastinal lymph node metastases in IG

· no differences in  rates of distant relapse (20.9% versus 18.5%).

· More frequent brain metastases in IG (75% vs. 21%)

· no difference in local RFS and distant metastasis–free survival

Deaths and Overall survival:

27% vs. 26% of patients died with comparable between groups, 17% due to tumor relapse (18% vs. 15%), 5% unrelated to therapy or tumor (5% in each arm), and 5% unknown.

no differences in death by resection type: 83% of the patients received a lobectomy, with death rates of 22% in both arms, and 14% of patients received a pneumonectomy, with a death rate of 4% (5 vs. 3%)

no differences on OS, medians not reached vs 59 months; (HR 0.594; 95% CI 0.165–2.131, p=0.424);

Toxicities:

generally mild, no grade 3/4 toxicity

most frequent pulmonary toxicities: grade 1/2 cough (13.4% vs. 7.7%) and dyspnoe (9% vs. 4.6%)

Pneumonia noted in 1 vs. 0 patients, bronchitis in 1 patient in each arm, no COPD exacerbations reported.

2b

Abwertung aufgrund von Studienlimitationen, geringer Präzision und Indirektheit

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In conclusion, the present study provides evidence that adjuvant chemotherapy does not have a negative effect on pulmonary function. However, a non-significant difference between D LCO decline in different therapy arms warrants further study.“(28)

“Although adjuvant chemotherapy with CPx is safe and characterized by less toxicity and better dose delivery than CVb, overall survival was not influenced by treatment arm in the context of this phase II trial.”(26)

Schlussfolgerung der Begutachterin: Multizentrische, in Deutschland, Belgien und Luxemburg durchgeführte Phase II unter Einschluss von 132 Patienten mit NSCLC nach vollständiger Resektion (Grad IB-T2N1) ohne adäquater Beschreibung der Randomisation (trotz veröffentlichtem Protokoll) und ohne Verblindung (daher Abwertung aufgrund von Studienlimitationen) zeigt Machbarkeit und Sicherheit einer Therapie mit Cisplatin und Pemetrexed bei geringen Nebenwirkungen. Trotz vollständigerer Therapie konnten keine Unterschiede hinsichtlich des Überleben und Anzahl der Rückfälle zwischen den Interventionsgruppen gezeigt werden.Abwertung aufgrund von fehlender Präzision (geringe Fallzahl) und Indirektheit (5% der Patienten waren im Stadium T3N1, 38% im Stadium IB und 57% im Stadium II).

Okamoto 2018

(29)

2009-2013

Japan

RCT

efficacy

and safety of S-1 versus cisplatin + S-1, followed by S-1 in patients with completely resected stage IB-III NSCLC

Inclusion criteria:

· adults (20-75 yrs)

· completely resected NSCLC (stage IB, II or III)  

· ECOG PS 0 or 1

· No previous CT or RT

· Adequate bone marrow and organ function

Exclusion criteria:

· drug allergies, cardiac infarction

· uncontrolled diabetes mellitus, problematic infections. Interstitial pneumonia, concomitant malignancies

n=141

mean age: 62.7±7.8 years

males: 76%

PS 0/1: 76/24%

Smoking history: 81%

IIA/B/IIIA: 31/26/42 %

Adeno/Squamous/large/other carcinoma: 63/28/3/6 %

Lobectomy: 92%

Pneumonectomy: 8%

within 8 weeks after surger:

Longterm S-1 (80 mg/m² oral) for 2 consecutive weeks repeated every 3 weeks for 1 year (n=100)

vs.

S-1 (80 mg/m² oral) for 3 consecutive weeks +  Cisplatin (60 mg/m²) d8

repeated every 5 weeks for four cycles, followed by  Longterm S-1 (80 mg/m² oral) for 2 consecutive weeks repeated every 3 weeks for 1 year (n=100)

Primary:

disease-free survival (DFS) at 2 years

Secondary:

OS

Safety

feasibility

Survival:

· Comparable 2-year DFS rate: 52 % (95% CI 40-63) vs. 61% (95% CI 48-70); HR 1.37 (95% CI 0.88–2.13).

· comparable OS: HR 0.94 (95% CI 0.48-1.85)

Safety:

· no significant differences in occurance of grade 3-4 AEs

· main grade 3-4 AEs: appetite loss: 4.3 vs. 11.6 %, decreased neutrophil count (5.7 vs. 10.1% and anemia ( 0vs. 5.8 %)

· no treatment related deaths

Feasibility:

· no differences in treatment completion: 45.7 % (95% CI 41.9-66.3) vs. 43.5 % (95%CI 44-69.4)

2b

Abwertung aufgrund von geringer Präzision und Indirektheit

 Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: „In conclusion, long-term adjuvant chemotherapy with S-1 was a feasible and promising treatment for patients with completely resected NSCLC, regardless of whether cisplatin was added. We recommend that S-1 monotherapy should be investigated further owing to its low toxicity and practical convenience.“

Schlussfolgerung der Begutachterin: Multizentrische, in Japan durchgeführte Phase II unter Einschluss von 141 Patienten mit NSCLC nach vollständiger Resektion (Grad IIA/B-IIIA) (daher Abwertung aufgrund von Indirektheit) mit adäquater Beschreibung der Randomisationgszuweisung, aber ohne verblindete Messpunkterhebung und Hinweise auf ein veröffentlichtes Protokoll mit vollständiger Auswertung aller Patienten zeigt Machbarkeit, Wirksamkeit und Sicherheit einer Langzeit-Monotherapie mit S-1 bei geringen Nebenwirkungen. Abwertung aufgrund von fehlender Präzision (geringe Fallzahl) und Indirektheit (42 % der Patienten waren im Stadium III, Studie wurde vollständig in Japan durchgeführt).  Die Studie wurde mit öffentlichen Fördergeldern unterstützt, 2 Autoren einer großen Gruppe berichten Interessenkonflikte (Forschungsförderung und Honorare durch pharmaceutische Firmen).

AE: Adverse event; CG: Control group; CI: Confidence interval; CRT: Chemo-radiotherapy; CT: Chemotherapy; CTCAE: Common Terminology Criteria for Adverse Events;  d: Tag; DCR: Disease control rate; DFS: Disease-free survival; DLT: dose-limiting toxicity; ECOG: Eastern Cooperative Oncology Group; FACT-L: Functional Assessment of Cancer Therapy-Lung; GHS: Global health scores; HR: Hazard ratio; IG: Intervention group; LCS: Lung Cancer Symptom Scale; N: Number of studies; n: Number of participants; NCI CTCAE: National Cancer Institute Common Toxicity Criteria; NR: not reached; NSCLC: Non-small cell lung cancer;  OR: Odds Ratio; OS: overall survival; PFS: Progression-free survival; PSI: Pulmonary Symptom Index; QoL: Quality of life; RECIST: Response Evaluation Criteria in Solid Tumors; RCT: Randomized controlled trial; RFS: Relapse-free survival; RoB: Risk of bias; RR: Relative Risk; RT: Radiotherapy; SAE: Serious adverse event;TTF: Time to treatment failure;  wks: weeks; yrs: years

Ahn 2015

(43)

KCSG-LU05-04

NCT00326378

RCT

Efficacy of consolidation CT with docetaxel and cisplatin after concurrent CRT with the same agents in locally advanced NSCLC

10/2005-04/2011 in 31 centers in Korea, China, Taiwan

Inclusion Criteria:

·  histologically documented NSCLC with inoperable stage IIIA/IIIB disease (proven by computed tomography (CT), magnetic resonance imaging, and/or positron emission tomography (PET))

·  N2 or N3 disease must have been confirmed by pathology or PET

·  ≥ 18 yrs,  ECOG 0-1,

·  measurable disease (RECIST); no prior CT, RT to the chest, immunotherapy, or biologic therapy; FEV1≥0.8 L (spirometry)

·  adequate bone marrow, renal, and hepatic function

·  no local progression or distant metastases, adequate organ function before the start of CC

Exclusion criteria:

·  pregnancy or lactating women, women without a pregnancy test ≤ 14 days before the first administration, childbearing potential without the use of adequate contraception

n=459 randomized, 420 analysed

median age: 61 yrs (31-79)

males: 82.6%

ECOG (0/1/2): 23.6/68.6/7.6

Pathology (Adeno/squamous/other): 51.7/32.1/12.9

Stage (IIIA/IIIB): 22.1/77.9

CRT for all patients:

docetaxel (20 mg/m² iv) + cisplatin (20 mg/m² iv. on days 1, 8, 15, 22, 29, and 36

+ thoracic RT 5 days/week (once-daily fractions, 2.0 Gy per fraction, total 66 Gy in 33 fractions)

4-8 wks after CRT:

consolidation CT (n=209):

docetaxel (35 mg/m² iv. for 1 hour)+ cisplatin (35 mg/m² for 1 hour) on days 1,8, repeated every 3 weeks

vs.

best supportive care (n=211)

for up to 3 cycles until unacceptable toxicity or evidence of disease progression

42.1% received planned 3  cycles, 54.1% finished ≥ 2 cycles, 31.6% did not receive any CC

reasons for no CC: early death (n=22), consent withdrawal or patient refusal (n=14), AEs (n=12), disease progression (n=10)

median follow-up: 50.7 months

Primary:

PFS

Secondary:

OS

Overall response rate

Patterns of failure

Toxicity (AEs graded with CTCAE 3.0)

Progression-free survival: no difference (p=0.36)

median: 9.1 months (95% CI 7.9-10.9) vs. 8.1 months  (95% CI 7.6-8.9); HR 0.91 (95% CI 0.73-1.12) p=0.36

Overall survival: no difference (p=0.44)

median: 21.8 (95% CI 17.7-24.7) vs. 20.6 months (95% CI 17.6-26.3);  HR 0.91 (95% CI 0.72-1.25)

Overall response: no difference (p=0.33)

 43.1% (with 2.9% complete response) vs. 38.4% (with 4.3% complete response)

Disease control rate: approximately 58% in both arms

Safety (during consolidation therapy):

slightly more febrile neutropenia (grade 3/4): 1.2/0.6 %

neutropenia (grade 3/4):  4.6 / 2.3 %

infection (grade 3/4): 2.3% vs. 2.9%

radiation pneumonitis (grade 3/4): 1.2% vs. 1.2%

treatment-related mortality (grade 3/4): 3.6% during CRT, 2.3% during CC.

1b

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In summary, this study confirms that the current strategy of consolidation treatment with cytotoxic chemotherapy after CCRT does not improve survival in patients with locally advanced (LA)-NSCLC. In future trials, CCRT without CC should remain the reference arm. Given that distant failure is the most common pattern of failure after CCRT in LA-NSCLC and the treatment outcome remains poor, we strongly believe that the concept ofconsolidation therapy needs to be further explored using less toxic, better tolerated agents.”

Schlussfolgerung der Begutachterin: Multizentrische asiatische RCT ohne Verblindung mit adäquat  beschriebener Randomisierung  und Studienregistrierung und modifizierter ITT-Analyse  nach Ausschluss von 17 (3.9%) Patienten, welche randomisierte Studienphase nicht begonnen (kein Unterschied zwischen Gruppen) mit adäquater Studiengröße und Nachbeobachtungszeit, aber ohne Wirksamkeitsnachweis, Interessenkonflikte der Autoren können nicht ausgeschlossen werden.

