The Efficacy of PARP inhibition in BRCA1- versus BRCA2-altered mCRPC, Journal Club - Christopher Wallis & Zachary Klaassen
December 14, 2021
Christopher Wallis and Zachary Klaassen discuss a publication entitled “Differential Activity of PARP Inhibitors in BRCA1-Versus BRCA2-Altered Metastatic Castration-Resistant Prostate Cancer.” Olaparib and Rucaparib are US Food and Drug Administration-approved for patients with metastatic castration-resistant prostate cancer (mCRPC) harboring BRCA1/2 mutations. This data aimed to study the relative efficacy of PARP inhibition in BRCA1- versus BRCA2-altered mCRPC.
Biographies:
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
Biographies:
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. Today, we are discussing a recent publication entitled Differential Activity of PARP Inhibitors in BRCA1-Versus BRCA2-Altered Metastatic Castration-Resistant Prostate Cancer. I'm Chris Wallis, an Assistant Professor in the Division of Urology at the University of Toronto. With me today is Zach Klaassen, an Assistant Professor in the Division of Urology at the Medical College of Georgia. This is the citation for this recent publication led by the senior author, Dr. Antonarakis.
The state of mCRPC has evolved quite dramatically in the last 10 years or so. However, there remains a generally poor prognosis in spite of a variety of new treatment approaches, including androgen receptor pathway inhibitors, taxane-based chemotherapy, and PARP inhibitors. In terms of the relevance of PARP inhibitors, we see that the germline variants most prevalent are BRCA1 and BRCA2, and so maybe targeted by agents which affect mismatch repair. When we look at the schematic here, we can see that there is synthetic lethality that is selective for cells with deficient homologous recombination repair pathways when we use PARP inhibitors. And so we currently have two FDA-approved PARP inhibitors, both rucaparib, and olaparib on the basis of the TRITON2 and PROfound studies, and both of these are indicated for men with advanced prostate cancer who received prior androgen-axis targeting agents with or without taxane therapy depending on the agent.
And while the indication spans both BRCA1 and BRCA2, as well as some other mutations, when we look at the post-hoc analysis, this data from TRITON2, but also supported by data from PROfound, we see a greater response rate, two PARP inhibitors in men who have BRCA2 compared with BRCA1. And so you can see here, partial response rates are at least 10% higher and progressive disease is much less common in men with BRCA2 compared with BRCA1.
And so the present study is a multi-center retrospective analysis of 123 patients at 12 academic institutions who received PARP inhibitors for mCRPC between 2014 and 2020. And they were included if they received treatment with any PARP inhibitor and had a deleterious germline or somatic mutation in BRCA1 or BRCA2, and a variety of different mutations were considered, provided that they were functionally deleterious.
The authors collected information in terms of demographic, clinical, and genomic data, including standard baseline characteristics, as well as the type and number of previous systemic therapies, and additionally assessed for concurrent alterations in P53, P10, RB1, SPOP, the androgen receptor, and TMPRSS2-ERG. The primary endpoint of interest was to compare the proportion of men with a PSA50 response between those who had BRCA1 and BRCA2 mutations. Additional secondary endpoints included PSA-progression-free survival, progression-free survival, and overall survival. The authors assumed a 60% PSA50 rate in BRCA2 patients and a 20% in BRCA1 patients, thus to achieve an 80% power with a one-sided alpha of 0.05, required 120 patients. The authors used logistic regression modeling to compare PSA50 rates and Cox proportional hazards models, and the Kaplan-Meier technique to compare time to event outcomes. The multi-variable models included age, Gleason score, M stage, baseline PSA, and prior taxane use as a binary variable. Additionally, separate models accounted for both P53 and P10 mutations.
At this point in time, I'm going to hand it over to Zach to walk us through the results of this interesting analysis.
