Improved Survival with Ipilimumab in Patients with Metastatic Melanoma
New England Journal of Medicine, 2010 · paper · Clinical trials
A phase 3 trial of 676 previously treated metastatic melanoma patients comparing ipilimumab, with or without a gp100 vaccine, against gp100 alone. Ipilimumab improved overall survival (median about 10 versus 6.4 months), with sometimes severe immune-related adverse events.
1 finding
Substance The 14 study-drug-related deaths are given as "(2.1%)" in the Abstract and in the Results (p6) but as "A total of 14 deaths (2.2%)" in the Table 3 footnote, which names the 643-patient safety population.
- p. 1 Abstract, Results
There were 14 deaths related to the study drugs (2.1%)
- p. 6 Results, Adverse Events, last sentence
There were 14 deaths related to the study drugs (2.1%), of which 7 were associated with immune-related adverse events.
- p. 18 Table 3 footnote
A total of 14 deaths (2.2%) were determined by the investigators to be related to the study drug
- p. 6 Adverse events, last sentence
There were 14 deaths related to the study drugs (2.1%)
Judge's note: On the p6 image the Results read "There were 14 deaths related to the study drugs (2.1%)" in the Adverse Events section, which reports the safety population. On the p18 upright image the Table 3 footnote reads "Included are all patients who received at least one dose of a study drug (643 patients). A total of 14 deaths (2.2%)". 14/643 rounds to 2.2%; 2.1% only works with the 676 randomized patients, and no denominator is stated for it. The paper does not explain the difference, so one quantity appears with two values.
Second check: another model agreed
Report a problem with this finding
7 minor (typos and the like; most checked by the judge only)
Wording The Discussion says "The long-term effect of ipilimumab in our study ... 1-year and 2-year survival rates of 45.6% and 23.5%" without saying these are the ipilimumab-alone group's rates (ipilimumab plus gp100: 43.6% and 21.6%).
- p. 7 Discussion, paragraph 3
which showed 1-year and 2-year survival rates of 45.6% and 23.5%, respectively
- p. 5 Results, Efficacy
were 43.6%, 45.6%, and 25.3% at 12 months
Judge's note: The p5 Results give 43.6/45.6/25.3% at 12 months and 21.6/23.5/13.7% at 24 months. The p7 Discussion quotes the ipilimumab-alone values without naming the group. The surrounding comparison with phase 2 monotherapy suggests the intended group, so the numbers are right; only the label is missing.
Second check: another model agreed
Report a problem with this finding
Wording Table 2 labels the medians 2.76 (2.73–2.79), 2.86 (2.76–3.02) and 2.76 (2.73–2.83) as "Time to progression — median (95% CI)", while the Figure 1 legend gives the same values as "median progression-free survival".
- p. 16 Table 2, Time to event
Time to progression — median (95% CI)
- p. 12 Figure 1 legend
The median progression-free survival (Panel B) was 2.76 months (95% CI, 2.73 to 2.79)
Judge's note: Both images checked: the values are identical. The Methods (p5) define progression-free survival as progression or death, and that is the prespecified endpoint. 'Time to progression' is a different standard endpoint, so the Table 2 label is inconsistent.
Second check: another model agreed
Report a problem with this finding
Typo The Statistical analysis section says "receipt or nonreceipt of previous interleukin therapy" for the stratification factor, which is called "previous interleukin-2 therapy" everywhere else.
- p. 5 Statistical analysis
previous interleukin therapy
Judge's note: The p5 image shows "previous interleukin therapy". The p3 Study design and the subgroup definitions say "interleukin-2". This is a minor terminology inconsistency; the meaning is clear.
Second check: none (judge only)
Report a problem with this finding
Typo The investigator list prints "S. Grabbe (Molnz)"; the city should be Mainz, which is spelled correctly in the next entry, "C. Loquai (Mainz)".
- p. 9 Appendix, participating investigators, Germany
S. Grabbe (Molnz)
Judge's note: The misspelling is visible on the p9 image, so it is not an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo The investigator list prints "G. Shuler (Erlangen)"; the surname may be a misspelling of Schuler.
- p. 9 Appendix investigators, Germany
G. Shuler (Erlangen)
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the Erlangen investigator is Gerold Schuler. The p9 image does print "G. Shuler (Erlangen)". Nothing in the files confirms the correct spelling.
Second check: none (judge only)
Report a problem with this finding
Typo Table 3, Pyrexia row, gp100-alone grade 3 cell is printed "2(1.5)" with no space, unlike the "n (%)" format used elsewhere.
- p. 17 Table 3, Pyrexia, gp100 alone grade 3
2(1.5)
Judge's note: The missing space is visible on the p17 upright image. Cosmetic only.
Second check: none (judge only)
Report a problem with this finding
Typo Table 3, Increase in aspartate aminotransferase row, ipilimumab-plus-gp100 total cell is printed "4(1.1)" with no space.
- p. 18 Table 3, Increase in aspartate aminotransferase
4(1.1)
Judge's note: Visible on the p18 upright image. Cosmetic only.
Second check: none (judge only)