Antonia 2017

PACIFIC

NCT02125461

(48-50)

RCT

Efficacy and safety of durvalumab as consolidation therapy in patients with stage III, locally advanced, unresectable NSCLC that had not progressed after platinum-based CRT

04/2014-04/2016 in 26 countries

Inclusion Criteria:

·  histologically or cytologically documented stage III, locally advanced, unresectable NSCLC who  received

·  no disease progression after this treatment, ≥ 18 yrs, WHO performance status 0-1, an estimated life expectancy ≥ 12 wks, completion of the last radiation dose within 1-14 days before randomization (after a protocol amendment, this criterion was changed to 1 to 42 days before randomization)

Exclusion criteria:

·  previous exposure to anti–PD-1 or PD-L1 antibodies,

·  receipt of immunotherapy or an investigational drug within 4 wks before the first dose (6 weeks for monoclonal antibodies),

·  active or previous autoimmune disease (within ≤ 2 yrs) or a history of primary immunodeficiency,

·  evidence of uncontrolled, concurrent illness or ongoing or active infections;

·  unresolved toxic effects ≥ grade 2 (CTCAE) and

·  ≥ grade 2 pneumonitis from previous CRT

n=713

age: 64 (23-90) yrs

70.1% males

disease stage (IIIA/IIIB/other): 52.9/44.7/2.4 %

WHO (0/1): 48.8/50.8 %

Tumour histologic type: (squamous, nonsquamous): 45.7/ 54.3 %

Smoking (current/former/never): 16.4/76.4/9.0 %

previous RT (<54 Gy/ ≥54 to ≤66 Gy/>66 to ≤74 Gy): 0.4/92.4/6.9 %

previous CT (induction /concurrent with RT): 26.8 /99.7 %

Best response to previous CT (complete/partial response/ stable disease): 2.2/48.1/47.1 %

CRT for all patients:

≥ 2 cycles of platinum-based CT (etoposide, vinblastine, vinorelbine, a taxane [paclitaxel or docetaxel], or pemetrexed) concurrently with

definitive RT (54 to 66 Gy), mean dose to the lung < 20 Gy (volume of lung parenchyma that ≥ 20 Gy) was

1-42 days after CRT: consolidation therapy (n=476): durvalumab (10 mg/kg of body weight iv)

vs.

Placebo (n=237)

every 2 wks

for up to 12 months until disease progression, could receive the drug again if disease control had been achieved

Median number of infusions:

20 (1-27) vs. 14 (1-26)

median relative dose intensity: 100% (29-100) vs. 100% (50-100)

Median follow-up: up to 25.2 months (IQR 14.1-29.5) months

Primary:

PFS (RECIST)

OS

Secondary:

12 /8 months DFS

objective response rate

duration of response

time to death or distant metastasis 24 months OS

Health-related QoL (time to deterioration of cough, dyspnoea, chest pain, haemoptysis (all QLQ-LC13), and global health

status or QoL)

Safety: 

(CTCAE, version

side-effect 4.03)

AEs, SAEs

vital signs, physical and laboratory examinations

AEs of special interest and immune-mediated AEs ( AEs of special interest that led to the use of systemic glucocorticoids, endocrine therapy, or other immune-suppressants, that were consistent with an immune-mediated mechanism, and that had no clear alternative cause)

Progression free survival (49):

·    median: 16.8 (95% CI 13.0-18.1) vs. 5.6 months (95% CI 4.6-7.8); HR 0.52 (95% CI 0.42-0.65;p<0.001)

·    12-month PFR rate: 55.9% (95% CI 51.0-60.4) vs. 35.3% (95% CI 29.0-41.7)

·    18-month PFS rate: 44.2% (95% CI 37.7-50.5) vs.27.0% (95% CI 19.9-34.5)

·    consistent across all pre-specified sensitivity and subgroup analyses

·    no differences between males and females, NSCLC disease stage and depending on best response, higher benefit for

o    younger age: < 65 vs. ≥ 65 yrs: HR 0.43 (95%CI 0.32-0.57) vs. 0.74 (95%CI 0.54-1.01)

o    nonsmoker vs. smoker: HR 0.29 (95%CI 0.15-0.57) vs. 0.59 (95%CI 0.47-0.73)

o    nonsquamous vs. squamous types: HR 0.45 (95%CI 0.33-0.59) vs. 0.68 (95%CI 0.50-0.92)

o    PD-L1 ≥25 vs. <25%: HR 0.41 (95%CI 0.26-0.65) vs. 0.59 (95%CI 0.43-0.82)

o    EGFR mutation: negative vs. positive: HR 0.47 (95%CI 0.36-0.60) vs. 0.76 (95%CI 0.35-1.64)

Overall survival (48):

·  median: NR (95% CI 34.7-NR) vs. 28.7 months (95% CI 22.9-NR); HR 0.68 (99.73 % CI 0.47-0.997; p=0.0025)

·  12-month rate: 83.1% (95% CI 79.4-86.2) vs. 75.3% (95% CI 69.2-80.4)

·  24-month rate: 66.3% (95% CI 61.7-70.4) vs. 55.6% (95% CI 48.9-61.8) (p = 0.005)

Time to death or distant metastases (48):

median: 28.3 (95% CI 24.0-34.9) vs. 16.2 months (95% CI 12.5-21.1), HR 0.53 (95% CI 0.41-0.68)

Frequency of new lesions (48):

22.5% vs. 33.8% with a lower incidence of new brain metastases (6.3% vs. 11.8%)

Time to second progression or death (48):

median: 28.3 (95% CI 25.1-34.7) vs. 17.1 months (95% CI 14.5-20.7); HR 0.58 (95% CI 0.46-0.73)

Overall response rate (48):

30.0% (95% CI 25.8-34.5) vs. 17.8% (95% CI 13.0-23.6) (p<0.001) with ongoing response at 18 months in 73.5 vs. 18 months

Safety (48) (during consolidation therapy):

·    3 or 4 AEs: 30.5% vs. 26.1%

·    Discontinuation of trial regimen due to AEs: 15.4% vs. 9.8%

o    Most frequent AEs leading to the discontinuation:

o    pneumonitis (4.8% vs. 2.6% of patients),

o    radiation pneumonitis (1.3% vs. 1.3%)

o    pneumonia (1.1% vs. 1.3%)

·    SAEs: 29.1% vs. 23.1%

·    Death due to AEs: 4.4% vs.6.4%

·    AEs of special interest (any grade or cause):  66.7% vs. 49.1% with 56.8% and 43.6% of patients, reporting grade 1 or 2 events.

Changes in health-related quality of life over 12 months (50):

·    Cough, dyspnoe, chest pain, fatique, appetite loss, physical functioning and global health status of QoL remained stable for both groups

·    No clinically important differences in:

o    time to deterioration for most symptoms, functioning terms, or global health status or QoL

o    proportion of patients with clinically relevant improvements in most symptoms

o    proportion of patients whose scores improved for global health status or QoL or most functioning terms

·    time to deterioration of other pain (ie, any pain other than in the chest, arms, or shoulders) longer in  durvalumab (HR 0.70; 05%CI 0.57-0.87)

·    emotional functioning favoured durvalumab

1b

 Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In conclusion, in the PACIFIC study, one of the coprimary end points was met at this planned interim analysis, and this study showed a significant increase in progression-free survival and no new safety signals with durvalumab in patients with stage III, unresectable NSCLC who had received chemoradiotherapy. These positive findings in an unselected patient population, irrespective of baseline expression of PD-L1 on tumor cells, suggest that durvalumab may be an effective adjuvant therapy in patients with stage III disease after standard treatment. Uncertainty about the potential mechanisms driving the interaction between immunotherapy and chemoradiotherapy warrants further investigation.” (49)

In conclusion, this trial showed a survival advantage with durvalumab therapy after concurrent chemoradiation therapy in patients with stage III, unresectable NSCLC (48).

Schlussfolgerung der Begutachterin: Methodisch gute multizentrische internationale RCT mit doppelter Verblindung (Patienten, behandelnde Ärzte und Erfasser von PFS und Response) und guter Beschreibung der Erzeugung und verblindeten Zuweisung der Randomisierungsfolge, alle geplanten Endpunkte werden für alle randomisierten Endpunkte berichtet, ausreichende Power und Nachbeobachtungszeit zum Nachweis von Unterschieden für PFS und OS (daher hohe Präzision der Ergebnisse), Studie wurde wie im Protokoll geplant nach Nachweis der Wirksamkeit in der 1. Zwischenanalyse ( nach 299 statt 491 beobachteten Todesfällen) abgebrochen, Finanzierung der Studie und Publikation durch Astra Zeneca, Interessenkonflikte der Autoren können nicht ausgeschlossen werden.

Feng 2015

(47)

NCT02430974

RCT

efficacy and safety of CT with icotinib in patients undergoing resection of stage IB to IIIA EGFR-mutated NSCLC

02/2011-12/2012 in 1 center in China

Inclusion Criteria:

·       histologically confirmed activating EGFR-mutated NSCLC between stage IB (with high risk factors) and stage IIIA, received an operation to remove the lung lesion completely

·       high-risk stage IB NSCLC were defined as poorly differentiated tumors, vascular invasion, wedge resection, tumor size > 4 cm, visceral pleural involvement or incomplete lymph node sampling,

·       activating EGFR-mutation-positive disease: ≥ 1 of 4 mutations (exon 19 deletion, or 18 G719X, 21 L858R, or 21 L861Q mutations)

·       18 years, ECOG 0 or 1, adequate hematological, biochemical and organ function

Exclusion Criteria:

·  single mutation ofexon 20 T790M, 20 insertions or 19 D761Y (resistant to EGFR-TKI)

·  systemic anticancer therapy prior to surgery, other malignancies before or during the study, any unstable illness, pregnancy or lactation

n=41

age: 56.6±10.3 yrs

69% males

95% adenocarcinoma

51% present or former smoker

lymph node status (N0/1/2): 59/10/31 %

stage (IB/II/IIIA): 44/26/31 %

EGFR mutations (19 delete, 21 L858R): 41/56%

CT for all patients:

4 cycles of platinum-based doublet CT (150 mg/m² paclitaxel + 80 mg/m² nedaplatin or 30 mg/m² lobaplatin on day 1 of a 3 wk cycle)

2 wks after CT:

consolidation therapy (n=21):

oral icotinib  (125 mg, 3x daily)

for 4 to 8 months or until occurrence of disease relapse, metastasis or unacceptable icotinib or CT toxicity

vs.

Best supportive care (n=20, 18 analyzed)

Primary:

DFS

Secondary:

toxicity (WHO toxicity grading scale for determining the severity of AEs)

follow-up for up to 24 months

Disease free survival: no difference (p=0.066)

12 months: 21/21 (100%) vs. 16/18 (88.9%); p=0.122

18 months: 20/21 (95.2%) vs. 15/18 (83.3%); p = 0. 225

24 months: 19/21 (90.5%) vs. 12/18 (66.7%); p = 0. 066

Toxicity:

4 patients in each group had ≥ 1 CT-related AE (19 vs. 22 %)

gastrointestinal reactions (grade 1/2): 3/0 vs. 3/1

marrow suppression: 4/0 vs. 3/0

neurotoxicity: 4/0vs. 3/1

liver and kidney damage: 4/0 vs. 2/1

allergic reactions: 3/0 vs. 2/1

2b

Abwertung aufgrund von schwer-wiegenden Studien-limitationen und geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: „The results suggest that chemotherapy plus orally icotinib displayed better DFS compared with chemotherapy only, yet the difference in DFS was not significant. We would think the preliminary result here was promising, and further trials with larger sample sizes might confirm the efficiency of adjuvant TKI in selected patients.“

Schlussfolgerung der Begutachterin: Monozentrische RCT aus China ohne Verblindung und Beschreibung der Zuweisung der Behandlung, abweichend vom Protokoll wurden 41 statt 58 rekrutiert, 2 (10 %) Patienten aus der Kontrollgruppe erhielten keine Chemotherapie und wurden aus der Analyse ausgeschlossen (daher Abwertung aufgrund von schwerwiegenden Studienlimitationen), Finanzierung der Studie durch ein „major drug discovery project“in China, es werden keine Interessenkonflikte der Autoren berichtet, Abwertung der Evidenz erfolgt aufgrund von Studienlimitationen und der geringen Präzision der Ergebnisse aufgrund der geringen Fallzahl und Nachbeobachtungszeit von bis zu  24 Monaten

Flentje 2016

(GILT)

(44)

RCT

Benefit on PSF of

adding 2 cycles of CT to best supportive

care in patients with at least stable disease control following CRT

05/2005- 05/2009 in 34 German centres

Inclusion Criteria:

·  histologically confirmed stage IIIA/ IIIB (TNM6) inoperable NSCLC

·  no malignant pleural effusion, weight loss

·  Staging based on bronchoscopy, thoracoabdominal CT, bone scintigraphy and brain CT or MRI

·  Patients received at least stable disease after CRT

n=279 received CRT, n=201 randomized

median age: 60.1 vs. 59.2 yrs (33.9-75.7), 50 %≥ 60 yrs

72% males

Karnovsky Index (80/90/100 %): 15.4/41.8/42.8%

Comorbidties (0/1/2/≥3): 19.9/37.3/35.3/7.5 %

<5% weight loss: 7.9%

Histology (squamous/adeno/large cell): 53.2/36.3/7.5%

Stage (IIIA/IIIB): 15.4/84.6%

CRT for all patients:

two cycles of oral vinorelbine (50 mg/m² days 1, 8 and 15) + cisplatin (20 mg/m² days 1–4) q4w) + concomitant RC (66 Gy)

4-6 wks after CRT:

consolidation therapy (n=96): 2 cycles oral vinorelbine (60–80 mg/m² days 1 and 8) + cisplatin (80 mg/m² day 1) q3w + best supportive care

vs.