Zachary Klaassen: Thanks Chris. So you can see here, this is the baseline demographic, clinical, and genomic data overall and by mutation type. This was a large table, so I've broken this down into two slides. So this is the first slide. As you can see here on the far right is BRCA2 patients, 110, and to the left of that is BRCA1, 13 patients. Overall, the median age of these patients was quite similar, 71 years in BRCA1, 66.5 in BRCA2. In terms of baseline PSA, slightly higher in the BRCA1 group at 47.2 median PSA, and 43.8 in the BRCA2 patients. Moving down to the M stage of diagnosis, 69.2% of patients in BRCA1 were M1 compared to 37.3 of BRCA2 patients. In terms of prior docetaxel use, 76.9% of patients who were BRCA1 had prior docetaxel compared to 58.2% of BRCA2 patients. The majority of these patients had both enzalutamide or abiraterone before enrollment in the trial. The majority of these patients also had bone metastases, 92% in the BRCA1 group and 85.8 in the BRCA2 group. Visceral metastases were more common in the BRCA2 group with 25.5% compared to the BRCA1 group at 7.7%, with about 50% of these patients also having no metastases. In terms of the origin of mutation, about a 50/50 split between germline and somatic mutation for both BRCA1 and BRCA2.
This is the second half of this table, looking at some of the mutation types; so in terms of concurrent P53 mutations, 54.5% of patients were BRCA1 and 35.9 were BRCA2; concurrent P10 mutations, 9.1% of BRCA1 patients, and 22.2% of BRCA2 patients; concurrent RB1, no patients in the BRCA1 group and 19.2% in the BRCA2 group; concurrent SPOP mutations, no patients in the BRCA1 group and 4% of patients in the BRCA2 group; concurrent AR mutations, 9.1% in the BRCA1 group and 19.4% in the BRCA2 group; and finally TMPRSS2-ERG concurrent mutations, 9.1% in BRCA1 and 14.1% in BRCA2.
This is a summary of the key genetic alterations in this cohort of 123 patients. You can see here BRCA2 by far the most common mutation at 89%. This is followed by P53 at 38%, P10 mutation at 21%, AR mutation at 18%, RB1 mutation at 17%, TMPRSS2-ERG at 13%, BRCA1 at 11%, and SPOP mutation at 3%. This is a waterfall plot of the best PSA response. You can see that BRCA2 is highlighted in red and BRCA1 is highlighted in blue, and roughly about two-thirds of these patients had a PSA decrease from baseline and if you look to the left of the patients who had an increase in PSA, a slightly higher proportion of BRCA1 patients compared to BRCA2.
This is a forest plot for the best PSA response and just a couple of highlights on these forest plots. So, particularly poor PSA response, which is with blue dots to the left of this figure, BRCA1, a missense mutation, as well as a concurrent AR mutation, and concurrent TMPRSS2-ERG mutation. In terms of good performers with blue dots to the right of the figure, concurrent SPOP mutation was associated with a strong PSA response, but remember there were only four patients with an SPOP mutation in this study. This is the Kaplan-Meier analysis of PSA-progression-free survival, again, BRCA1 in blue and BRCA2 in red, and although not statistically significant, you can see there was improved PSA-progression-free survival for BRCA2 versus BRCA1, with a Cox proportional hazard model of BRCA2 versus BRCA1 of 1.94, and a nonsignificant 95% confidence interval.
This is the forest plot for PSA-progression-free survival. Again, you can see poor performers compared to the overall or less robust performers I should say, is BRCA1 missense mutations, as well as monoallelic mutations. Similarly, as we saw in the PSA response, concurrent AR mutation was also associated with a less PSA-progression-free survival, as well as P53. On the other side of the bracket, in terms of being associated with a better response, again SPOP mutation and subsequently patients that had homozygous deletion versus the rest of the deletions. Looking at the Kaplan-Meier analysis for progression-free survival, again, BRCA2 with a more favorable looking Kaplan-Meier curve, although not statistically significant with a hazard ratio for BRCA1 versus BRCA2 of 2.08 and a 95% confidence interval of 0.99 to 4.4.
A similar-looking plot for progression-free survival. We see a little bit better performance for BRCA1 compared to the previous slides of PSA response and progression-free survival of PSA. In terms of less robust performers for progression-free survival, missense mutations, concurrent P53 mutations, AR mutations, and again we see this signal in the SPOP mutations of being favorable responders, as well as patients that had no prior docetaxel or cabazitaxel therapy. Finally, we'll look at Kaplan-Meier analysis for overall survival. This was statistically significant favoring BRCA2, looking at the hazard ratio, worse survival for BRCA1 with a hazard ratio of 3.01 and 95% confidence interval of 1.32 to 6.83. So, improved overall survival for BRCA2 versus BRCA1. A final forest plot for you to have a look at is the overall survival forest plot, BRCA1 as mentioned before, poor performer for overall survival as was missense mutations, concurrent P53, P10, and looking again on the other side of the spectrum, we see slightly improved outcomes with patients that did not receive docetaxel or cabazitaxel chemotherapy.
