MK-3475 (Merck's Anti-PD1) for melanoma...ASCO abstracts and my thoughts

Evaluation of immune-related response criteria in patients with advanced melanoma treated with the anti-PD1 monoclonal antibody MK-3475
Abstract 3006, J Clin Oncol
Hodi, Ribas, Hamid, Robert, Weber, Wolchok, et al

3 melanoma cohorts treated with MK-3475 2mg/kg every 3 weeks, 10mg/kg every 3 OR 2 weeks in a phase 1 trial.  Tumor imaging was performed every 12 weeks.  411 patients enrolled.  192 were on MK-3475 for 28 weeks or more as of 10/18/2013.  Tumor flare was seen in 7 patients. In these patients, best overall response was:  complete response = 1, partial response = 4, stable disease = 2.  Atypical delayed response was seen in 6 patients.  The 51 patients with progressive disease, but also WITH evidence of response using immune criteria had favorable overall survival compared with 145 patients who had progressive disease and NO evidence of response via immune criteria.  Conclusion: Conventional criteria such as RECIST may underestimate the benefit of MK-3475 in approximately 10% of treated patients.


Clinical efficacy and correlation with PD-P1 expression in patients with melanoma treated with anti-PD1 monoclonal antibody MK-3475

Abstract 3005, J Clin Oncol
Kefford, Ribas, Hamid, Robert, Daud, Wolchok, Joshua, Hodi, Weber, Patnaik, et al

135 patients received MK-3475 10mg/kg every 2 weeks or every 3 weeks or 2 mg/kg every 3 weeks.  Response was checked every 12 weeks.  Biopsy was required in the 60 days before starting med and PD-L1 expression was assessed.  As of 10/18/2013, all patients had 13 months or more follow-up.  In patients with measurable disease, overall response rate was 41%.  Objective responses were observed as late as 64 weeks, with some conversions to complete response seen as late as 72 weeks,  Median response duration was not reached; responses were ongoing for 87% of responders.  Tumor PD-L1 expression was evaluable in 71 patients with measurable disease.  Of these patients, PD-L1 expression was associated with improved overall response rate (51% vs 6%) and progression free survival (median 12 vs 3 months).  14% of patients experienced drug-related 3/4 adverse events, but there were no treatment-related deaths.  "Conclusion:  MK-3475 induces durable responses...and...although tumor PD-L1 positivity was associated with improved overall response rate and progression free survival, antitumor activity was also observed in patients with low PD-L1 expression."

Lesion-specific patterns of response and progression with anti-PD1 treatment in metastatic melanoma

Abstract 9077
Lyle, Lee, Menzies, Chan, et al (Researchers out of Australia)

Bi-dimensional measurements were taken of every met via CT at baseline and every 12 weeks thereafter.  27 patients evaluated with f/u of up to 54 weeks, had total of 442 discrete mets at baseline.  13 of 27 had had objective response, 11 of these by the first scan.  Median time to best response in patients with an objective response was 24 weeks.  Only 1 of 27 patients had a complete response.  However, 228 of 442 tumors had a complete response and 81% of patients had 1 or more met with a complete response at first scan.  Mets with complete response were smaller than those without it. Complete response rate was highest in lung vs other sites combined. Only 1 complete response met subsequently progressed.  Of 80 new or growing mets at first scan, only 4 (5%) subsequently had an objective response.  "Conclusion: Response to anti-PD1 therapy is rare in mets that are new or growing at first scan."

Baseline tumor size as an independent prognostic factor for overall survival in patients with metastatic melanoma treated with the anti-PD1 monoclonal antibody MK-3475

Abstract 3015, J Clin Oncol
Joseph, Schaap, Wolchok, Joshua, Ribas, Hodi, Hamid, Daud, Weber, et al.