Best supportive care (n=105) alone

until time of restaging (around 12 wks after randomization)

median relative dose intensity: 89.6 % received both cycles with a median dose intensity of 96.5 % for cisplatin and 93.3 % for vinorelbine

Subsequent treatment:

CT: 45.8 vs. 58.1 %

Symptomatic RT: 47.9 vs. 41.9 %

Primary: PFS

Secondary :

OS

objective response rate (RECIST 1.0)

Disease control rate

toxicity (CTCAE 3.0)

Quality of life

(LCSS)

Follow-up every 3 months

PFS: no differences (p = 0.63)

median: 6.4 (5.0–8.7) vs. 5.5 (3.8–7.4) months; HR: 0.93 (95% CI 0.69–1.26)

better PFS for higher age >60 yrs: HR 0.72 (95%CI 0.53-0.97)

OS: no differences (p = 0.51)

median: 20.8 (13.5–25.3) vs. 18.5 (13.6–24.7)

4-year OS rate: 25.2 and 21.4 %

Response: no differences (p=0.12)

Complete response: 3.1 vs. 1.0 %

partial response: 26.0 vs. 23.8 %

stable disease: 37.5 vs. 31.4 %

progressive disease: 12.5 vs. 30.5 %

not evaluable: 20.8 vs. 13.3 %

ORR: 29.1 % (95% CI 20.4-39.4) vs. 24.8 % (95% CI 16.8-34.2)

Disease control rate: no  differences (p=0.127)

66.7 % (95% CI 56.0-76.0) vs. 56.2 % (95% CI 46.0-66.0)

Progression during follow-up:

80.2 % vs. 90.5 %

Death from malignancy: benefit for IG (p=0.049)

74.2 vs. 86.8/74.2 %

Cancer-specific survival: no differences (p=0.333)

 25.5 vs. 22.7 months

Safety (during consolidation therapy):

·    Predominant side effects (grades 3 / 4): leukopenia (26.7 % vs. 0%) and neutropenia (22.1 vs. 0%)

·    febrile neutropenia occurred in < 2 % (1.0  vs. 0 %)

·    Non-haematological toxicities (only grade 3): nausea/vomiting (2.3/3.5 % vs. 0%) and fatique (3.5% vs. 0%)

·    5 deaths (2 vs.3)

1b-

Abwertung aufgrund von schwer-wiegenden Studien-limitationen

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Taking together the results from the GILT study and available evidence from controlled trials, CC after definitive concurrent CRT does not improve outcome in unselected patients. Further analyses should investigate the possible benefit in subpopulations.“

Schlussfolgerung der Begutachterin: Multizentrische RCT aus Deutschland ohne Verblindung, berichteten  Informationen zur Randomisierung  und Studienregistrierung, welche aufgrund der langsamen Rekrutierung vorzeitig abgebrochen wurde (geplant war die Randomisierung von 282 Patienten, rekrutiert wurden in 4 Jahren und 26 Kliniken nur 201 Patienten, die mediane Nachbeobachtungszeit wird nicht berichtet (daher Abwertung aufgrund von schwerwiegenden Studienlimitationen), Finanzierung der Studie und Publikation durch das Institute de Recherche Pierre Fabre, Interessenkonflikte der Autoren können nicht ausgeschlossen werden.

Hanna 2008

(45, 46)

RCT

efficacy of consolidation docetaxel in patients with inoperable stage III NSCLC patients

Multicenter study, USA

03/2002-08/2006

Inclusion Criteria:

·       histologic or cytologic confirmation of unresectable NSCLC, stage IIIA or IIIB

·       measurable or assessable disease

·       FEV1 ≥1 L by spirometry within 42 days of study treatment

·       ECOG 0 to 1, unintended weight loss

·       adequate bone marrow, renal and hepatetic function

·       eligibility for consolidation therapy required completion of initial CRT within 4 to 8 weeks of random assignment without local progression or distant metastases, ECOG PS 0-2 at random assignment, adequate bone marrow and hepatic function, absence of symptomatic peripheral neuropathy

Exclusion criteria:

·  disease extending into the cervical region

·  prior CT or RT

·  symptomatic peripheral neuropathy, malignant effusions, superior sulcus (Pancoast) tumors, or significant cardiac disease

n=166 (45)

age (<70 /≥70 yrs): 75.8/24.2%

males: 69%

ECOG 0: 61%

stage IIIB: 63%

FEV1 >2: 51%

PET scan: 68%

Current smoker: 45%

CRT for all patients:

cisplatin (50 mg/m² i.v. days 1, 8, 29, 36) + etoposide (50 mg/m² i.v. days 1–5,  29–33) + RT (1.8 Gy daily at 5 days/ wk for a total of 25 fractions (45 Gy) to the primary tumor and mediastinum + boost to the primary and involved nodes to 1.8-Gy daily in 8 fractions (14.40 Gy), total dose of 59.40 Gy in 33 fractions)

4-8 wks after CRT:

consolidation therapy (n=82):

docetaxel (75 mg/m² every 3 weeks for 3 cycles)

vs.

observation (n=84)

Prophylactic granulocyte colony-stimulating factor support during consolidative chemotherapy permitted

Primary:

OS

Secondary:

PFS

toxicity

Overall survival (OS) (45): no differences depending on consolidation therapy or age group (p=0.75)

·    median from registration: 24.2 (95% CI 18.6 -34.8) vs. 26.1 months (95% CI 18.6-32.0)

·    3-year rate: 32.4 % (95%CI 21.1-44.1) vs. 36.7 % (95% CI 25.3-48.9)

·    5-year rate: 16.4 % (95%CI 7.4-28.7) vs. 23.8 % (95% CI 13.0-36.5)

Progression-free survival (OS) (45): no differences

median from registration: 10.8 (95% CI 8.6 -13.3) vs. 10.3 months (95% CI 8.2-15.2)

Toxicity: (46)

·    high hematologic toxicities (grade 3/4) with docetaxel: neutropenia (24.7 %), febrile neutropenia (10.9%), anemia (1.3%)

·    non-hematologic toxicities (grade 3-5): higher with docetaxel: infections (11.0 vs. 0%), pneumonitis (9.6 vs. 1.4%), treatment-related deaths (5.5 vs. 0%)

1b-

Abwertung aufgrund von Studien-limitationen

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Consolidation docetaxel after PE/XRT results in increased toxicities but does not further improve survival compared with PE/XRT alone in patients with stage III inoperable NSCLC.“(46)

“Consolidation docetaxel after EP/XRT does not improve survival in LA-NSCLC. Fit older adults with LANSCLC benefit from concurrent chemoradiation similarly as younger patients but experience higher rates of hospitalization and toxicity.” (45)

Schlussfolgerung der Begutachterin: Multizentrische RCT aus USA ohne Verblindung, berichteten Informationen zur Randomisierung und öffentliche Studienregistrierung (daher Abwertung aufgrund schwerwiegender Studienlimitationen). Fallzahl wurde zuerst aufgrund der langsamen Rekrutierung und hohen Drop-out-Rate nach der Chemo-Radiotherapie (243 rekrutiert, von denen 166 hinsichtlich der Konsolidierungstherapie randomisiert wurden) erhöht, anschließend wurde die Studie aufgrund der in einer geplanten Zwischenanalyse festgestellten fehlenden Wirksamkeit vorzeitig abgebrochen wurde (geplant war die Rekrutierung von 180 in der Konsolidierungsphase zu rekrutierenden Patienten, Zwischenanalyse erfolgte auf Grundlage von Daten von 147 Patienten,  die mediane Nachbeobachtungszeit wird nicht berichtet, Finanzierung der Studie und Publikation durch Sanofi-Aventis, wenige kompensierte Interessenkonflikte der Autoren

AE: Adverse event; CC: Consolidation chemotherapy; CG: Control group; CI: Confidence interval; CRT: Chemo-radiotherapy; CT: Chemotherapy; CTCAE: Common Terminology Criteria for Adverse Events;  DFS: Disease-free survival; ECOG: Eastern Cooperative Oncology Group; FEV1: Forced expiratory volume in 1 second;  HR: Hazard ratio; IG: Intervention group; LCSS: Lung Cancer Symptom Scale; N: Number of studies; n: Number of participants; NR: not reached; NSCLC: Non-small cell lung cancer;  OR: Odds Ratio; OS: overall survival; PFS: Progression-free survival; QoL: Quality of life; RECIST: Response Evaluation Criteria in Solid Tumors; RCT: Randomized controlled trial; RoB: Risk of bias; RR: Relative Risk; RT: Radiotherapy; SAE: Serious adverse event; wks: weeks; yrs: years

Gambazzi 2019

(27)

ISRCTN 16281786

Switzerland

Screening of patients 10/2011 to 08/2014

RCT

diagnostic performance of PET-CT with contrast-enhanced CT (CE-CT) of the chest in post-treatment surveillance of NCSCL patients

Inclusion criteria:

·  patients with NSCLC ≥ 18 years

·  18F-FDG–PET–positive tumor

·  tumor stage I to III (or IV, if patients presented with solitary, completely resected brain metastases)

·  had completed a curative-intent treatment for NSCLC (anatomic surgical lung resection alone or + neoadjuvant chemotherapy or + adjuvant chemotherapy, or + radiotherapy/ chemotherapy alone

Exclusion criteria:

·      insufficient knowledge of the German language,  impaired kidney function (glomerular filtration rate <30 mL/min)

n=98

median age: older with PET-CT 67 (IQR 61-74) vs. 61 (IQR 56-70)

males: 70 %

COPD: 49 %

adenocarcinoma: 64 %

squamous cell carcinoma: 34 %

NSCLC I/II/III/IV: 46/29/21/3 %

therapy:

surgery alone: 62.5 %

surgery+adjuvant CTX: 20%

surgery+neoadjuvant CTX: 9 %

RTX:3 %

During 2 years of follow-up: clinical examination and imaging procedure  at 6-month intervals

in case of radiologic suspicion of cancer recurrence, the patient completed the study and underwent a diagnostic workup:

 PET-CT (5 MBq/kg body weight with a Biograph MCT) + low-dose CT for attenuation correction of the PET images (n=50)

vs.

contrast-enhanced CT (CE-CT) of the chest (64 or a 320 detector-row CT scanner) (n=46)

+ nonscheduled brain CT, bronchoscopy, or therapeutic-intent surgical intervention

in case of no suspicious radiologic finding but symptoms of recurrence between scheduled imaging scans completed the study as well and underwent a diagnostic workup

Study termination at 24 months without a suspicious radiologic finding

if scheduled imaging shows a suspicious radiologic finding:

