Symptomatic and Health Related Quality of Life Benefits With Olaparib in Men With mCRPC Carrying Mutations in the Homologous Recombination Repair Genes, Journal Club - Christopher Wallis & Zachary Klaassen

May 24, 2022

Christopher Wallis and Zachary Klaassen focus this discussion on the article published in The Lancet titled "Pain and health-related quality of life with olaparib versus physician's choice of next-generation hormonal drug in patients with metastatic castration-resistant prostate cancer with homologous recombination repair gene alterations (PROfound)." The authors of this study aimed to assess pain and patient-centric health-related quality of life (HRQOL) measures in patients within the PROfound trial.


Christopher J.D. Wallis, MD, Ph.D., Assistant Professor in the Division of Urology at the University of Toronto.

Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Georgia Cancer Center

Read the Full Video Transcript

Christopher Wallis: Hello, and thank you for joining us for this UroToday Journal Club discussion. We're discussing a recent publication coming out of the PROfound trial, examining Pain and health-related quality of life with olaparib versus physician's choice of next-generation hormonal drug and patients with metastatic castration-resistant prostate cancer with homologous recombination repair gene alterations: An open-label, randomized, phase III trial. I'm Chris Wallis, an assistant professor in the Division of Urology at the University of Toronto, and with me today is Zach Klaassen, an assistant professor in the Division of Urology at the Medical College of Georgia. This paper was recently published in Lancet Oncology, led by the senior author, Dr. Fred Saad.

By the way of background, homologous recombination repair or DNA repair defects are seen between 1 in 6 and 1 in 4 patients with metastatic castration-resistant prostate cancer when we consider both germline and somatic variants. There are many seminal papers describing this, but three of them have been highlighted here. These show that these alterations are associated with a poor prognosis and resistance to standard therapies, including targeting of the androgen axis. However, these alterations do open up novel therapeutic targets and the use of PARP inhibitors can target these pathways.

PROfound is a phase III randomized trial among men with mCRPC who had previously received androgen signaling agents and who had relevant homologous recombination repair mutations, with a focus on cohort A comprising those with BRCA1, BRCA2, or ATM alterations. Patients were randomized to receive olaparib or a switch from their initial androgen signaling agent to an alternative agent. The primary outcome is radiographic progression-free survival, and the secondary was overall survival, focusing again on cohort A of those with BRCA1, BRCA2 and ATM mutations. As we see here in the initial publication led by Dr. de Bono, the use of olaparib was associated with significant improvements in imaging-based progression-free survival and cohort A, and follow up work from led by Dr. Hussain showed that this translated into improvements in overall survival as well.

On the basis of these data, the FDA approved olaparib for patients with homologous recombination repair gene-mutated mCRPC, however, we need to take a greater scope and think about what our treatment goals are in this disease space. And while improving survival is a clearly important outcome, we also need to consider efforts to provide the best health-related quality of life we can for our patients for as long as possible.

And so the present analysis addresses the second point and seeks to assess the effective olaparib compared to androgen access targeting agent switch on pain, health-related quality of life, symptomatic skeletal-related events, and time to first opiate use. We've already alluded somewhat to the design of the PROfound trial, but to summarize, this enrolled adult men with metastatic castration-resistant prostate cancer, who'd had progression on previous abiraterone enzalutamide, according to RECIST version 1.1 for soft tissue, Prostate Cancer Working Group 3 for bony lesions who had ongoing ADT, previous taxane-based chemotherapy was allowed, but not mandated, and patients had to have adequate organ and marrow function. Patients in the screening process underwent an investigational assay based on the FoundationOne CDX platform to identify alterations in one of 15 pre-specified genes involved in homologous recombination repair. And you see the list there.

This study was designed, as we alluded to previously, with the biomarker-driven stratification to cohort A, focused on BRCA1, BRCA2, and ATM, and cohort B, focused on the other 12 homologous recombination repair defects. Within each cohort, there was 2:1 randomization to olaparib or an androgen access targeting agent switch, and randomization was further stratified by use of prior taxane chemotherapy and measurable disease. Treatment was continued until disease progressed or there was unacceptable toxicity, and patients who are initially randomized to an androgen access agent switch could cross over to receive olaparib at the time of progression. This, again, highlights this stratified design of the trial with randomization within each of the cohorts independently.

The primary outcome, as we alluded to before, is imaging-based progression-free survival and cohort A, and then there are also a number of secondary outcomes. These oncologic secondary outcomes have already been reported and the presence manuscript we're discussing today focuses on remaining secondary outcomes, including time to pain progression, pain interference score, health related quality of life is measured by FACT-P, symptomatic skeletal-related events, and the time to first opiate use. And these outcomes are all primarily assessed in cohort A among patients with BRCA1, BRCA2, and ATM alterations, but also assessed in the overall study cohort.

In terms of outcome assessment, patients completed the BPI-SF at baseline, and then every 4 weeks until 24 weeks after they had evidence of disease progression. It stopped at this point because it was felt that subsequent measurements would not reflect the active treatment within the randomization strata and would reflect subsequent post-protocol therapies. ERT medication module was used to quantify analgesic use, the FACT-P questionnaire was completed at baseline, and then every 8 weeks, again until 24 weeks after progression, and symptomatic skeletal-related events were clinically assessed every 4 weeks until week at 24, and then every 8 weeks thereafter. Notably, these were defined based on clinical symptoms and the requirement for palliative treatment, and did mandate imaging-based confirmation of these skeletal-related events.

