Pembro no better than chemo in ipi refractory melanoma??? Disconcerting.... but I think I know why!

So this sounds weird....

Final analysis of a randomised trial comparing pembrolizumab versus investigator-choice chemotherapy for ipilimumab-refractory advanced melanoma. Hamid, Puzanov, Dummer,...Ribas, et al.  Eur J Cancer. 2017 Sep 26. 

PURPOSE:  To evaluate the protocol-specified final analysis of overall survival (OS) in the KEYNOTE-002 study of pembrolizumab versus chemotherapy in patients with ipilimumab-refractory, advanced melanoma.

In this randomised, phase II study, eligible patients had advanced melanoma with documented progression after two or more ipilimumab doses, previous BRAF or MEK inhibitor or both, if BRAF^V600 mutant-positive. Patients were randomised to pembrolizumab 2 mg/kg or 10 mg/kg every 3 weeks or investigator-choice chemotherapy. Crossover to pembrolizumab was allowed following progression on chemotherapy. The protocol-specified final OS was performed in the intent-to-treat population. 

A total of 180 patients were randomised to pembrolizumab 2 mg/kg, 181 to pembrolizumab 10 mg/kg and 179 to chemotherapy. At a median follow-up of 28 months (range 24.1-35.5), 368 patients died and 98 (55%) crossed over to pembrolizumab. Pembrolizumab 2 mg/kg and 10 mg/kg  resulted in a non-statistically significant improvement in OS versus chemotherapy; median OS was 13.4 and 14.7, respectively, versus 11.0 months, with limited improvement after censoring for crossover. Two-year survival rates were 36% and 38%, versus 30%. Progression-free survival, objective response rate and duration of response improved with pembrolizumab versus chemotherapy, regardless of dose. Grade III-V treatment-related adverse events occurred in 24 (13.5%), 30 (16.8%) and 45 (26.3%) patients, respectively.

Improvement in OS with pembrolizumab was not statistically significant at either dose versus chemotherapy.

So, these are beat up melanoma patients who have already failed ipi or BRAF/MEK or both. (Meaning they had already been treated with those drugs and progressed!)  THEN, patients were randomly place in these groups:  180 to Pembro 2mg/kg, 181 to pembro 10mg/kg, and 179 to chemo (exact treatment not noted, old school chemo suggested).   All patients who progressed on chemo were allowed to "cross over" to the pembro arms.  At median follow-up 28 months later - 368 of 540 patients had died and 55% of the chemo patients had crossed over to pembro.  (Meaning less than half of the 179 chemo patients, stayed on chemo...they do not break down deaths in that particular group...while the rest switched over to pembro.)  In the end of this business, researchers found no significant difference in OS  (with 2 year survival rates at 36 and 38% for the pembro groups vs 30% for the chemo group) between the 3 groups...though PFS, RR, and duration of response was better for the pembro groups than for chemo.

Here's what I think is important:

1.  The data here is really muddy due the cross over component in the chemo group combined with the lack of information related to their specific outcomes.  More than half of the chemo ratties ended up on pembro...a drug known to take time to work...time that was wasted on chemo!  Ultimately, 459 of the 540 patients in this study ended up on pembro.
2.  Folks already treated with ipi do less well on anti-PD-1 than those who take anti-PD-1 at the start.  This is not news:  Sequential nivo then ipi = ORR of 41%. Ipi followed by nivo = ORR of 20%!!!! FDA! Are you listening???????
3.  The dose of Pembro 2mg/kg vs Pembro 10mg/kg made no difference in results.  So, as we are gradually learning in immunotherapy generally, the larger dose for an infinite amount of time, does not necessarily gain better responses.

Hang tough, ratties.  You are some amazing peeps!!! - c

Late Note:  10/28/2017 - This article I just reviewed from the Society for Melanoma Research 2016 Congress, published 29 January 2017 - confirms my suspicion that "crossover" accounted for their "results"!!!

Adjusting overall survival when accounting for treatment switch in the KEYNOTE-002 study of pembrolizumab versus chemotherapy in patients with ipilimumab-refractory advanced melanoma.  Drummer, Robert...Pavlick, et al.  

In the randomized phase 2 KEYNOTE-002 (NCT01704287) study, pembrolizumab 2 mg/kg and 10 mg/kg Q3W significantly improved PFS versus investigator-choice chemotherapy at the second interim analysis in ipilimumab-refractory advanced melanoma, with no difference between pembrolizumab doses. Intent-to-treat (ITT) analyses showed a numeric, but not statistically significant, trend toward improved OS with pembrolizumab 2 mg/kg Q3W versus chemotherapy. Final analysis of the KEYNOTE-002 study will be presented at this year's European Society for Medical Oncology congress. Here, we aim to adjust for the bias introduced by crossover from the chemotherapy to pembrolizumab arms. We applied both the 2-stage method for crossover adjustment to OS, and the rank-preserving structural failure time (RPSFT) correction as prespecified in the statistical analysis plan. The resulting adjusted survival curves for chemotherapy were evaluated against external literature-based evidence. With crossover correction for the pembrolizumab 2 mg/kg arm, the HR was 0.79 for the RPSFT and 0.58 for the 2-stage method. When comparing against external literature-based evidence, the adjusted OS for chemotherapy from the 2-stage method was found to fit best with external historical data. Results of the pembrolizumab 10 mg/kg arm will be included in the presentation. In conclusion, the OS benefit of pembrolizumab versus chemotherapy using the best-fitting crossover adjustment was statistically significant.

Just say'n!!! Come on researchers!!!  Don't write crap just so you have something to publish! - c