Diagnostic workup reveals no cancer

Diagnostic workup reveals cancer recurrence

Symptomatic recurrence before first scheduled imaging

Symptomatic recurrence between scheduled imaging

Diagnostic performance:

Sensitivity

Specificity

Positive predictive value

130 surveillance PET-CTs vs.  138 surveillance CE-CTs were performed

No differences between clinical outcomes:

·   Study termination at 24 months without a suspicious radiologic finding: 36 vs. 46 % (p=0.336)

·   No cancer after diagnostic work-up of suspicious radiologic finding: 22 vs. 17% (p=0.571)

·   Diagnosis of cancer recurrence after diagnostic work-up: 28 vs. 30% (p=0.793)

·   Symptomatic recurrence before first scheduled imaging: 10 vs. 4% (p=0.287)

·   Symptomatic recurrence between scheduled imaging: 4 vs. 2% (p=0.607)

·   Time to suspicious radiologic finding: no difference (p=0.528)

No difference in diagnostic performance

·   Sensitivity: 88 % (95%-CI 62 to 98) vs. 93 % (95%-CI 68 to 100)

·   Specificity: 62 % (95%-CI 42 to 79) vs. 72 % (95%-CI 53 to 87)

·   Positive predictive value: 56 % (95%-CI 35 to 76) vs. 64 % (95%-CI 41 to 83)

·    

1b-

Abwertung aufgrund geringer Präzision der Ergebnisse

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In summary, the results of our randomized pilot study suggest that PET-CT is not superior to CE-CT in detecting cancer recurrence during the 2 years after curative-intent treatment of NSCLC. Sensitivity, specificity, and positive predictive value were comparable between the two imaging procedures. In addition, the number of potentially curable recurrences was similar in the two groups. If imaging investigations are discussed at an interdisciplinary board, both methods may thus have a similar performance in post-therapeutic surveillance of lung cancer. Considering the substantially higher costs associated with PET-CT compared with CECT, the latter appears to be the preferred option for routine surveillance (at least short-term) after curative intent treatment of NSCLC. However, some patients with a specific tumor stage or histopathology may require PET-CT follow-up based on their individual characteristics“

Schlussfolgerung des Begutachters: Randomisierte kontrollierte Pilotstudie mit moderatem Verzerrungspotential (fehlende Informationen zur Zuweisung der Behandlung, keine Verblindung der Patienten, Ärzte und Endpunkterfasser), registriertes Protokoll, alle geplanten Endpunkte werden berichtet, kein Drop-out), welche Ergebnisse zu klinischen Endpunkten und zur diagnostische Güte einer Nachbeobachtung von insgesamt 96  Patienten mit kurativer Behandlung eines NSCLC über 2 Jahre berichtet (aufgrund der kleinen Fallzahl Abwertung aufgrund geringer Präzision).

Matucci 2019

(24)

Search until 06/2019

Systematic Review

treatment decision from staging and diagnostic accuracy

of molecular imaging using 18F-FDG PET/CT in staging

patients with SCLC

Inclusion criteria:

·      patients with histologically proven SCLC;

·       18F-FDG PET/CT performed for disease staging compared to staging without 18F-FDG PET/CT

Exclusion criteria:

o   Use of 18F-FDG PET only without hybrid technology for staging,

o   studies evaluating the prognostic value of 18F-FDG PET/CT or its use for radiotherapy planning

o   articles evaluating the role of 18F-FDG PET/CT for treatment response assessment or restaging after treatment

o   review articles, letters, comments, editorials, conference proceedings

·      case reports or small case series.

Search in 3 databases

N=9 (1 RCT, 2 prospective, 6 retrospective)

1 multicenter, 8 single center

n=791 patients (23 to 309)

median age: 58 to 69.6 years

males: 22 to 90 %

Quality assessment (QUADAS-2 (4)):

·   bias in reference standard judged as high in 6 studies

·   bias in patient selection high in 3 studies

with no explanations

18F-FDG PET/CT performed for disease staging

vs.

Staging without 18F-FDG PET/CT (e.g., using conventional imaging methods:

contrast-enhanced CT or bone scintigraphy)

composite reference standard:  histology, further imaging, or  clinical /biochemical follow-up

Primary outcome: Change of binary SCLC staging using 18F-FDG PET/CT (number of SCLC patients with a change of staging obtained by 18F-FDG PET/CT (upstaging from limited to extended stage disease or downstaging) / overall number of SCLC patients who underwent the 18F-FDG PET/CT scan)

secondary outcomes:

Diagnostic accuracy

Survival

Diagnostic accuracy (N=3):

·    better in all studies with PET/CT:

·    prospective studies: 57/59 (96.6 %) vs. 54/59 (91.5 %) 19/20 (95 %) vs. 17/20 (85 %)

·    retrospective study: 23/23 (100 %) vs. 18/23 (78.3 %)

Change of binary SCLS staging (N=6, n=277):

·    range from 5.1 to 21.7 % with a pooled value of 15 % (95%-CI 9 to 21) with moderate heterogeneity between results of different studies (I²=50 %)

·    with lower change in prospective vs. retrospective studies: 11 % (95%-CI 2 to 25) vs. 17 % (95%-CI 12 to 23)

Survival (N=1 RCT, n=309): difference in survival was not shown

1a-

Abwertung aufgrund von Inkonsistenz

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “ 18F-FDG PET/CT is a very useful imaging method in staging SCLC patients, as it may change the binary SCLC stage, resulting in change of therapeutic management, in ∼15% of cases. Further well-designed studies on the use of this hybrid imaging method for SCLC staging are needed.”

Schlussfolgerung des Begutachters: Systematische Übersicht mit geringen Einschränkungen der Qualität (Registrierung des Protokolls erst bei Einreichung, keine Beschreibung des Vorgehens bei Datenextraktion und Qualitätsbewertung, Schwierigkeit der Interpretation von Tests auf Publikationsbias bei 6 Studien), welche Therapieänderungen und diagnostische Güte einer PET-CT-Untersuchung zur Differenzierung zwischen limited (confined to the ipsilateral hemithorax including contralateral mediastinal and ipsilateral supraclavicular lymph nodal metastases and can be included within a radiation field). und extended stage disease (spread beyond the ipsilateral hemithorax including hematogenous metastases and malignant pleural or pericardial effusion) beschreibt.  Die Übersichtsarbeit schließt insgesamt 9 Studien ein (davon 1 RCT und 2 weitere prospektive Kohortenstudien). In die Metaanalyse zur Häufigkeit des Therapiewechsels gehen Ergebnisse aus 6 Studien ein, deren moderate Heterogenität zwischen den Ergebnissen der Einzelstudien nur teilweise erklärt werden konnte (daher Abwertung aufgrund von Inkonsistenz).

Mitchell 2016

(25)

Search from 2000 to 06/2015

Systematic Review

assessment for the efficacy and diagnostic performance of different imaging modalities for the pre-treatment staging of SCLC

Inclusion criteria:

o   full-text article in a  peer-reviewed, published 2000 to 2014

o   diagnostic performance: comparison to a common reference standard.

o   comparative effectiveness:  different tests or test strategies must have been given to separate groups of patients.

o   ≥ 85% of patients must have been undergoing staging for SCLC.

o   ≥  85% of patients must have been ≥  18 (or separate reporting)

o   CT performed on a multidetector scanner.

o   PET/CT must have been performed on an integrated PET/CT device

o   ≥ 10 patients per imaging test or test strategy participated.

o   Data reported for ≥  50% of the patients initially enrolled.

o   diagnostic performance:  sufficient data to allow calculation of sensitivity and specificity with their 95%-CI

o   comparative effectiveness:  the choice and/or timeliness of treatment or the outcomes of treatment must be reported for each study group.

Exclusion criteria:

o   Studies for the staging of recurrent SCLC

o   Data on imaging tests performed after any form of treatment

Search in 3 databases

N=7

Quality assessment (modified QUADAS-2 (4)):

presence of metastases to osseous structures

Any direct comparisons of imaging tests  or any combination:

• MDCT

• PET/CT

• MRI

• PET/MRI

• EBUS

• EUS-FNA

• Bone scintigraphy

Test concordance

diagnostic accuracy: Sensitivity  Specificity  Positive-predictive value, Negative-predictive value  Positive or negative likelihood ratio

Choice of treatment (eg, surgery, chemotherapy, or radiation)

Timeliness of treatment

Tumor response

Harms due to overtreatment or undertreatment

 Survival

 Quality of life

Presence of metastases to osseous structures:

PET-CT vs. scintigraphy (N=2 with moderate risk of bias, n=123, GRADE: low):

·    Sensitivity: favours PET/CT: 100 (95%-CI 88 to 100) vs. 37 % (95%-CI 22 to 55), but no difference in second study

·    Specificity: no difference in first study, favours PET-CT in second study: 100 (95%-CI 83 to 100) vs. 58% (95%-CI 26 to 77)

·    PET/CT is more sensitive than scintigraphy

PET-CT vs. CT (N=1 with moderate risk of bias, n=29, GRADE: low):

·    Sensitivity: favours PET/CT: 80 (95%-CI 49 to 94) vs. 30 (95%-CI 11 to 60)

·    specificity = 100% in both groups

·    PET/CT more sensitive than CT

Presence of any distant metastases:

PET/CT vs. standard staging (N=1 with high risk of bias, n=73):

·    Sensitivity: 92 (95%-CI 76 to 98) vs. 46 % (29 to 65)

·    Specificity: 96 (95%-CI 92 to 100) vs. 100 % (95%-CI 86 to 99)

·    Standard staging plus PET/CT is more sensitive than standard staging alone (GRADE: low)

2a

Abwertung aufgrund von Studienlimitationen und geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “First, compared to MDCT and bone scintigraphy (imaging modalities commonly used for staging), FDGPET/CT is more sensitive for detecting osseous metastases in patients with SCLC.

Second, for patients with SCLC who have undergone standard staging, the addition of FDGPET/CT increases sensitivity for detecting distant metastases.

Due to the sparse evidence available, we were unable to draw conclusions about the comparative specificity of FDGPET/CT compared to these other modalities. “

…"Our results suggest potential resource implications. Although FDG-PET/CT may be more expensive than CT and bone scintigraphy, some patients may undergo all three tests

when being evaluated for osseous and other distant metastases outside the brain. Based on the current review of literature, we determine that eliminating routine use of bone scintigraphy from SCLC staging protocols in favor FDG-PET/CT could potentially result in significant cost savings without loss of sensitivity.“

Schlussfolgerung des Begutachters: Systematische Übersicht mit geringen Einschränkungen der methodischen Qualität, welche sich aus der sehr breiten Fragestellung (Bildgebung zum Staging vor der Behandlung bei SCLC) ergeben (geringe Anzahl von Studien zu heterogenen Themen). Es lagen ausschließlich Ergebnisse zur diagnostischen Genauigkeit vor, von diesen wurden die Ergebnisse extrahiert, aus denen sich Schlussfolgerungen zum Einsatz einer PET-CT ergeben. Ergebnisse zur Differenzierung von limited und extended disease wurden nicht extrahiert, da die Studie in Matucci 2020 eingeschlossen wurde. Die Autoren bewerten den Evidenzgrad mit low (2 Abwertungsstufen). Abwertungen basieren auf dem moderaten Verzerrungspotential in den Studien und der geringen Präzision der Ergebnisse.