In phase 1 clinical trial, 411 patients given MK-3475 2mg/kg every 3 weeks, 10mg/kg every 3 weeks, or 10mg/kg every 2 weeks.  Response was checked every 12 weeks.  Of the 411, 365 had measurable tumors at baseline and median f/u of 10 months as of the October 2013 cutoff date. "Median overall survival was not reached, and 1 year overall survival was 71% in all 411 patients and 69% in the 365 patients included in this analysis.  While tumor size greater than 90mm was associated with a worse prognosis, these patients did have a median overall survival of 14 months in the combined data set, suggesting they derive benefit from MK-3475."  "Conclusions: Baseline tumor size is the strongest independent prognostic factor in patients with metastatic melanoma treated the MK-3475." 

Randomized comparison of two doses of the anti-PD1 monoclonal antibody MK-3475 for ipi-refractory and ipi-naive melanoma.

Abstract 3000, J Clin Oncol
Hamid, Robert, RIbas, Wolchok, Hodi, Weber, et al

2 cohorts - ipi naive and ipi refractory.  Patients were randomized to 2 or 10mg/kg every 3 weeks.   Arms were well balanced for known prognostic factors.  No significant differences were noted in overall response rate between the doses.  "Conclusions:  MK-3475 2 mg/kg every 3 wks and 10mg/kg every 3 wks provided similar efficacy and safety in ipi naive and refractory patients."

My thoughts:
1.   Patients with an immune response do better, even if scans don't show it.  To fail to look at the immune response in addition to the scan results may fail to provide an accurate picture of patient progression or success.
2.  PD-L1 expression of the tumor, facilitated a much better response to anti-PD1.  However, some folks with low expression responded as well.  Does that mean PD-L1 expression is pretty essential for a good response to anti-PD1 (51% vs 6% ORR) or do we just not have a good test?
3.  Two abstracts show us that smaller tumors respond better.  So why are we not treating more NED patients???????
4.  Patients seem to respond to anti-PD1 at the start since new or growing tumors at first scan, rarely responded....but sometimes they did.
5.  For MK-3475....2mg vs 10 mg/kg....neither did better than the other.  So...for all of you out there getting less than the 10mg/kg...no worries.  A lesser dose works just as well.
6.  And, as I've pointed out before....give it time!!!  Anti-PD1 can have positive effects...LATER!
------------------------------------------------------------------------------------
Results of my sister arm....in my Nivo trial at Moffitt -

Updated survival, toxicity, and biomarker of nivolumab with/without peptide vaccine in patients naive to, or progressed on, ipilimumab
Abstract 3009, J Clin Oncol
Weber, Kudchadkar, Gibney, et al.

105 patients with unresectable melanoma, who had failed at least one regimen were enrolled and given nivo at 1, 3, or 10mg/kg.  34 patients were ipi naive.  10 patients had failed ipi with grade 2 or less adverse events.  20 patients had filed ipi with grade 3/4 adverse events.  All of these patients were HLA A0201 positive and were given peptide vaccines along with nivo.  41 patients had grade 2 or less ipi reactions and were not HLA restricted and were given nivo alone.  Median f/u was 15 months.  Median progression free survival was 4.2 months.  Median overall survival was 16.7 months, with a one year survival of 65%.  Median overall survival was similar for ipi naive or ipi refractory.  Of 20 patients with grade 3/4 ipi reactions, only one had a subsequent grade 3/4 nivo reaction and it was different from the one experienced on ipi.  Conclusion:  Median overall survival of 16.7 months with nivo was observed in previously treated melanoma patients naive to or failing ipi.  Prior adverse effects to ipi were not replicated with nivo.

My take:  So...you can fail ipi and/or have a bad reaction to it and you may still respond to nivo and are NOT doomed to experience a repeat of your adverse reaction.  Median overall survival for MK-3475 and Nivo are looking pretty similar. 

Much more to come out of ASCO in the coming weeks.  Stay tuned.  And...how bout a hand for all those ratties?!!!  Wishing you all my best - c