In terms of statistical analysis, as you would expect, the study was designed empowered on the basis of the primary outcome of radiographic progression-free survival, and all of these secondary outcomes we're discussing are assessed on the basis of a hierarchical multiple testing procedure, which was not corrected for the multiplicity of secondary outcomes examined. In terms of the safety population and toxicity assessments, this included all patients who received at least one dose of the trial drug and the analyses we're looking at today, looking at patient reported outcomes and health related quality of life, are performed in a modified intention-to-treat population that excludes data collected after patients started new anti-cancer therapy. So the patients were censored at the initiation of subsequent therapy, but the feeling that their patient reported outcomes may be more at trivial to these post therapies at that time.

The authors use mixed models for repeated measures to assess their patient reported outcomes, and in their multivariate models, included treatment effects, visit interaction between these two, the baseline function, prior taxane use, and measurable disease as fixed effects. They use the Kaplan-Meier technique for time to event outcomes considered 6 and 12 month landmark outcome statistics. Finally, they used the logistic regression to assess the proportion of patients with changes in patient reported outcomes at landmark intervals.

At this point in time, I'm going to hand over to Zach to walk us through this important secondary analysis of the PROfound trial.

Zachary Klaassen: Thanks, Chris, for that great introduction and review the methods for this PROfound analysis. As you can see here in the trial profile, there was 4,425 patients that were screened, and ultimately, there was 245 patients in cohort A that were included in the efficacy analysis, 162 of which were assigned to the olaparib group, and 83 of which were assigned to the control group.

This is a table one baseline characteristics. We'll focus on the middle of this table, which is the patients with BRCA1, BRCA2, ATM mutations, otherwise known as cohort A. You can see that in terms of olaparib versus control, these were well balanced groups, with a median age of late sixties with most commonly two-thirds of patients being white, 23% of patients being metastatic at initial diagnosis, and more than two-thirds of patients having a Gleason score greater than 8. In terms of the alterations, most commonly was BRCA2, 49% of patients with olaparib and 57% in the control group. And interestingly, the median baseline PSA was higher in the control group, at 112.9, compared to 62.2 in the olaparib group.

In terms of measurable disease at baseline, just over 50% in both arms. In terms of metastasis of baseline, bone only, 35% with the olaparib group and 28% in the control group. Despite being a pretreated cohort, the majority of patients were ECOG 0 and 1, with slightly more ECOG 1 in the control group. With regards to previous new hormonal drugs. The most common was enzalutamide, at 42% in the olaparib group and 48% in the control group. And with regards to previous taxane use, majority of these patients had previous docetaxel, at 46% in the olaparib group and 39% in the control group.

This is the Kaplan-Meier analysis for time to pain progression in patients without opioid use at baseline. For the next several figures, olaparib will be red and the control group will be blue. The median time to pain progression was not reached in both arms, with a hazard ratio not significant, but favoring olaparib at 0.45 and 95% confidence interval of 0.20 to 1.06. Similar figure looking at time to progress of pain severity in patients without opioid use at baseline. Again, the median not reached in both groups, with a hazard ratio of 0.56 and a 95% confidence interval of 0.25 to 1.34. This looks at time to first opioid use for cancer-related pain and those without opioid use at baseline. The median was 18 months in the olaparib group and was 7.5 months in the control group, with a statistically significant hazard ratio favoring olaparib, at 0.61 and a 95% confidence interval of 0.38 to 0.99.

This is an important figure. This shows the change in overall pain interference. You can see on the x-axis is weeks, so this is time in baseline, and on the y-axis is the least square means change from baseline in pain interference score. And you can see at the top here, these are all significant p-values, favoring the olaparib group over time. So essentially, from week 4 all the way through week 24, and overall, there was a significant benefit olaparib in overall pain interference.

There's a similar file looking at the specific domains of the BPI-SF subscale score. Again, we can see a benefit for olaparib in all these domains, including general activity, mood, walking ability, normal work, relationships with others, sleep, and enjoyment of life. This shows the time to deterioration in health related quality of life. Across several of these metrics, we can see here on the left, and in summary for this figure, these were comparable between the olaparib and the control group, as noted by the p-values on the right. This final figure looks at the Kaplan-Meier analysis of time to first symptomatic skeletal-related event, which significantly benefited the olaparib group. The median was not reached in each of these arms. However, the hazard ratio of favors olaparib, at 0.37 and a 95% confidence interval of 0.20 to 0.70.

In PROfound, compared to the control group, olaparib was associated with symptomatic improvement in the burden of pain, better preserved health related quality of life, delayed time to first symptomatic skeletal-related event, and delayed time to first opioid use. Furthermore, PRO data are considered important for treatment decisions making for both of patients and physicians, and the importance of improving health related quality of life and reducing the burden of pain in patients with mCRPC is highlighted by the substantial, symptomatic and emotional burden association with this disease and the associations between these factors and overall survival.

So in conclusion, collectively, the symptomatic and health related quality of life benefits with olaparib compared with two controlled drugs support the clinical benefit of improved rPFS and OS that has already been reported in men with mCRPC and HRR gene alterations who had progression while receiving previous next-generation hormonal drugs. Health related quality of life is an important consideration in patient with metastatic CRPC, and these analyses have shown that compared to two controlled drugs, olaparib was associated with better control of disease symptoms via reduction in the burden of pain and better preserved health related quality of life. Thank you very much for your attention to this UroToday Journal Club discussing the recently published PROfound health related quality of life.