Wu 2017

(26)

Search until 01/2017

Systematic Review

diagnostic accuracy of 18F-FDG PET/CT for the detection of adrenal metastasis in lung cancer

Inclusion criteria:

·  studies or subsets in studies on the diagnostic performance of 18F-FDG PET/CT in the assessment of adrenal mass in lung cancer

·  studies that used histopathologic examination and/or close clinical and radiologic follow-up for ≥ 6 months as the reference standard for confirmation of malignant or benign adrenal masses

·  for per-lesion statistics, sufficient information to calculate the true-positive (TP), false-positive (FP), true- negative (TN), and false-negative (FN) values

Exclusion criteria:

·  non-original research articles with limited data, such as letters, case reports, reviews, comments, and conference abstracts

·  studies with potentially over- lapping data

·  studies with < 10 analyzable lesions

Search in 5 databases

Inclusion of N=9 studies ( 1 prospective, 8 retrospective, 3 with consecutive recruitment)

n=707 patients (with n=810 adrenal masses from lung cancer; benign: 50.7%; metastatic: 49.3%

Quality assessment (QUADAS (28)): QUADAS scores> 10: N=6 no conclusive evidence of a threshold effect for 18F-FDG PET/CT

18F-FDG PET/CT

For diagnostic accuracy of adrenal mass in lung cancer

vs.

Biopsy/ Radiological Follow-Up

Sensitivity

Specificity

Positive likelihood ratio

Negative likelihood ratio

Diagnostic odds ratio

Sensitivity:

·    range from 71 to 97 % with a pooled value of 88.7 % (95%-CI 85.2 to 91.7) and substantial heterogeneity between results of different studies (I²=80.8 %)

·    and no differences depending on sample size, consecutive patient recruitment or QUADAS-score

Specificity:

·    range from 81 to 100 % with a pooled value of 90.8 % (95%-CI 87.5 to 93.4) and small heterogeneity between results of different studies (I²=26.4 %)

·    and no differences depending on sample size, consecutive patient recruitment or QUADAS-score

Positive Likelihood Ratio: 8.552 (95%-CI 6.197–11.802) and small heterogeneity between results of different studies (I²=9.7 %)

Negative Likelihood Ratio: 0.090 (95%- CI 0.039–0.211) and substantial heterogeneity between results of different studies (I²=83.9 %)

Diagnostic odds ratio: 96.825 (95%-CI 40.402 to 232.05)

2a

Abwertung aufgrund von Studienlimitationen und Inkonsistenzen

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: „Our study showed that 18F-FDG PET/CT plays a vital role in the differentiation of adrenal metastasis from a benign adrenal mass in lung cancer patients, with excellent diagnostic performance. The literature focusing on the use of 18F-FDG PET/CT in this setting still remains limited to retrospective studies and prospective studies are needed in future.“

Schlussfolgerung des Begutachters: Systematische Übersichtsarbeit mit geringer Einschränkung der Qualität (kein Studienprotokoll, unklare doppelte Studienextraktion, keine Ausführliche Darstellung des Verzerrungsrisikos der eingeschlossenen Studien, trotz Nutzung eines passenden Tools). Die Anwendbarkeit von 18F-FDG PET/CT für die Diagnose und Differenzierung von benignen und metastatischen Nebennierentumoren bei Lungenkarzinom wurde bewertet. Es wurden 9 Studien (davon 8 retrospektiv) mit 707 Patienten und 810 Proben eingeschlossen. 18F-FDG PET/CT stellt mit einer Sensitivität von 89 % (95%-KI 85 - 92) und einer Spezifität von 91 (95%-KI 88 - 93) ein gutes Mittel zur Diagnostik dar, wobei die Ergebnisse der Studien eine bedeutsame Heterogenität aufweisen (daher Abwertung aufgrund von Inkonsistenzen) Es erfolgte eine Abwertung aufgrund von Studienlimitationen da die Ergebnisse zum Großteil auf retrospektiven Studien basieren und genauere Angaben zum Verzerrungsrisiko fehlen.

CI: confidence interval; DWI: Diffusion weighted imaging; CT: N: number of studies; n: number of participants; NSCLC: Non-small cell lung cancer;  OR: Odds Ratio; RoB: Risk of bias; RR: Relatives Risiko

Patienten

Patienten (m/w) mit NSCLC im Stadium I/II, bei denen aufgrund eingeschränkter   funktioneller Operabilität keine Lobektomie, jedoch eine parenchymsparende (sublobäre) Resektion durchgeführt werden kann (Segment-/Keilresektion)

Intervention

stereotaktische Strahlentherapie

Kontrolle

Segmentresektion oder

Keilresektion (separate Betrachtung)

Zielkriterien (Outcomes)

rezidivfreies Überleben

Gesamtüberleben

Komplikationsraten (incl. Letalität) innerhalb von 90 Tagen postinterventionell

Studiendesign

Systematische Übersichtsarbeiten mit Volltextveröffentlichung ab 2019

Propensity-Matched- Studien und multizentrische (prospektive und retrospektive) Kohortenstudien mit Volltextveröffentlichung ab 2008

NSCLC: Non-small cell lung cancer; RCT: Randomisierte kontrollierte Studie; SBRT: VATS: video-assistierte thoraskopische Lobektomie

Cao 2019

(12)

Search in 2 databases until 01/2018

Systematic review

Efficacy of stereotactic body radiation therapy (SBRT) or surgery on overall survival for patients with stage I NSCLC

Inclusion Criteria:

· comparative OS for patients with NSCLC

· who underwent SBRT or surgical resection for NSCLC

Exclusion Criteria:

· Abstracts, case reports, conference

· presentations, editorials, expert opinions, and publications not written in English

N=32 (2 RCTs, 24 propensity-matched cohort studies, balancing a number of covariables on patient characteristics, preoperative risk factors, and tumor characteristics)

N=23 selected for quantitative analysis

age (IQR): 71-78 vs. 68-78 yrs

adenocarcinoma and

SCC were the most common types

Up to 70% of patients who underwent SBRT did

not have a pretreatment pathologic diagnosis of NSCLC with differences between Europe and the US

clinical staging:

IA: 71-84 vs. 70-82 %

IB: 16-29 vs. 18-34 %

SBRT (varied in dosage and fractions among centers and within each institution,

depending on the location, size, and type of the tumor)

vs.

surgery (subgroup with sublobular resection)

Primary:

overall survival

Secondary:

cancer-specific

survival

disease-free survival

freedom from locoregional

recurrence

freedom from distant recurrence

perioperative mortality and morbidity

only results from confounder-adjusted analyses were extracted

Overall survival (N=6 with >n=3000 patients  compared SBRT with sublobular resection without matching):

·       Better survival after sublobar resection vs. SBRT: OR: 1.54 (95% CI 1.36-1.75; p2=32 %), all studies stated better survival with surgery

2a-

Abwertung aufgrund von sehr schwer-wiegenden Studienlimitationen

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “The current evidence suggests surgery is superior to stereotactic body radiation therapy in terms of mid- and long-term clinical outcomes; stereotactic body radiation therapy is associated with lower perioperative mortality. However, the improved outcomes after surgery may be due at least in part to an imbalance of baseline characteristics. Future studies should aim to provide histopathologic confirmation of malignancy and compare stereotactic body radiation therapy with minimally invasive anatomical resections.”

Schlussfolgerung der Begutachterin: Systematische Übersicht zum Vergleich einer SBRT und einer Operation mit einer Suche bis Januar 2018. Einschluss von 32 Studien mit Studienteilnehmern mit NSCLC im Stadium I (davon 2 randomisierte Studien und 23 Studien mit Propensity-Score-Matching). 6 Studien berichteten Ergebnisse zum Vergleich einer des Gesamtüberlebens von Patienten, welche mit sublobulärer Resektion oder SBRT behandelt wurden. Es liegen keine Ergebnisse aus gematchten Analysen vor. Die systematische Übersichtsarbeit weist eine moderate Einschränkung der Qualität auf (fehlendes Protokoll, fehlende Informationen zum doppelten Screenen und Gründe zum Ausschluss potentiell relevanter Volltexte fehlen, keine Heterogenitätsbetrachtungen und Untersuchungen zum Publikationsbias). Aufgrund fehlender Informationen zur Adjustierung sollte von einem Selektionsbias ausgegangen werden (daher Abwertung aufgrund sehr schwerwiegender Studienlimitationen). Die Ergebnisse bestätigen ein besseres Überleben der Patienten nach sublobulärer Resektion (ohne weitere Differenzierung) im Vergleich zur SBRT bei moderater Heterogenität der Ergebnisse der Einzelstudien, wobei alle Studien einen Vorteil für die resektierten Patienten zeigen. Insgesamt wird die Qualität der Evidenz wegen sehr schwerwiegender Studienlimitationen (Systematische Übersicht auf Grundlage von Kohortenstudien) auf 2a- abgewertet.

Crabtree 2013

(16)

US

RTOG 0236: opened in 2004

ACOSOG Z4032: closed 01/2010

Pooled individual patient data analysis of 3 prospective clinical trials (RTOG0236, ACOSOG4032, ASOGZ4033)

Comparison of the selection criteria and short-term outcomes of SBRT, sublobar resection and radiofrequency ablation

Inclusion criteria:

· High-risk and inoperable patient population with early-stage NSCLC

· RTOG0236: phase II, multicenter, North American trial on the efficacy of SBRT in inoperable stage I NSCLC patients (n=59) with results for local control, treatment-related toxicity, disease-free survival, overall survival, and patterns of failure

· ACOSOG Z4032: RCT on sublobar resection vs. sublobar resection plus brachytherapy in stage I NSCLC (n=222) who were considered too high risk for surgery with results on time to local recurrence, morbidity, overall and failure-free survival, regional or distant recurrence, quality of life, pulmonary function and Adverse events

· ACOSOG Z4033: trial on the efficacy of RFA in high-risk patients with stage IA NSCLC on 2-yr survival,  local, regional, and distant recurrence and procedure-related morbidity and mortality after radiofrequency ablation

Propensity-score analysis SBRT vs. sublobular resection for patients with clinical T1-tumors from RTOG 0236 (n=32) and ACOSOG Z4032 (n=202) with available pulmonary function test

SBRT (n=55)

(3 fractions of 20 Gy each within 8 to 14 days)

vs.

sublobar resection (n=211)

vs.

radiofrequency ablation (n=51)*

*not extracted

Early mortality

Adverse events

propensity-matched analysis among patients with clinical stage 1A in RTOG 0236 and ACOSOG Z4032 using age, PS, FEV1, and DLCO.

Adverse events: no differences in:

· No treatment-related mortality

· 30-day overall mortality: 0/55 vs. 3/211

· 90-day overall mortality: 0/55 vs. 5/211

· Grade ≥3/5 after 30 days in 9.1/0 vs. 28.0 /1.4% of patients

· Grade ≥3/5 after 90 days in 21.8/0 vs. 33.2/1.4 % of patients

no difference in frequency of AEs≥ 3 between matched cohorts with sublobular resection (n=202) and SBRT (n=32) at

· 30-days: OR 2.37 (95 %-CI 0.75-9.90; p=0.18)

· 90-day: OR 1.92 (95% 0.71-6.08; p=0.25)

· patients were divided into 5 strata with different risks of AEs for patients with sublobular resection:

· stratum 1 (n=42, median age: 70 yrs, DLCO: 30%, FEV1:was 39 %):

o   90-day grade ≥ 3 AE: 48 %

o   90-day mortality:  2.4

· strata 5 (n=34, median age: 73 yrs, DLCO: 69 %, FEV1: 56 %)

o   90-day grade ≥ 3 AE: 24%

o   no mortality at 90 days

2b-

Abwertung aufgrund von geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Among appropriately matched patients, no difference was seen in early morbidity between sublobar resection and stereotactic body radiotherapy. These results underscore the need for a randomized trial to delineate the relative survival benefit of each modality and to help stratify patients considered high risk.”

