Pembrolizumab plus Chemotherapy in Metastatic Non-Small-Cell Lung Cancer
2018 · paper · Clinical trials
A double-blind phase 3 trial of 616 patients with untreated metastatic nonsquamous lung cancer, adding pembrolizumab or placebo to pemetrexed and platinum chemotherapy. Pembrolizumab improved overall survival (hazard ratio 0.49) and progression-free survival, with similar rates of severe adverse events.
0 findings
None that passed a second check.
6 minor (typos and the like; most checked by the judge only)
Reference or label The claim that patients with a score of less than 1% are "a population for which single-agent PD-1 and PD-L1 inhibition have a small chance of benefit.12,14-19" cites ref 12 ("Pembrolizumab versus chemotherapy for PD-L1–positive non–small-cell lung cancer") and ref 15 (carboplatin + paclitaxel +/− bevacizumab, with or without atezolizumab, IMPower150), neither of which is a single-agent study in a <1% population by its own title.
- p. 11 Discussion, 2nd paragraph
inhibition have a small chance of benefit.12,14-19
- p. 15 Reference 15
+ paclitaxel +/− bevacizumab, with or without atezolizumab in 1L non-squamous metastatic NSCLC (IMPower150)
- p. 15 Reference 12
Pembrolizumab versus chemotherapy for PD-L1–positive non–small-cell lung cancer
Only the authors can say what is right here.
Judge's note: Read the Discussion on p11 and the reference list on p15. Ref 12 is titled as a PD-L1-positive population, and ref 13 describes KEYNOTE-024 as "PD-L1 TPS >=50%". Ref 15 is by title a chemotherapy-plus-bevacizumab combination with or without atezolizumab. Refs 14 and 16-19 fit the claim by title. The citation block may be meant as loose background, so only the authors can say whether 12 and 15 were intended.
Second check: another model agreed
Report a problem with this finding
Wording Methods describe the ECOG score as "on a 5-point scale, with higher scores indicating increasing disability", while the Table 1 footnote and Figure 1 legend say "performance-status scores range from 0 to 5" (six levels).
- p. 2 Methods, Patients
1 (on a 5-point scale, with higher scores indicat-
- p. 4 Table 1, footnote ‡
performance-status scores range from 0 to 5
- p. 5 Figure 1 legend
range from 0 to 5, with 0 indicating no symptoms
Judge's note: Read the Methods crop (p2) and the Table 1 footnote crop (p4); the Figure 1 legend wording on p5 was checked in the text layer only. A 0-to-5 range has six levels, so the two descriptions of the same scale do not agree. Nothing in the article reconciles them. Minor descriptive inconsistency; either form could be loose shorthand.
Second check: another model agreed
Report a problem with this finding
Wording Discussion says "The median overall survival in our trial may change with longer follow-up because there was substantial censoring of the Kaplan–Meier curves around the time that the medians were reached in this analysis", but Results state "The median overall survival was not reached in the pembrolizumab-combination group".
- p. 11 Discussion, 3rd paragraph
Meier curves around the time that the medians were reached in this analysis
- p. 7 Results, Overall Survival
The median overall survival was not reached in the pembrolizumab-combination group
- p. 11 Discussion, third paragraph
around the time that the medians were reached in this analysis
- p. 6 Figure 1A
Pembrolizumab combination
Judge's note: Read p11, Figure 1A on p6, and the Results text for p7 in the text layer. Only the placebo-combination group reached a median (11.3 months); the pembrolizumab curve in Figure 1A stays above 60%. The plural "medians were reached" is therefore loose for overall survival. It is a wording ambiguity only (it could mean the placebo median); no number is affected.
Second check: another model agreed
Report a problem with this finding
Wording Text says "The only adverse events that were reported in at least 10% of the patients that were more frequent in the pembrolizumab-combination group were diarrhea and rash (Fig. S6A in the Supplementary Appendix)", but Table 2 shows other events at 10% or more that are numerically more frequent with pembrolizumab (e.g. peripheral edema 19.3% vs 12.9%, increased lacrimation 17.0% vs 10.9%, pyrexia 19.5% vs 14.9%).
- p. 11 Results, Adverse Events, 2nd paragraph
The only adverse events that were reported in at least 10% of the patients that were more frequent in the pembrolizumab-combination group were diarrhea and rash
- p. 12 Table 2, rows Peripheral edema / Increased lacrimation / Pyrexia
78 (19.3)
- p. 11 Adverse Events, paragraph 2
reported in at least 10% of the patients that were more frequent in the pembro-
Only the authors can say what is right here.
Judge's note: Read p11 and Table 2 on p12. The table values are as the reviewer quotes, so as worded "only ... more frequent" does not match Table 2. The sentence cites Fig. S6A, which is in the Supplementary Appendix and not among the files; it may apply a criterion (such as a confidence interval for the between-group difference excluding zero) that the article text does not state. Query to authors: state the criterion for "more frequent". The missing-conjunction point in reviewer note 2 is house style and not a defect.
Second check: another model agreed
Report a problem with this finding
Wording Text says "The only adverse event of grade 3 or higher that was more frequent in the pembrolizumab-combination group was febrile neutropenia (Fig. S6B in the Supplementary Appendix)", but Table 2 shows several grade 3-5 events numerically more frequent with pembrolizumab (fatigue 5.7% vs 2.5%, neutropenia 15.8% vs 11.9%, asthenia 6.2% vs 3.5%, diarrhea 5.2% vs 3.0%).
- p. 11 Results, Adverse Events, 2nd paragraph
The only adverse event of grade 3 or higher that was more frequent in the pembrolizumab-combination group was febrile neutropenia
- p. 12 Table 2, Grade 3, 4, or 5 columns (Fatigue, Neutropenia, Asthenia, Diarrhea)
23 (5.7)
Only the authors can say what is right here.
Judge's note: Read p11 and Table 2 on p12. The grade 3-5 columns are as quoted, so the unqualified "only ... more frequent" does not match the table. The sentence cites Fig. S6B, which is in the Supplementary Appendix and not among the files; it may apply an unstated criterion for the between-group difference. Query to authors: state the criterion in the text.
Second check: another model agreed
Report a problem with this finding
Typo Disclosure paragraph prints "Dr. Rodriguez-Abreu," without the accent, whereas the byline prints "D. Rodríguez‑Abreu".
- p. 13 Disclosures, after 'Supported by Merck.'
Dr. Rodriguez-Abreu,
- p. 1 Author byline
D. Rodríguez‑Abreu
Judge's note: Read the byline crop (p1) and the disclosure crop (p13). The accent is present in the byline and absent in the disclosure paragraph on the page image, so this is not an extraction artefact. The same difference is visible between ref 12 (accented) and ref 13 (unaccented) on p15. The Appendix spelling was not checked. Cosmetic.
Second check: none (judge only)