Pembrolizumab Plus Olaparib in Misses Its Mark in Previously-Treated mCRPC

Evan Y. Yu, MD

Pembrolizumab (Keytruda) in combination with olaparib (Lynparza) did not elicit improvements in radiographic progression-free survival (rPFS) or overall survival (OS) when used in patients with molecularly unselected, previously treated metastatic castration-resistant prostate cancer (mCRPC).¹

Findings come from the phase 3 KEYLYNK-010 trial (NCT03834519) which evaluated pembrolizumab plus olaparib vs abiraterone (Zytiga) and enzalutamide (Xtandi) in this patient population. Findings showed that the combination of pembrolizumab plus olaparib resulted in more grade 3 greater treatment-related adverse events vs novel hormonal agents (NHA). Further, there were no new safety signals identified.

Patients aged 18 years and older with mCRPC who had progressed on or after abiraterone or enzalutamide and docetaxel were enrolled. Patients were also required to have an ECOG performance status of 1 or less and were randomized 2:1, including 529 patients enrolled in the combination arm and 264 patients in the NHA arm.

In the experimental arm, pembrolizumab at 200 mg every 3 weeks for 35 cycles or less plus olaparib at 300 mg twice a day was administered to patients while those in the control group received abiraterone at 1000 mg every day if they received prior enzalutamide or enzalutamide at 160 mg each day if they received prior abiraterone.

The co-primary end points of the study included rPFS and OS. Secondary end points were time to first subsequent therapy, objective response rate, and safety.

Findings of the trial showed that the median time to first subsequent therapy was a 7.2-months (95% CI, 6.7-8.1) in the pembrolizumab plus olaparib arm compared with 5.7 months (95% CI, 5.0-7.1) in the NHA arm (HR, 0.86; 95% CI, 0.71-1.03). Patients in the pembrolizumab plus olaparib arm had an ORR of 16.8% (95% CI, 12.3%-22.1%) vs 5.9% (95% CI, 2.4%-11.7%).

The disease control rate in these groups were 32.4% (95% CI, 26.5%-38.6%) and 19.3% (95% CI, 12.7%-27.6%), respectively. Additionally, complete responses were observed in 1.6% of patients given pembrolizumab plus olaparib vs 0 given an NHA. A total of 15.2% vs 5.9% of patients experienced a partial response.

“I do think that this combination may have more efficacy in those patients who have soft tissue measurable disease than those who have bone metastases. Adverse events were as expected with these agents. The most common treatment-related adverse events were anemia, gastrointestinal adverse events, like decreased appetite, diarrhea, and some skin rash. Those are the typical things., said Evan Y. Yu, MD, in an interview with Targeted Oncology™.

In the interview, Yu, a professor, Division of Oncology, Department of Medicine, University of Washington School of Medicine, further discusses these data from the KEYLYNK-010 trial.

Targeted Oncology: Can you talk about some of the updates seen in mCRPC over the last 2 years. What unmet needs still exist?

Yu: There's a lot of new things that have happened in metastatic castration-resistant prostate cancer recently. I would say the thing that's most impactful is the FDA approval of 177Lu-PSMA-617 [Pluvicto]. That's a radioligand therapy that targets PSMA and was studied in patients who have metastatic castration-resistant prostate cancer who have previously received at least 1 novel hormonal therapy agent and received prior taxane chemotherapy like docetaxel. There's a clear overall survival benefit to that and that is now commercially available.

There have been some production facility issues, but it seems to be ramping up now. I think that's going to make a major difference in the lives of our patients with metastatic castration-resistant prostate cancer. I would say the other findings that I think may not yet be primetime but are being considered by the field are the MAGNITUDE [NCT03748641] and the PROpel study [NCT03732820] and these are looking at use of PARP inhibitors for first-line metastatic castration-resistant prostate cancer. I think for those with DNA repair deficiency, that the addition of niraparib [Zejula] in addition to abiraterone [Zytiga] has led to significant radiographic progression-free survival benefit. Not yet overall survival as that data is not mature, but at least for those that are biomarker positive, certainly generally driven by BRC2 mostly, that those patients seem to have radiographic progression-free survival benefit.

The PROpel study looked at using olaparib combined with abiraterone and all comers. After the fact, they did next generation sequencing and then sub-classified people into DNA repair deficiency vs DNA repair proficient patients and showed that in all comers and in both subgroups that there was radiographic progression-free survival benefit as well. I think the challenges for these studies are that 1, there's not yet an overall survival benefit because these data are not yet mature. It's not clear that these studies were statistically powered to be able to show overall survival benefit. But the mechanism is unclear for those patients who do not have DNA repair deficiency. I think that those are things that need to be clarified in the field before it becomes prime time.