Schlussfolgerung der Begutachterin: Gemeinsame Auswertung der individuellen Patientendaten von drei prospektiv geplanten bis dahin unveröffentlichten multizentrischen Kohortenstudien (wobei eine einarmige und eine randomisierte Studie dem Thema entsprachen) auf Basis von in den Jahren 2004-2010 in den USA behandelten nicht-operablen NSCLC-Patienten im klinischen Stadium IA. Es wurden innerhalb von 90 Tagen auftretende Ergebnisse zu Morbidität und Mortalität von 32 mit SBRT behandelte Patienten mit 202 mit einer sublobularen Operation behandelten Patienten verglichen, nachdem die Patienten nach Alter, Performance Status und Lungenfunktionsparametern (FEV1, DLCO) gematcht wurden. Nach der Operation wurden innerhalb von 30 und 90 Tagen nach Studienbeginn etwas mehr Nebenwirkungen (CTC-AE, Version 3.0 ≥ 3) beobachtet. Da die Patienten in verschiedene Studien eingeschlossen wurden, kann auch hier ein Selektionsbias aufgrund nicht erfasster Konfounder nicht sicher ausgeschlossen werden. Zusätzlich ist die geringe Grüße der mit SBRT behandelten Gruppe und die geringe Anzahl der Ereignisse zu beachten (daher Abwertung aufgrund geringer Präzision).

Grills 2010

(17)

USA

Treated 02/2003-8/2008

Single-center retrospective study with confounder-adjustment

Comparison of outcomes between SBRT and wedge resection for

stage I NSCLC

Inclusion criteria:

· Stage I NSCLC (T1-2N0M0)

· Patients who underwent SBRT were treated on a prospective phase II trial from 11/2005

· Wedge resection cases were consecutively selected since 2003

· primary tumors with or without a pathologically microscopic satellite lesion and without nodal or distant metastasis)

· ineligible for anatomic lobectomy (95 % were medically inoperable, 1 refused surgery).

Exclusion criteria:

Patients with completion lobectomy (thoracic surgeon discretion)

n=124

mean age: 78 vs. 74 yrs

females: 60 %

smoker: 98 %

Adenocarcinoma/SCC/ Adenosquamous cell carcinoma: 62 vs.65/33 vs. 25/0 vs. 10 %

T1N0: 71 vs. 81 %

T2N0: 29 vs. 19 %

Mean FEV1: 1.31 vs. 1.39 L

diffusing capacity of lung for carbon monoxide were: 10.14 vs. 12.0 mL/min/mmHg

Mean CCI: 4 vs. 3

Image-guided SBRT (n=58)

SBRT was volumetrically prescribed as 48 (T1) or 60 (T2) Gy in 4-5 fractions

vs.

wedge resection (n=69) with curative intent and a goal of negative oncologic margins (52 % with VATS)

Survival

Recurrences (local, regional, distant)

Complications (NCT CTC, Version 3)

Median potential follow-up is 2.5 years

Overall survival at 30 months:

·       higher with wedge: 72 vs. 87 % (p=0.01), without pathological T4: 70 vs. 85 % (p=0.02); OR 3.32 (95 %CI 1.61-6.86)

Cause-specific survival at 30 months:

No differences shown: 93 vs. 94 %

Recurrences

No differences stated after 30-months for

·       regional recurrence: 4 vs. 18 %, without pathological T4: 0 vs. 18 % (p=0.07)

·       distant metastasis: 19 vs. 21 %

·       freedom from any failure: 77 vs. 65 %

a trend toward reduced local recurrences with SBRT: 4 vs. 20 %, without pathological T4: 5 vs. 24 % (p=0.05)

·       locoregional: 9 vs. 27 %, without pathological T4: 5 vs. 29 % (p=0.03)

Confounder-adjusted analyses:

·       After SBRT: only histology and presence of a synchronous primary tumor predicted DM after SBRT.

·       After wedge resection: visceral pleural invasion and stage (IB) predicted DM

·       Median time to LR: 10.5 vs. 25 months

·       median time to RR or DM:  9 vs. 25 months

Complications:

SBRT:

·       ≥ grade 2 radiation pneumonitis:  11% (9% grade 2, 2%grade 3), no grade 4 or 5 events.

·       temporarily required corticosteroids: 11 %

·       Rib fractures: 11 % (7% symptomatic needing pain medication)

·       grade 1 skin toxicities: 38%

·       Acute or chronic treatment-related myositis:  4% to 10% and related to chest wall proximity

·       No treatment-related deaths

Wedge resection:

·       median length of hospital stay: 5 days (range, 0 to 29 days)

·       Readmission ≤ 30 days postoperatively: 10 %

·       chest tube at discharge: 3 %

·       postoperative pneumonia, empyema, and chylothorax: 3%, 2%, and 2 %

·       postoperative wound infections / arrhythmias: 3% 6 %

·       30-day mortality: 0 %.

3b

Abwertung aufgrund von Studien-limitationen und geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In conclusion, both lung SBRT and wedge resection are reasonable options for patients with stage I NSCLC who are ineligible for lobectomy. SBRT reduced LR, RR, and LRR. As would be expected for a nonrandomized population of patients selected for surgery versus RT (medically inoperable) at physician discretion, OS was higher after surgery, but CSS was identical. Patients who underwent SBRT had fewer serious treatment-related complications and rarely required hospitalization.

Schlussfolgerung der Begutachterin: Monozentrische US-amerikanische retrospektive Kohortenstudie auf Basis von prospektiv gesammelten einer Phase II-Studie zur Behandlung von 55 NSCLC-Patienten im klinischen Stadium IA mit SBRT sowie einem Register aller 69 mit Keilresektion ab 02/2003 operierten Patienten (aufgrund der geringen Fallzahl und monozentrischen Durchführung Abwertung aufgrund geringer Präzision). Eine Konfounder-Adjustierung erfolgte allein zur Ermittlung prognostischer Faktoren für das Überleben in beiden Behandlungsgruppen ohne Vergleich der Behandlungsgruppen (daher Abwertung aufgrund schwerwiegender Studienlimitationen). Es konnte ein Vorteil im Überleben nach 2 Jahren nach Keilresektion (ohne Konfounderadjustierung) nachgewiesen werden. SBRT reduzierte das Auftreten von Rekurrenzen (trotz der deutlich höheren Anzahl untersuchter Lymphknoten), während die mediane Zeit bis zum Auftreten der Rekurrenzen bei den mit SBRT behandelten Patienten kürzer war. Behandlungsspezifische Komplikationen nach der Behandlung werden beschrieben, wobei keine durch die Behandlung bedingten Todesfälle auftraten. Patienten, welche mit SBRT behandelt wurden, waren älter, wiesen mehr Komorbiditäten auf, so dass von einem Selektionsbias auszugehen ist Das Staging basiert in der SBRT-Gruppe auf klinischen (via Mediastinoskopie) und in der operierten Gruppe auf pathologischen Kriterien. Die Studie wurde in die systematische Übersichtsarbeit von Cao 2019 eingeschlossen und hier zusätzlich extrahiert, da Ergebnisse zu mit Keilresektion operierter Patienten vorliegen.

Matsuo 2014

(18)

Japan

01/2003-12/2009

Monozentrische propensity-adjusted retrospective cohort study

Efficacy of SBRT vs. sublobular resection on survival pf patients with stage I NSCLC

Inclusion criteria:

· Consecutive patients who underwent SBRT or sublobular resection because of medical comorbidities for clinical stage I NSCLC

· Clinical stage was determined with CT and 18-FDG-PET or CT alone

Exclusion criteria:

· tumour diameter >50 mm or

· those without histological confirmation of NSCLC

n=180, 2x53 patients were matched with

median age: 76 years

PS 0–1

median tumour size: ≈20 mm

median FEV1 ≈1.8 L

median CCI = 1

SBRT (n=115)

48 Gy in 4 fractions to the isocentre for a peripheral tumour and 60 Gy in 8 fractions for a centrally located tumour

since 06/2006, the dose was increased to 56 Gy in 4 fractions for a peripheral tumour with a diameter > 30 mm.

vs.

Sublobular resection (n=65)

via VATS,

surgery as wedge resection (n=39)

or segmentectomy (n=26) at discretion of the surgeon with generous margins

median follow-up periods: 

6.7 vs. 5.3 yrs

Overall survival

Recurrences:

Local: primary tumour recurrence and recurrence in the involved lobe

Regional: ipsilateral nodal recurrence in hilum or mediastinum, or as recurrence in the ipsilateral lung

Distant: other recurrence

Propensity-score matching according to age, gender, PS,

tumour diameter, FEV1 and CCI

Overall survival: no differences shown, slightly better with resection

·       No treatment related deaths

·       5-yr OS: 40.4 % (95 %CI 26.6-53.7) vs. 55.6% (95%CI 40.2–68.5)

·       Median OS: 3.8 vs. 5.3 yrs

Cause-specific death: similar between groups:

35.3% (95%CI 22.4–48.5) vs. 30.3% (95%CI, 17.8–43.7%)

Recurrences:

·       Local: slightly higher with SBRT after 5 yrs: 28.3 vs. 14.1% (p=0.059)

·       Regional: 14.3 vs. 9.2% (p=0.241)

·       Distant: 35.9 vs. 36.1% (p=0.674)

3b

Abwertung aufgrund von Studien-limitationen und geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In conclusion, SBRT achieved comparable cause-specific survival to sublobular resection in the PSM cohort. SBRT can be an alternative treatment option to sublobular resection in high-risk patients who cannot tolerate lobectomy because of medical comorbidities.”

Schlussfolgerung der Begutachterin: US-amerikanische retrospektive Kohortenstudie auf Basis von konsekutiv in den Jahren 2003-2009 monozentrisch in Japan behandelter NSCLC-Patienten im klinischen Stadium I. Es wurden 115 Patienten mit SBRT und 35 mit einer sublobularen Operation behandelt (davon 39 mit Keilresektion und 26 mit Segmentektomie) und im Median über 5 Jahre beobachtet. Insgesamt konnten 2x53 Patienten nach Alter, Geschlecht, Performance Status, Tumordurchmesser, FEV1 und Komorbiditätsscore gematcht werden (aufgrund der geringen Teilnehmeranzahl und monozentrischen Durchführung Abwertung wegen geringer Präzision). Nach SBRT wurden über einen Nachbeobachtungszeitraum von 5 Jahren etwas mehr lokale Rekurrenzen und Todesfälle beobachtet. Auch in dieser Studie kann ein Selektionsbias aufgrund nicht erfasster Konfounder nicht sicher ausgeschlossen werden. Die Autoren beschreiben eine vollständige histologische Bestätigung des NSCLC bei allen mit SBRT, aber nicht bei allen operierten Patienten (daher Abwertung aufgrund von Studienlimitationen).

