NEW YORK – Treatment with Bristol Myers Squibb's PD-1- and CTLA4-blocking immune checkpoint inhibitors led to "robust" responses in metastatic castration-resistant prostate cancer patients with DNA mismatch repair deficiencies, researchers at the European Society for Medical Oncology Congress said.
Niven Mehra, a medical oncologist at the Radboud University in the Netherlands and one of the study's main investigators, presented the results from the Phase II INSPIRE study at the meeting in Barcelona, Spain, on Monday. The results were published in tandem in the Annals of Oncology.
In the past, immune checkpoint inhibitors have seen limited success treating mCRPC, especially in biomarker unselected patient populations. "Immunotherapy in prostate cancer in an all-comers strategy is clearly a failed approach," said Christopher Sweeney, a medical oncologist at the Royal Adelaide Hospital in Adelaide, Australia, during a discussion of the data. He estimated that sponsors spent more than $700 million to study multiple drugs in more than 7,305 patients. And yet, these all-comer studies have all turned up negative, Sweeney bemoaned.
Prior data suggested that using BMS's immune checkpoint inhibitors together — specifically, 1 milligram per kilogram of Opdivo (nivolumab) and 3 milligrams per kilogram of Yervoy (ipilimumab) — could have higher efficacy for mCRPC relative to single-agent checkpoint inhibition. But, historically, the toxicities were too high for the strategy to gain a strong foothold in the advanced prostate cancer treatment setting.
Tweaked dose strategy, biomarker selection
Investigators designed the INSPIRE study to determine the efficacy and safety of the same two-drug immunotherapy approach, but in a different dose: 3 milligrams per kilogram of Opdivo and 1 milligram per kilogram of Yervoy for four doses followed by a flat dose of 480 milligrams of Opdivo for up to one year.
Researchers focused on outcomes with this strategy across four molecularly selected subtypes: mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-high) cancers; tumor mutational burden-high (TMB-high) cancers; BRCA1/2-mutated tumors; and cancers with biallelic inactivation of CDK12. Collectively, Mehra said, these biomarker subtypes make up around 15 percent of mCRPC cases.
Of note, to determine patient eligibility for the TMB-high subset, the researchers set a TMB threshold of at least 7.1 mutations per mega base.
Among 65 mCRPC patients in the trial's efficacy evaluable population, none of whom had received prior immune checkpoint inhibitors, the disease control rate was 38 percent after six months on Opdivo-Yervoy.
Breaking down the patient population by biomarker subgroup, Mehra said that 21 patients had dMMR tumors, eight patients had high TMB, 16 patients had CDK12 inactivation, and 20 patients had BRCA1/2 mutations.
"There were clear differences in the disease control rate between the molecular subtypes," Mehra said. The disease control rates were 81 percent for dMMR cases; 25 percent for TMB-high patients; 19 percent for patients with CDK12-inactivated tumors; and 15 percent among those with BRCA1/2-mutated tumors.
The overall response rate was 38 percent in the full efficacy-evaluable group and 75 percent for dMMR cases; 22 percent for those with CDK12-inactivated tumors; and 23 percent in those with BRCA1/2-mutated tumors. None of the patients with high TMB responded.
The median progression-free survival was highest at 32.7 months in dMMR cases. Among those with high TMB, the median progression-free survival was 3.8 months, 3.5 months for those with BRCA1/2 mutations, and 1.6 months for patients with CDK12-inactivated tumors. Median overall survival was not yet reached among dMMR patients, 10.3 months in both the CDK12-inactived and BRCA1/2-mutated subsets, and 7.1 months for those with high TMB.
The prostate-specific antigen responses showed a similar trend as the other outcomes. The PSA response was highest in the dMMR patient population with 86 percent of these patients experiencing deep PSA responses. "Again, we saw the mismatch repair group doing spectacularly," Mehra said. "These findings underscore the need for early testing and treatment of patients with mismatch repair deficient prostate cancer with dual immune checkpoint inhibitors."
Toxicities still limiting
Despite the promising efficacy results with Opdivo-Yervoy in the dMMR mCRPC patient population, Mehra emphasized that future research is still needed. This is especially true given as only about 5 percent of metastatic prostate cancer patients harbor mismatch repair deficiencies.
Additionally, Mehra pointed out that the treatment regimen is not without its serious side effects.
"This dual immune checkpoint blockade is toxic," he said. Out of the trial's safety-evaluable population, 48 percent experienced a grade 3 or higher treatment-related toxicity, and two patients died due to the treatment's side effects. Based on this risk, Mehra and colleagues suggested in their Annals of Oncology paper that dual immune checkpoint inhibitor strategies be reserved for fit patients likely to benefit from the drugs, and that in future studies, researchers try to identify potential biomarkers for early recognition of immune checkpoint inhibitor-related toxicity.
"Studies should also investigate if we can limit the use of dual immune checkpoint inhibitors to one or two cycles without compromising treatment efficacy," the authors wrote. Given the rapid early PSA decline noted in the dMMR patient subgroup, they suggested it might be possible that one or two doses of Opdivo and Yervoy could sufficiently achieve deep responses.
"This treatment is not without risk," Sweeney agreed.
Commercial landscape, future questions
Merck's anti-PD-1 agent Keytruda (pembrolizumab) is a US Food and Drug Administration-approved option for patients with any type of refractory cancers that are TMB-high, dMMR, or MSI-high. Patients with refractory mCRPC with these molecular features are also eligible for Keytruda under this tissue-agnostic approval.
In Europe, however, the checkpoint inhibitor's label is narrower. The European Commission in 2022 approved Keytruda for five dMMR or MSI-high tumor types: colorectal, endometrial, gastric, biliary tract, or small intestine cancer.
Although Opdivo-Yervoy isn't approved for mCRPC, in their Annals of Oncology publication, Mehra and coauthors shared their view that the INSPIRE trial data, combined with previous studies, might be enough to justify the use of this combo and other immune checkpoint inhibitors in dMMR mCRPC.
The data are "enough to justify consideration by the health authorities," they wrote, especially given the unmet need this patient population faces in Europe.
"Conditional approval by the European Medicines Agency could be an appropriate initial step, allowing patients access to these promising therapies while awaiting the results of a larger, collaborative international study," they wrote. In such a study, researchers could dig deeper into the relative benefit of Opdivo-Yervoy combined versus a single-agent anti-PD-1 agent like Keytruda and compare both approaches to standard-of-care, first-line treatment.
This comparison between single-agent and dual-checkpoint blockade, for now, remains an open question. In their paper, Mehra and coauthors pointed out that in a prior retrospective analysis, the overall response rate was 48 percent among 71 dMMR mCRPC patients treated with single-agent anti-PD-1 checkpoint inhibitors, and the median progression-free survival was 8.3 months.
In the INSPIRE trial, the 75 percent overall response rate and 32.7-month median progression-free survival observed with the Opdivo-Yervoy combo appeared higher than what has been seen with a single-agent approach. "Although direct comparisons are lacking, these data suggest that dual immune checkpoint inhibitors may be more effective compared to immune checkpoint inhibitor monotherapy," wrote the authors.
Even though the data from INSPIRE were encouraging in this particular dMMR group, the researchers underscored the need for better predictive immunotherapy biomarkers in mCRPC. To this end, Mehra and colleagues are planning to conduct translational research within this trial and other studies of dual immune checkpoint inhibitors, aiming to identify genomic correlates associated with response.
"We need to invest in biomarkers for personalized medicine," Sweeney concluded in his discussion. "Yes, it is hard … I'm not going to disagree with that. But it is harder to have futile, toxic therapy."