Can you talk about how the KEYLYNK-010 study came about?

The KEYLYNK study combines pembrolizumab and olaparib. There are many reasons why this came about. I think that there are all kinds of rationale for combination therapy with PARP inhibitors in prostate cancer, but the question would be, would there be a role for combining a PARP inhibitor with immunotherapy? There have been some studies that have shown that when you give a PARP inhibitor, mostly in breast and ovarian cancer that upregulates PD-L1 expression in the tumor, that might make the patient more amenable to response to something like a PD-1 inhibitor like pembrolizumab. Another rationale is there's some preclinical data that shows that when you give a PARP inhibitor, even to patients that are DNA repair proficient, that you can accumulate single strand DNA, cytosolic DNA, that then triggers a pathway called the CGAS sting pathway.

That increases interferons and serves as an immune attractant to attract more immune cells into the tumor microenvironment. That might be more conducive to a PD-1 inhibitor working in prostate cancer, which is generally considered to be a rather immune-exclusive microenvironment that doesn't have a lot of immune cells in that area. These serve as a preclinical rationale. Then we did a phase 2 study, the KEYNOTE-365 [NCT02861573] study, that if you looked at the waterfall plots, 20 plus percent of patients with soft tissue had significant disease shrinkage responses. When it came to confirming those responses, those numbers were smaller because many of those patients had bone metastases, and by the RECIST 1.1 confirmation, you couldn't prove those responses by soft tissue responses with bone mets. The truth of the matter is we saw significant activity and are in a phase 2 study. This then launched into the randomized phase 3 KEYLYNK-010 trial.

What are some key takeaways from these results?

The results that we presented were unfortunately negative for the primary end points. This was a randomized, phase 3, open-label trial that took patients with metastatic castration-resistant prostate cancer and progressive disease after either abiraterone or enzalutamide, but not both, and docetaxel. They were randomized in a 2:1 fashion to pembrolizumab plus olaparib vs either abiraterone or enzalutamide. If you received prior enzalutamide, you'd receive abiraterone and if you received prior abiraterone, you'd receive enzalutamide. We ended up randomizing approximately 800 patients.

We saw about half the patients that receive abiraterone and about half the patients had received enzalutamide. About a quarter of the patients were DNA repair gene altered and this is what we would expect from other studies in the field. BRCA mutated was a little bit less than 10%. The primary end point of radiographic progression-free survival, that hazard ratio was 1.02. There was no benefit to pembrolizumab plus olaparib, but we did see in the subgroup of BRCA mutated patients significant benefit. For the primary end point of overall survival, that hazard ratio of 0.94 was also not statistically significant, with a median of 15.8 months for pembrolizumab plus olaparib vs 14.6 months for the switch to the novel hormonal therapy agent. The 2 subgroups that stand out are the BRCA mutated patients where the hazard ratio was 0.52 and the confidence interval didn't cross 1.

The other population was a small population. It was very interesting as the North American population seemed to do better, although the hazard ratio did slightly cross 1 there. What we did see is a significant benefit in objective response rate that was 16.8% for pembrolizumab plus olaparib, 5.9% for those with novel hormonal therapy, and the confidence intervals did not cross.

What do these results mean for the future treatment landscape?

I do think that this combination may have more efficacy in those patients who have soft tissue measurable disease than those who have bone metastases. Adverse events were as expected with these agents. The most common treatment-related adverse events were anemia, gastrointestinal adverse events, like decreased appetite, diarrhea, and some skin rash. Those are the typical things.

In summary, this randomized, controlled trial did not show statistically significant improvement in radiographic progression-free survival or overall survival of pembrolizumab plus olaparib vs the active comparator novel hormonal therapy agent for an unselected patient population molecularly unselected. For that reason, the trial was stopped early with guidance from the external data monitoring committee. Objective response rate, however, was higher for pembrolizumab plus olaparib and the expected BRCA altered groups did do better with pembrolizumab plus olaparib.

_REFERENCE:

_{Yu EY, Park SH, Goh JCH, et al. Pembrolizumab + olaparib vs abiraterone (abi) or enzalutamide (enza) for patients (pts) with previously treated metastatic castration-resistant prostate cancer (mCRPC): randomized open-label phase III KEYLYNK-010 study. Ann Oncol. 2022;33(suppl 7):1362MO. doi:10.1016/annonc/annonc1070}