Anti-PD-1 results in melanoma patients: outcomes plus responses to retreatment

So, for those of us lucky enough to respond to anti-PD-1, given the BE-ATCH that is melanoma, it does cross your mind, "Now what???!!!"

For a little background, here are the 2017 ASCO articles, that I included in:  ASCO 2017: Outcomes after stopping immunotherapy in melanoma

Here are the related articles from 2018:  ASCO 2018 - Outcomes for melanoma peeps after stopping anti-PD-1 

Along with this from 2017:  After stopping anti-PD1 treatment...how long can melanoma patients maintain a complete response????

Finally, if once you were an anti-PD-1 responder, but find yourself one no longer, there was this from 2018:   Anti-PD-1 after progression

Now, there's this:

Responders to anti-PD1 therapy: Long-term outcomes and responses to retreatment in melanoma (mel).  2019 ASCO.  Warner, Palmer, Shoushtari, ...Wolchok, Postow, et al.  J Clin Oncol 37, 2019.

Background: Little is known about patients (pts) who discontinue anti-PD1 therapy after a complete response (CR) outside of clinical trials. There are also limited data about retreatment with a second course of anti-PD1 upon disease progression. Methods: We retrospectively studied pts (n = 398) at MSKCC with unresectable mel (non-uveal) who received greater than/= to 1 dose of single-agent anti-PD1 and were followed greater than/= to 3 months (mos) after treatment cessation. CR was defined radiographically or by a negative biopsy of residual tissue. Overall survival (OS) and time to treatment failure (TTF, time until next therapy or death) were calculated from time of CR. When to stop therapy and whether to retreat after progressive disease (PD) were at the discretion of the treating oncologist. A subset of pts received a second course of single-agent anti-PD1 greater than/= to 3 months after initial discontinuation; retreated pts were evaluable if they had radiographic or clinical evaluation to assess retreatment efficacy. Results: 102 pts (25.6%) achieved CR (n = 89 radiographic, n = 13 pathologic). Median follow-up was 22.6 mos for survivors who had a CR. Estimated 3-year OS from time of CR was 82.5%. For pts who had a CR, therapy was discontinued due to CR (n = 72), toxicity (n = 24), or other reasons (n = 6). The median duration of treatment for CR pts was 9.4 mos (range 1.6-36.1). 20 CR pts later had progressive disease (PD). Median TTF has not been reached; at 3-years the estimated treatment-free survival for CR pts was 72.3%. 34 pts received a second course of anti-PD1 for PD after a median of 11.6 mos off treatment (range 3.5-28.6). Best responses to the second course of treatment were: 2 CRs (5.9%), 3 with tumor shrinkage (8.8%), 9 (26.5%) with SD, and 20 with PD (58.8%). Of these pts who had had a CR (n = 8) or some lesser degree of tumor shrinkage (n = 13) to the initial course of anti-PD1 treatment, only 1 and 2 pts responded, respectively, to retreatment. Median duration of retreatment was 10.9 wks. Conclusions: In this largest dataset to knowledge of mel pts treated with a second course of anti-PD1, response rate was low, even in pts who had achieved a response initially. Further study is needed into the necessary duration of initial anti-PD1 treatment and optimal strategies for initial responders who discontinue and later develop PD.

Here, when looking back over the records of 398 melanoma patients treated with at least one dose of anti-PD-1 (I think they should have chosen those who had had at least a year of treatment, but that's just me!) 102 of those peeps had a complete response.  Median f/u of them was 22.6 months.  3 year OS from CR was 82.5%.  20 of these complete responders later had progressive disease.  Additionally, they looked at 34 patients who were given a second course of anti-PD-1 for progression after 11.6 months off treatment. (At least in the abstract, they do not break down these 34 as to whether they were complete responders, partial responders, folks with stable disease or what.)  Of these, 2 gained a CR, 3 had tumor shrinkage, 9 had stable disease and 20 continued with progressive disease.  Of those who responded to this second round of anti-PD-1, 1 had had a CR to their first exposure to anti-PD-1 and 2 had had "a lesser degree of tumor shrinkage" to their first round of anti-PD-1.  The obvious conclusion was "In this largest dataset to knowledge of mel pts treated with a second course of anti-PD1, response rate was low, even in pts who had achieved a response initially. "

So - good news vs bad news sort of thing.  Pretty good data on CR and durability of those responses, with OS of 82.5% of CR at 3 years.  Pretty tough to see the limited response rates for retreatment with anti-PD-1.  HOWEVER!  I think there is a major flaw in this retrospective.  We KNOW that immunotherapy takes a minute to work.  We KNOW that it usually takes more than one dose to make a difference.  SO....including folks who had been given only one dose of anti-PD-1 in this data set would obviously skew the results to having fewer responders of all stripes.  But, again - that's just me.

Hang tough, ratties, for there is still beauty...

...after the storm. - c