In the next several slides we will work through a multi-variable model, looking at PSA50 response, PSA-progression-free survival, progression-free survival, and overall survival. So this model is for the multi-variable analysis of PSA50 response, and the only significant factor in this model was BRCA1 versus BRCA2 with an odds ratio of 0.20 and a 95% confidence interval of 0.04 to 0.76. Looking at the multi-variable analysis of PSA-progression-free survival, there were actually no significant predictors in this model for progression-free survival for PSA. Moving to the multi-variable analysis for progression-free survival, the notable variable that was significant in this model was prior docetaxel or cabazitaxel therapy. So yes versus no, with a hazard ratio of 1.66 and a 95% confidence interval of 1.02 to 2.71. And finally, the multi-variable analysis for overall survival, not surprisingly again, BRCA1 versus BRCA2 with a hazard ratio of 3.50 and a 95% confidence interval of 1.36 to 9.02, and similarly to progression-free survival prior docetaxel or cabazitaxel, yes versus no, with a hazard ratio of 2.53 and a 95% confidence interval of 1.34 to 4.79.
So several discussion points from this study are notable. The clinical activity of PARP inhibitors is diminished in metastatic CRPC patients with BRCA1 compared with BRCA2 mutations, and this study showed that this was not driven by differences in baseline demographics or clinical characteristics, and generally, there were improved outcomes in patients with biallelic versus monoallelic mutations, but not in those with germline versus somatic mutations. The presence of concurrent P53 mutations, which was 55% of BRCA1 compared to 36% of BRCA2 cancers was associated with a worse response to PARP inhibitors.
As I mentioned in the previous slides, numerically better outcomes were reserved in patients with concurrent SPOP mutations, but again, this should be seen as a precaution as this represented a very small subset, only 4% of the total population.
So in conclusion, this is the first dedicated study examining the potential differential effect of PARP inhibitor efficacy in BRCA1-versus BRCA2-associated prostate cancer. PARP inhibitor activity is attenuated in mCRPC patients with BRCA1 mutations and sensitivity is highest in those with BRCA2 mutations, and this is likely secondary to a greater number of monoallelic mutations and a higher prevalence of concurrent P53 alterations in BRCA1 patients. Taken together, additional treatment options are needed for BRCA1-altered metastatic CRPC patients.
Thank you very much, and we hope you enjoyed this UroToday Journal Club discussion.
Christopher Wallis: Hello, and thank you for joining us for this UroToday Journal Club. Today, we are discussing a recent publication entitled Differential Activity of PARP Inhibitors in BRCA1-Versus BRCA2-Altered Metastatic Castration-Resistant Prostate Cancer. I'm Chris Wallis, an Assistant Professor in the Division of Urology at the University of Toronto. With me today is Zach Klaassen, an Assistant Professor in the Division of Urology at the Medical College of Georgia. This is the citation for this recent publication led by the senior author, Dr. Antonarakis.
The state of mCRPC has evolved quite dramatically in the last 10 years or so. However, there remains a generally poor prognosis in spite of a variety of new treatment approaches, including androgen receptor pathway inhibitors, taxane-based chemotherapy, and PARP inhibitors. In terms of the relevance of PARP inhibitors, we see that the germline variants most prevalent are BRCA1 and BRCA2, and so maybe targeted by agents which affect mismatch repair. When we look at the schematic here, we can see that there is synthetic lethality that is selective for cells with deficient homologous recombination repair pathways when we use PARP inhibitors. And so we currently have two FDA-approved PARP inhibitors, both rucaparib, and olaparib on the basis of the TRITON2 and PROfound studies, and both of these are indicated for men with advanced prostate cancer who received prior androgen-axis targeting agents with or without taxane therapy depending on the agent.
And while the indication spans both BRCA1 and BRCA2, as well as some other mutations, when we look at the post-hoc analysis, this data from TRITON2, but also supported by data from PROfound, we see a greater response rate, two PARP inhibitors in men who have BRCA2 compared with BRCA1. And so you can see here, partial response rates are at least 10% higher and progressive disease is much less common in men with BRCA2 compared with BRCA1.