Paul 2016

(19)

US, SEER database

10/2007.06/2012, follow-up until 21/2013

Retrospective cohort study with propensity matched comparative analysis Comparison of cancer specific survival after sublobar resection vs. SABR for tumors ≤ 2 cm and sublobar resection or lobectomy vs. SABR for tumors ≤5 cm

Inclusion criteria:

patients (≥ 66 yrs) who underwent SABR or lung resection (open or VATS sublobar resection or lobectomy)

treatment during 10/ 2007 to 06/2012 from one month before to six months after diagnosis of a first primary NSCLC (ICD-9-CM)

stage I (T1 or T2, N0, M0) cancer, with tumor size ≤5 cm.

definitive surgery bases on the most extensive procedure

Exclusion criteria:

Patients with multiple cancers or a second diagnosis of lung cancer

patients without pathological confirmation

patients who received CT before SABR or lung resection

n=690 (primary analysis)

Mean age: 76 yrs

Women: 63 %

Whites: 90 %

Patients undergoing SABR were older (77.6 v 75.6 yrs), had a higher prevalence of chronic pulmonary disease (88.7 vs. 80.2 %) and congestive HF (28.4 vs. 22.7 %), more T1 (94.9 vs. 83.9 %), more SCC (41.1 vs. 26.3 %) and adenocarcinoma (52.4 vs. 47.5 %), but less other histology (6.6 vs. 26.3 %), more undetermined tumor grade ( 52 vs. 7.2 %), higher tumour size: 0-10 /11-20 mm: 8.7/91.3 % vs. 24.8/75.2 %

Primary analysis:

SABR (n=275, 39.9%)

vs. thoracoscopic sublobar lung resection (n=415, 60.1%) (363 (87.5 %) with wedge or 52 (12.5 %) with segmental resections) for a tumor sized ≤2 cm

Secondary analysis:

SABR (n=714)

vs.  thoracoscopic resection* (n=2253)

(562 (24.9%) with wedge resection, 92 (4.1%) with segmental resection, and 1599 (71.0%) lobectomy) for tumors sized ≤5 cm

mean follow-up: 3 (0 to 6.25) yrs

*not extracted

Overall survival

Cancer specific survival

propensity score for receiving SABR

based on potential confounding variables (age, sex, race, marital status, income, residence location, tumor histology, size and stage, and comorbidities reported by Elixhauser et al., pre-procedural diagnostic test (mediastinoscopy or PET scan))

Overall survival:

Patients with tumor sized ≤2 cm (primary analysis):

·       144 (52.4%) vs. 138 (33.3%) died during follow-up

·       3-yr OS: 52.2 vs. 68.4 %

·       Propensity-matched cohort stated longer survival with

o      sublobular resection (n=233) (HR 1.80; 95 %CI 1.33-2.43; p<0.001)

o      wedge resection (n=185) (HR 1.71; 95 % CI 1.24-2.35; p=0.001)

o      Segmental resection (n=48) (HR 2.09; 95 %CI 1.12-3.89; p=0.02)

·        

Cancer specific survival:

·       Patients with tumor sized ≤2 cm (primary analysis): 37 (13.5%) vs. 44 (10.6%) died from lung cancer

·       3-yr CSS: 82.6 vs. 86.4 %

·       Propensity-matched cohort stated no difference between

o      Sublobular resection (n=233) and SABR (HR 1.32; 95 %CI 0.77-2.26; p=0.32)

o      Wedge resection (n=185) and SABR (HR 1.44; 95 %CI 0.79-2,62; p=0.23)

o      Segmental resection (n=48) vs. SABR (HR 1.43; 95 %CI 0.53-3.84; p=0.48)

2b-

Abwertung aufgrund von Studien-limitationen

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Our propensity matched analysis suggests that patients undergoing thoracoscopic surgical resection, particularly for larger tumors, might have improved cancer specific survival compared with patients undergoing SABR. Despite strategies used in study design and propensity matching analysis, there are inherent limitations to this observational analysis related to confounding, similar to most studies in healthcare that compare non-surgical technologies with surgery. As the adoption of SABR for the treatment of early stage operable lung cancer would be an important paradigm shift in lung cancer care, it warrants further thorough evaluation before widespread adoption in practice.”

Schlussfolgerung der Begutachterin: US-amerikanische retrospektive Kohortenstudie auf Basis von Daten SEER-Datenbank von älteren NSCLC -Patienten im klinischen Stadium 1, von denen in den Jahren 2007 bis 2012 415 Patienten mit sublobularer Resektion und 275 Patienten mit SABR behandelt wurden. Es wurden SABR-Patienten und vergleichbare Patienten mit sublobularer Resektion (Keilresektion oder Segmentresektion) gematcht (Propensity-Score-Matching). Die Vergleichbarkeit konnte für die im Register gespeicherten demografischen Informationen des Patienten (Alter, Geschlecht, Race, Beziehungsstatus, Wohnort), das Auftreten von Komorbiditäten, Tumoreigenschaften (Histologietyp, Tumorgröße und pathologisches Stadium) und vorhergehende diagnostische Tests hergestellt werden. Für Patienten mit einer Tumorgröße unter 2 cm konnte ein längeres Gesamtüberleben gezeigt werden, wobei Patienten mit Segmentresektion die beste Prognose haben. Komorbiditäten und der Gesamtzustand des Patienten wurden über den Score von Elixhauser et al. berücksichtigt. Ein Selektionsbias auf Grundlage nicht gemessener Konfounder (Herz-Lungenfunktion, Stärke der Komorbiditäten, funktioneller Status, Präferenzen des Patienten), welche Entscheidung zur Operation beeinflussen, kann nicht ausgeschlossen werden (daher Abwertung aufgrund von Studienlimitationen). Die Autoren beschreiben eine Unterschätzung des Stadiums bei SABR und häufig fehlende Informationen zum Staging der Lymphkonten bei operierten Patienten.

Port 2014

(20)

2001-2012

US: Prospective databases: thoracic surgical and a radiation oncology lung cancer database

Retrospective analysis of a single-center cohort study with propensity-score matching and confounder adjustment

Efficacy of SBRT vs. wedge resection on survival in stage IA NSCLC

Inclusion criteria:

patients who underwent wedge resection, wedge with brachytherapy or SBRT for clinical stage IA NSCLC

Exclusion criteria:

patients with pure ground glass opacities

n=164 patients, 99 were matched by age, sex and histology

PET in 83 vs. 93 %

Patients with SBRT were older: 76 (70-79) vs. 72 (66-77) yrs

males: 52 vs. 34 %

tumor size: 1.65±0.62 cm

pTNM (for wedge: TNxM/IA/IB/II/III): 3/74/13/5/4 %

histology:

Adeno/SCC(other: 82/15/3 %

SBRT (n=23)

· referral despite adequate cardiopulmonary function because the combination of prior lobar resection(s) and a new central lesion would have mandated a completion pneumonectomy or

· adequate cardiopulmonary reserve for a limited resection; but centrality of their lesions

· median dose to the tumor:  48 Gy (30-60 Gy) over 4(3-5) fractions

· Patients were monitored after treatment by CT or PET-CT every 6 months

vs.

wedge resection (n=76) in patients with peripherally located tumor with

· adequate pulmonary function, but with a history of prior lobar resection

· borderline cardiopulmonary reserve precluding lobar resection

· intraoperative brachytherapy determined by the surgeon’s assessment

· curative intent with a margin equal to the diameter of the tumor or ≥ 1 cm margin

Recurrence

local (recurrent tumor at the primary site, staple line, or involved lobe)

regional (in the ipsilateral uninvolved lobe or ipsilateral hilar and mediastinal nodes)

Distant (any disease in contralateral nodes or distant sites)

Perioperative data (extent of resection, 30-day operative mortality, complications according to NCI-CTC)

Follow-up by routine CT scan of the chest every 6 months,

if clinically warranted with PET

 until 11/2013

Overall survival:

·       3-yrs-OS: 75 % (95 %CI 56.2-94.2 %) vs. 87 % (95 %CI 77.5-96.3 %,

·       mortality higher with SBRT only: OR 3.46 (95 %CI 1.13-10.59) (adjusted for age, sex, histology and comorbidity)

Recurrence free survival:

3-yrs RFS:  higher after wedge resection: 72% (95% CI 51.9 to 91.1 vs. 88% (95% CI 78.3 to 96.7)

Disease-free survival:

No differences of 3-yrs DFS shown:  59% (95% CI 37.5 to 79.5) vs. 77% (95% CI 65.6 to 88.8 ) (p=0.066)

Recurrences:

·       Any: more frequent after SBRT: 7 (30.4 %) vs. 7 (9.2 %) (p=0.016)

·       Distant recurrence: 4 (17 %) vs. 4 (5 %)

·       Median time to recurrence : 14 (5-31) vs. 16 ( 8-32) months

Complications:

no mortality during SBRT or within 30 days after surgery

After SBRT:

·       Fatigue (grade I, 60.5 %)

·       Chest pain (grade II, n=4)

·       Pneumonia (n=1)

after surgery:  7 (9 %) patients:

·       prolonged air leak (grade I, n=3)

·       pneumonia (grade II, n=1)

·       arterial fibrillation (grade I-II, n=2)

·       pulmonary embolus (grade IV, n=1)

median length of stay: 5 (3-6) days

R

3b

Abwertung aufgrund von Studien-limitationen und geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Patients with clinical stage IA non-small cell lung cancer treated by SBRT appear to have higher overall disease recurrence than those treated by wedge resection. However, there was no significant difference in disease-free survival. A randomized trial is needed to define the role of SBRT in the potentially operable patient.”

Schlussfolgerung der Begutachterin: Monozentrische US-amerikanische retrospektive Kohortenstudie auf Basis von prospektiv gesammelten Daten der Jahre 2001 bis 2012 zur Behandlung von NSCLC-Patienten im klinischen Stadium IA mit SBRT sowie einer historischen Kontrollgruppe aus einem Register aller operierter Patienten. Insgesamt konnten 99 Patienten nach Alter, Geschlecht und Histologie gematcht werden, von denen 23 mit SBRT und 76 mit Keilresektion in den Jahren 2001 bis 2012 in einer Klinik in den USA behandelt wurden (aufgrund der geringen Fallzahl und monozentrischen Durchführung Abwertung aufgrund geringer Präzision). Auch nach dem Matching war die mit SBRT-behandelte Gruppe um mediane 4 Jahre älter. Nebenwirkungen ab dem Grad II traten in der SBRT-Gruppe bei 5 (21 %) und in der Operationsgruppe bei 3 (4 %) der Patienten auf. Nach SBRT wurden mehr Rekurrenzen beobachtet und es verstarben mehr mit SBRT behandelte Patienten. Für die Ergebnisse zum Gesamtüberleben erfolgte eine zusätzliche Adjustierung für Alter, Geschlecht, Histologie und das Auftreten von Komorbiditäten. Ein Selektionsbias aufgrund der Lage des Tumors (mehr zentrale Läsionen mit SBRT), des vorhandenen Altersunterschiedes und weiterer nicht erfasster Konfounder kann nicht ausgeschlossen werden und bei nicht-operierten Patienten ist eine Unterschätzung des Stadiums möglich, da hier sowohl Informationen zum pathologischen TNM-Staging als auch dem Grad der Differenzierung fehlen (daher Abwertung aufgrund von Studienlimitationen).

Shirvani 2012

(21)

US, SEER-Database

2001-2007

Retrospective cohort study with propensity-score matching

Comparison of survival outcomes of  lobetomy, sublobar resection, conventional radiation, SABR and observation on Survival

Inclusion criteria:

patients (≥ 66 yrs) without prior malignancy with stage I/II NSCLC and reported in the SEER-Medicare cohort

Exclusion critera:

· Patients with inadequate Medicare records

· any second cancer diagnosed within 120 days of the index lung cancer

· tumors > 5 cm, distant metastases or nodal disease at presentation

· absence of pathologic confirmation, and the use of non-standard therapies (CT, radiofrequency ablation, pneumonectomy, multimodality therapy)

n=10 923

median age: 75 yrs

females: 54.1 %

moderate-to severe comorbidity: 29 %

lobectomy (59%)

vs.

sublobar resection (n=1277, 11.7%)

vs.

conventional radiation (14.8%)

vs.

observation (12.6%)

vs.

