Real-World Outcomes of Stage IV NSCLC with PD-L1 ≥ 50% Treated with First-Line Pembrolizumab: Uptake of Second-Line Systemic Therapy

Round 1

Reviewer 1 Report

I find the results very interesting as they focus on the differences between real-world data and clinical trial data. However, to claim that second-line chemotherapy improves prognosis, additional data needs to be presented.

1: Please present the Kaplan-Meier curves of overall survival (OS) from the start of pembrolizumab treatment for the group that received second-line chemotherapy and the group that did not receive it. Please include the results of the log-rank test for comparison between the groups.

2: Please present the Kaplan-Meier curves of OS from the end time of pembrolizumab treatment for the group that received it and the group that did not receive it. Please include the results of the log-rank test for comparison between the groups.

Author Response

1: Please present the Kaplan-Meier curves of overall survival (OS) from the start of pembrolizumab treatment for the group that received second-line chemotherapy and the group that did not receive it. Please include the results of the log-rank test for comparison between the groups.

Thank you for this comment. The data had been included in the results (Median OS was 5.6 months (95% CI 4.8-6.5) in patients who did not receive additional therapy after progression and 22.2 months (95% CI 18.3-26.2) in patients who received subsequent therapy (p<0.001).) however we included the figure to further highlight the difference.

2: Please present the Kaplan-Meier curves of OS from the end time of pembrolizumab treatment for the group that received it and the group that did not receive it. Please include the results of the log-rank test for comparison between the groups.

Thank you for this suggestion to use the end of first line pembrolizumab as a proxy for disease progression that enables a comparison of the BSC and second line treated patient. The following was added to the methods and results respectively:

“Comparison of OS was made between the patients who received best supportive care and second line treatment using date of last pembrolizumab infusion as the baseline date, acting as a proxy for progression.”

Reviewer 2 Report

The present article is interesting because it focuses on the most recent questions about lung cancer we are facing, presenting a very large case series due to the real-world approach. Apart from its clinical interest, its novelty is controversial since the structure and the analysis is very similar to that conducted by Cortellini et al. (Cortellini A, Cannita K, Tiseo M, Cortinovis DL, Aerts JGJV, Baldessari C, Giusti R, Ferrara MG, D'Argento E, Grossi F, Guida A, Berardi R, Morabito A, Genova C, Antonuzzo L, Mazzoni F, De Toma A, Signorelli D, Gelibter A, Targato G, Rastelli F, Chiari R, Rocco D, Gori S, De Tursi M, Mansueto G, Zoratto F, Filetti M, Bracarda S, Citarella F, Russano M, Cantini L, Nigro O, Buti S, Minuti G, Landi L, Ricciardi S, Migliorino MR, Natalizio S, Simona C, De Filippis M, Metro G, Adamo V, Russo A, Spinelli GP, Di Maio M, Banna GL, Friedlaender A, Addeo A, Pinato DJ, Ficorella C, Porzio G. Post-progression outcomes of NSCLC patients with PD-L1 expression ≥ 50% receiving first-line single-agent pembrolizumab in a large multicentre real-world study. Eur J Cancer. 2021 May;148:24-35. doi: 10.1016/j.ejca.2021.02.005. Epub 2021 Mar 12. PMID: 33721704.) 

Moreover, I have some other questions: 

1. METHODS. It would be better to define the second line systemic therapy in this paragraph: e.g. Second line systemic therapy included carboplatin doublet, other chemotherapy and tyrosine kinase inhibitors for patients… 

2. In the Statistical section you said OS was calculated from date of treatment initiation using Kaplan-Meier curves and compared using log rank test. In the result section, however, there was no curve. It would be interesting to include the Kaplan Meier curve in the paper. 

3. Since it is still in doubt corticosteroids use may compromise the efficacy of immunotherapy, it would be interesting to include these data in your analysis. 

Author Response

1. METHODS. It would be better to define the second line systemic therapy in this paragraph: e.g. Second line systemic therapy included carboplatin doublet, other chemotherapy and tyrosine kinase inhibitors for patients… 

Thank you for this comment. The following has been added to the methods to improve clarity: 

“Second line therapy was defined as any systemic therapy including chemotherapy or targeted therapy delivered after first line pembrolizumab. Treatment duration was de-fined as time of first treatment administration to last treatment administration.”

2. In the Statistical section you said OS was calculated from date of treatment initiation using Kaplan-Meier curves and compared using log rank test. In the result section, however, there was no curve. It would be interesting to include the Kaplan Meier curve in the paper. 

Thank you for this comment. The data had been included in the results (Median OS was 5.6 months (95% CI 4.8-6.5) in patients who did not receive additional therapy after progression and 22.2 months (95% CI 18.3-26.2) in patients who received subsequent therapy (p<0.001).) however we included the figure to further highlight the difference.

3. Since it is still in doubt corticosteroids use may compromise the efficacy of immunotherapy, it would be interesting to include these data in your analysis. 

Unfortunately this information was not readily available for our dataset. The impact of steroids and potentially PPIs would be interesting to explore. The following was added to the limitations:

“There was a lack of information on use of other medications like steroids that could in-fluence immunotherapy outcomes.”