And so the present study is a multi-center retrospective analysis of 123 patients at 12 academic institutions who received PARP inhibitors for mCRPC between 2014 and 2020. And they were included if they received treatment with any PARP inhibitor and had a deleterious germline or somatic mutation in BRCA1 or BRCA2, and a variety of different mutations were considered, provided that they were functionally deleterious.
The authors collected information in terms of demographic, clinical, and genomic data, including standard baseline characteristics, as well as the type and number of previous systemic therapies, and additionally assessed for concurrent alterations in P53, P10, RB1, SPOP, the androgen receptor, and TMPRSS2-ERG. The primary endpoint of interest was to compare the proportion of men with a PSA50 response between those who had BRCA1 and BRCA2 mutations. Additional secondary endpoints included PSA-progression-free survival, progression-free survival, and overall survival. The authors assumed a 60% PSA50 rate in BRCA2 patients and a 20% in BRCA1 patients, thus to achieve an 80% power with a one-sided alpha of 0.05, required 120 patients. The authors used logistic regression modeling to compare PSA50 rates and Cox proportional hazards models, and the Kaplan-Meier technique to compare time to event outcomes. The multi-variable models included age, Gleason score, M stage, baseline PSA, and prior taxane use as a binary variable. Additionally, separate models accounted for both P53 and P10 mutations.
At this point in time, I'm going to hand it over to Zach to walk us through the results of this interesting analysis.
Zachary Klaassen: Thanks Chris. So you can see here, this is the baseline demographic, clinical, and genomic data overall and by mutation type. This was a large table, so I've broken this down into two slides. So this is the first slide. As you can see here on the far right is BRCA2 patients, 110, and to the left of that is BRCA1, 13 patients. Overall, the median age of these patients was quite similar, 71 years in BRCA1, 66.5 in BRCA2. In terms of baseline PSA, slightly higher in the BRCA1 group at 47.2 median PSA, and 43.8 in the BRCA2 patients. Moving down to the M stage of diagnosis, 69.2% of patients in BRCA1 were M1 compared to 37.3 of BRCA2 patients. In terms of prior docetaxel use, 76.9% of patients who were BRCA1 had prior docetaxel compared to 58.2% of BRCA2 patients. The majority of these patients had both enzalutamide or abiraterone before enrollment in the trial. The majority of these patients also had bone metastases, 92% in the BRCA1 group and 85.8 in the BRCA2 group. Visceral metastases were more common in the BRCA2 group with 25.5% compared to the BRCA1 group at 7.7%, with about 50% of these patients also having no metastases. In terms of the origin of mutation, about a 50/50 split between germline and somatic mutation for both BRCA1 and BRCA2.
This is the second half of this table, looking at some of the mutation types; so in terms of concurrent P53 mutations, 54.5% of patients were BRCA1 and 35.9 were BRCA2; concurrent P10 mutations, 9.1% of BRCA1 patients, and 22.2% of BRCA2 patients; concurrent RB1, no patients in the BRCA1 group and 19.2% in the BRCA2 group; concurrent SPOP mutations, no patients in the BRCA1 group and 4% of patients in the BRCA2 group; concurrent AR mutations, 9.1% in the BRCA1 group and 19.4% in the BRCA2 group; and finally TMPRSS2-ERG concurrent mutations, 9.1% in BRCA1 and 14.1% in BRCA2.
This is a summary of the key genetic alterations in this cohort of 123 patients. You can see here BRCA2 by far the most common mutation at 89%. This is followed by P53 at 38%, P10 mutation at 21%, AR mutation at 18%, RB1 mutation at 17%, TMPRSS2-ERG at 13%, BRCA1 at 11%, and SPOP mutation at 3%. This is a waterfall plot of the best PSA response. You can see that BRCA2 is highlighted in red and BRCA1 is highlighted in blue, and roughly about two-thirds of these patients had a PSA decrease from baseline and if you look to the left of the patients who had an increase in PSA, a slightly higher proportion of BRCA1 patients compared to BRCA2.
This is a forest plot for the best PSA response and just a couple of highlights on these forest plots. So, particularly poor PSA response, which is with blue dots to the left of this figure, BRCA1, a missense mutation, as well as a concurrent AR mutation, and concurrent TMPRSS2-ERG mutation. In terms of good performers with blue dots to the right of the figure, concurrent SPOP mutation was associated with a strong PSA response, but remember there were only four patients with an SPOP mutation in this study. This is the Kaplan-Meier analysis of PSA-progression-free survival, again, BRCA1 in blue and BRCA2 in red, and although not statistically significant, you can see there was improved PSA-progression-free survival for BRCA2 versus BRCA1, with a Cox proportional hazard model of BRCA2 versus BRCA1 of 1.94, and a nonsignificant 95% confidence interval.