SABR (1.1%, n=124)

median follow up: 3.2 years,

Overall survival

Lung-cancer specific survival

adjusted for pre-specified, clinically relevant patient, tumor, and treatment characteristics

exploratory analysis wherein propensity-score matching to compare SABR patients with matched controls (matched for race, gender, education level, median income, comorbidity score, histology, tumor grade, tumor size, and receipt of lymph node sampling)

Overall survival:

Unadjusted Analyses:

·       30-day mortality:  lowest for SBRT (0%) vs. conventional radiation (0.6%) vs. sublobar resection (1.2%) vs. lobectomy (1.3%) vs. observation (8.6%)

·       At 90-days: 0.8% vs. 4.1% vs. 4.1% vs. 5.6% vs. 20.7%

·       At 6 months: 4% vs. 6.5% vs. 7.8% vs. 13.8% vs. 32.8%,

·       At 2 years: lowest mortality for lobectomy (18.3%) vs. sublobar resection (25.1%) vs. SABR (41.1%) vs. conventional radiation (56.7%) vs. observation (73.4%)

Multivariate analyses:

Within the initial 6 months:

·       better OS (vs. lobectomy) with

·       best short-term OS with SABR: HR 0.48 (95 %CI 0.38-0.63)

·       Sublobular resection: HR 0.95 (95 %CI 0.86 -1.05)

After the initial 6 months:

·       Best results with lobectomy,

·       worse OS (vs. lobectomy) with

·       sublobular resection: HR 1.40 (95 %CI 1.28-1.54)

·       SABR: HR 1.56 (95 %CI 1.21-2.00)

Propensity-Matched Analyses: no difference between SABR and sublobular resection shown: HR 1.22 (95 %CI 0.79-1.89)

Lung-cancer specific survival (LCSS):

Multivariate analyses:

Within the initial 6 months:

·       best LCSS (vs. lobectomy) with

·       SABR: HR 0.59 (95 %CI 0.36-0.96)

·       Sublobular resection: HR 1.07 (95 %CI 0.91 -1.25)

After the initial 6 months:

·       worse LCSS (vs. lobectomy) with

·       Sublobular resection:  HR 1.55 (95 %CI 1.33-1.82)

·       SABR: HR 1.81 (95 %CI 1.11-2.95)

Propensity-Matched Analyses: no difference between SABR and sublobular resection shown: HR 0.47 (95 %CI 0.19-1.15)

2b-

Abwertung aufgrund von Studien-limitationen

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “In summary, our analysis of patients with early-stage NSCLC lung cancer in the contemporary period supports lobectomy as the optimal treatment strategy for fit older adults. Our findings also raise intriguing questions regarding the comparative effectiveness of SABR in certain patient subsets. Surgical intervention comes at the price of perioperative mortality and SABR may offer an alluring compromise, namely a lower risk of early periprocedural mortality with promising long-term survival outcomes.”

Schlussfolgerung der Begutachterin: US-amerikanische retrospektive Kohortenstudie auf Basis von Daten der US-amerikanischen SEE-Datenbank aus den Jahren 2001-2007 Zum Vergleich von 5 Behandlungsoptionen für NSCLC-Patienten im klinischen Stadium I und II. Es wurden 124 Patienten mit stereotaktischer Strahlentherapie und 1277 mit einer sublobularen Operation behandelt (ohne weitere Differenzierung) und im Median über 3,2 Jahre beobachtet. Es liegen Ergebnisse aus Konfounderadjustierten und propensity-gematchten Analysen vor, wobei die gematchten Analysen einen direkten Vergleich der mit stereotaktischer Strahlentherapie und Teilresektion behandelten Patienten ermöglichen. Insgesamt konnten 2x112 Patienten nach Race, Geschlecht, Bildung, Einkommen, Komorbiditätsscore, Tumoreigenschaften (Histologie, Tumorgrad, Größe) und vorhandener Lymphknotenbeprobung gematcht werden. Nach stereotaktischer Strahlentherapie wurde im Vergleich zur Teilresektion ein geringfügig (nicht-signifikant) schlechteres Gesamtüberleben und besseres lungenkrebsspezifisches Überleben beobachtet.  beobachtet. Auch in dieser Studie kann ein Selektionsbias aufgrund nicht erfasster Konfounder (Lungenfunktionsstatus und Gesamtzustand) nicht ausgeschlossen werden, da davon ausgegangen werden kann, dass Patienten mit besserer Lungenfunktion in der Regel operiert werden (daher Abwertung aufgrund von Studienlimitationen).

Varlotto 2013

(22)

US, databases from 5 institutions

1999-2008

Multicenter retrospective study with propensity-score matching

Effectiveness of SBRT vs. lobectomy and sublobular resection in stage I NSCLC

Inclusion criteria:

· lung cancer patients from 4 institutions who underwent surgical resection 1999-2008

· negative preoperative staging with PET-CT (no ct/pet performed in 87), negative lymph nodes at surgery, T1-T2 tumors and pathologically-confirmed squamous cell, adenocarcinoma, or NSCLC NOS histology

· did not received adjuvant or neoadjuvant CT or RT

· patients from 1 institution who underwent SBRT/SABR 2000-2008 received a prescription of a tumor dose ≥ 100 Gy₁₀10 and had biopsy-proven NSCLC, but did not undergo lymphadenectomy or lymph node staging,  mean dose: 60 Gy in 3 fractions (range: 48-60 Gy in 3 -5 fractions

n=317

SBRT/SABR patients were older (73.3 vs. 68.3 yrs), had higher mean tumor diameter (3.0 vs. 2.6 cm),  CCI (4.2 vs. 3.0) und proportion of NOS NSCLC (43 vs. 8 %) vs. surgery patients.

males: 48 vs. 55 %

DLCO<40/50 %: 39/60 % vs. 13/32 %

Lobectomy* (n=132)

vs.

wedge-resection (n=48)

vs.

SBRT/SABR (n=137)

Median follow-up: 25.8 (18.8 for SBRT vs. 30.0 months  for surgery) months (time from surgery or last SBRT)

*not extracted

survival

total recurrence control: absence of any recurrence (local, regional, or disseminated)

local and loco-regional control: absence of any recurrence in the ipsilateral lung, and the ipsilateral lung, bronchial stump/suture line, and N1-N3 nodal areas

propensity matching for pathology, age, tumor size and sex (all outcomes)

CCI or aspirin use (only OS)

Counfounder-adjustment for probability of treatment with SBRT/SABR, propensity-score matching for age, aspirin use, Charlson comorbidity score, diabetes, NSAID use, non–small cell pathology, percentage predicted FEV1, and hypertension

Overall survival

matched-pair comparison: better with SBRT (p=0.003)

2-yr OS: 50.8 vs. 94.1 %

3-yr OS: 42.3 vs. 86.3 %

5-yr OS: 31.7 vs. 86.3 %

· Multivariate analysis without propensity score (PS) correction correlated better OS with surgery, lower Charlson comorbidity score, and adenocarcinoma histology

· After adjustment for selection bias with propensity scores, OS correlated only with CCI (and the PS used for adjustment)

Total recurrence control:

matched-pair comparison: no difference shown (p=0.468)

2-yr OS: 95.2 vs. 94.1 %

3-yr OS: 78.0 vs. 86.3 %

5-yr OS: 78.0 vs. 86.3 %

· Without the propensity score, we associated improved TRC with surgery (vs. SBRT/SABR) and no other variables

· After adjustment for selection bias with propensity scores, tumor grade correlated with TRC (p=0.026), but not SBRT/SABR (p=0.636).

Locoregional control:

matched-pair comparison: no difference shown (p=0.059)

2-yr OS: 83.6 vs. 100 %

3-yr OS: 77.1 vs. 92.9 %

5-yr OS: 77.1 vs. 92.9 %

Dissiminated:

matched-pair comparison: no difference shown (p=0.819)

2-yr OS: 92.0 vs. 95.2 %

3-yr OS: 92.0 vs. 84.0 %

5-yr OS: 92.0 vs. 84.0 %

3b

Abwertung aufgrund von Studien-limitationen und geringer Präzision

Zusammenfassende Beurteilung

Schlussfolgerungen der Autoren der Studie: “Patients with stage I NSCLC treated with SBRT/SABR had similar TRC and LRC as patients treated with surgery but worse OS, on a matched-pair analysis. However, after adjustment for treatment selection, overall survival was no longer significantly worse for patients treated by SBRT/SABR. Our results suggest that randomized trials are needed to eliminate selection bias in treatment assignment in order to accurately compare outcomes between these approaches. We greatly encourage the completion of the ongoing prospective, randomized trials comparing SBRT/SABR to surgery.”

Schlussfolgerung der Begutachterin: US-amerikanische retrospektive Kohortenstudie unter Einschluss von NSCLC-Patienten im klinischen Stadium 1 und einer Behandlung mit SBRT aus einer Klinik mit langer Nachbeobachtungszeit (mediane 18,8 Monate) oder Keilresektion aus 4 Kliniken in den Jahren 1999-2008, wobei 137 Patienten mit SBRT und 48 Patienten mit Keilresektion behandelt wurden. Es werden Ergebnisse für gematchte Paare (Matching für Alter, Geschlecht, Tumorgröße und Pathologie, Komorbiditäten und Aspirineinnahme) berichtet, welche ein besseres Überleben für Patienten mit Keilresektion zeigen. Nach einer Propensity-Score-Matching für die Variablen, welche in dieser Studie als prädiktiv für die Wahl der Therapie waren, konnte kein Unterschied im Überleben oder Auftreten der Rekurrenzen in Abhängigkeit von der Therapie nachgewiesen werden.

Zu beachten ist die kleine Studiengröße (daher Abwertung aufgrund geringer Präzision), die explorative Bestimmung der Variablen für die Bestimmung des Propensity –Scores und die häufig ungenauere Bestimmung von prognostisch relevanten Informationen (Tumorgrad, Lymphknoteninvasion, Histologie) in der mit SBRT/SABR behandelten Patientengruppe. Ein Selektionsbias auf Grundlage nicht gemessener Konfounder (Lungenfunktion, Rauchen, funktioneller Status oder Ausmaß des Lymphknotenbefalls), welche Entscheidung zur Operation beeinflussen, kann nicht ausgeschlossen werden (daher Abwertung aufgrund von Studienlimitationen). Die Studie wurde in die systematische Übersichtsarbeit von Cao 2019 eingeschlossen und hier zusätzlich extrahiert, da Angaben zu mit Keilresektion operierten Patienten vorliegen.

Wu 2020

(23)

US, National Cancer database, 2004-2014

Retrospective cohort study with propensity-score matching

Efficacy of sublobular resection, SBRT or ablation for stage I NSCLC

Inclusion criteria:

patients treated with sublobular resection, ablation or SBRT for primary clinical stage I NSCLC (8th American Joint Committee on Cancer stage edition. Patients with American Joint Committee on Cancer) T1 through T2a tumors, lymph node-negative (N0), non-metastatic (M0), primary NSCLC

Exclusion criteria:

Patients who received CT, immune therapy, or hormone therapy

n=53 973

patients treated with SBRT (vs. SLR) were older (74±9 vs. 70±10 yrs), had more frequently SCC (34 vs. 25 %, a Charlson–Deyo score of 0 (58 vs. 46 %) and were diagnosed after 2009 (80 vs. 63 %) from non-metropolitan regions, and treated in facility located in central US (54 vs. 37 %)

males: 43 %

whites: 90 %

well and moderately /poorly differentiated: 24.8/19.5 vs. 63.1/27.1 %

SBRT (n=22 134)

vs.

ablation* (n=1388)

vs.

Sublobular resection (SLR)(n=30 451)

*not extracted

Primary: overall survival

With propensity-score matching based on age, sex, race, year of diagnosis, facility type, facility location, insurance status, income, education level, residence, Charlson–Deyo score, tumor size, histology, and pathologic grade

Complications:

30-day post-treatment unplanned hospital readmission rate among patients was greater with SLR vs. SBRT (4.0% vs 0.4% (p

Overall survival

After propensity-score matching for 2x9967 patients:

·       Shorter OS with SBRT vs. SLR:

·       Median OS: 40.5 vs 60.4 months; HR: 1.559 (95% CI1.497-1.623; p < .001)

·       Longer OS with wedge resection (median OS: 58 months) vs. SBRT (HR, 0.668; 95% CI 0.640-0.697; p < .001)

·       Longer OS with segmental resection (median OS: 71.4 months) vs. SBRT (HR 0.529; 95% CI 0.488-0.573; p <.001>

Subgroup with tumor size ≤ 2 cm (SBRT: n=5647, SLR: n= 5699):

·       Higher mortality with SBRT vs. SLR: median 45.0 vs. 67.5 months; HR 1.626 (95% CI 1.538-1.720; p <.001>

·       Longer OS with wedge resection (median OS: 64.3 months) vs. SBRT (HR 0.640; 95% CI 0.604-0.679; p < .001)

·       Longer OS with segmental resection (median OS: 77.5 months) vs. SBRT (HR 0.482; 95% CI, 0.426-0.546; p <.001>

2b-

Abwertung aufgrund von Studien-limitationen

Zusammenfassende Beurteilung