This is the forest plot for PSA-progression-free survival. Again, you can see poor performers compared to the overall or less robust performers I should say, is BRCA1 missense mutations, as well as monoallelic mutations. Similarly, as we saw in the PSA response, concurrent AR mutation was also associated with a less PSA-progression-free survival, as well as P53. On the other side of the bracket, in terms of being associated with a better response, again SPOP mutation and subsequently patients that had homozygous deletion versus the rest of the deletions. Looking at the Kaplan-Meier analysis for progression-free survival, again, BRCA2 with a more favorable looking Kaplan-Meier curve, although not statistically significant with a hazard ratio for BRCA1 versus BRCA2 of 2.08 and a 95% confidence interval of 0.99 to 4.4.
A similar-looking plot for progression-free survival. We see a little bit better performance for BRCA1 compared to the previous slides of PSA response and progression-free survival of PSA. In terms of less robust performers for progression-free survival, missense mutations, concurrent P53 mutations, AR mutations, and again we see this signal in the SPOP mutations of being favorable responders, as well as patients that had no prior docetaxel or cabazitaxel therapy. Finally, we'll look at Kaplan-Meier analysis for overall survival. This was statistically significant favoring BRCA2, looking at the hazard ratio, worse survival for BRCA1 with a hazard ratio of 3.01 and 95% confidence interval of 1.32 to 6.83. So, improved overall survival for BRCA2 versus BRCA1. A final forest plot for you to have a look at is the overall survival forest plot, BRCA1 as mentioned before, poor performer for overall survival as was missense mutations, concurrent P53, P10, and looking again on the other side of the spectrum, we see slightly improved outcomes with patients that did not receive docetaxel or cabazitaxel chemotherapy.
In the next several slides we will work through a multi-variable model, looking at PSA50 response, PSA-progression-free survival, progression-free survival, and overall survival. So this model is for the multi-variable analysis of PSA50 response, and the only significant factor in this model was BRCA1 versus BRCA2 with an odds ratio of 0.20 and a 95% confidence interval of 0.04 to 0.76. Looking at the multi-variable analysis of PSA-progression-free survival, there were actually no significant predictors in this model for progression-free survival for PSA. Moving to the multi-variable analysis for progression-free survival, the notable variable that was significant in this model was prior docetaxel or cabazitaxel therapy. So yes versus no, with a hazard ratio of 1.66 and a 95% confidence interval of 1.02 to 2.71. And finally, the multi-variable analysis for overall survival, not surprisingly again, BRCA1 versus BRCA2 with a hazard ratio of 3.50 and a 95% confidence interval of 1.36 to 9.02, and similarly to progression-free survival prior docetaxel or cabazitaxel, yes versus no, with a hazard ratio of 2.53 and a 95% confidence interval of 1.34 to 4.79.
So several discussion points from this study are notable. The clinical activity of PARP inhibitors is diminished in metastatic CRPC patients with BRCA1 compared with BRCA2 mutations, and this study showed that this was not driven by differences in baseline demographics or clinical characteristics, and generally, there were improved outcomes in patients with biallelic versus monoallelic mutations, but not in those with germline versus somatic mutations. The presence of concurrent P53 mutations, which was 55% of BRCA1 compared to 36% of BRCA2 cancers was associated with a worse response to PARP inhibitors.
As I mentioned in the previous slides, numerically better outcomes were reserved in patients with concurrent SPOP mutations, but again, this should be seen as a precaution as this represented a very small subset, only 4% of the total population.
So in conclusion, this is the first dedicated study examining the potential differential effect of PARP inhibitor efficacy in BRCA1-versus BRCA2-associated prostate cancer. PARP inhibitor activity is attenuated in mCRPC patients with BRCA1 mutations and sensitivity is highest in those with BRCA2 mutations, and this is likely secondary to a greater number of monoallelic mutations and a higher prevalence of concurrent P53 alterations in BRCA1 patients. Taken together, additional treatment options are needed for BRCA1-altered metastatic CRPC patients.
Thank you very much, and we hope you enjoyed this UroToday Journal Club discussion.