Read the Full Video Transcript

Christopher Wallis: Hello, and thank you for joining us for this UroToday Journal Club. Today, we are discussing the recently published TALAPRO-1 study: talazoparib monotherapy in metastatic castration-resistant prostate cancer with DNA repair alterations; an open-label, phase II trial. I'm Chris Wallis, an Assistant Professor in the Division of Urology at the University of Toronto. With me today, is Zach Klaassen, Assistant Professor in the Division of Urology at the Medical College of Georgia.

This is the citation for this recent publication in Lancet Oncology.

By way of background, alterations in DNA repair pathways are relatively common in advanced prostate cancer, and this paper from Dr. Nicolosi and colleagues in JAMA Oncology highlights this. Among 3,600 men with prostate cancer, germline variants were seen in 17.2%, with BRCA1 and 2 variants comprising over 30%.

In addition to the prognostic association, this offers the opportunity for a biomarker-driven targeted treatment, and so this highlights the mechanism of action of PARP inhibitors, which function to trap PARP on damaged DNA, prevent DNA repair, and induce cellular death. This depends on homologous recombination repair deficiency within the cell, as you can see here on the right side of the screen.

So in prostate cancer, one of the first demonstrations of the differential efficacy of PARP inhibitors is seen here in the work of Dr. Mateo, published in 2015, in this phase II trial among approximately 50 patients with extensively pre-treated mCRPC. Among the 16 patients with mutations in DNA repair genes, we can see that the benefits, both in terms of radiographic progression-free survival and overall survival with olaparib, were substantially larger than in those who were biomarker negative.

Now, this led to the subsequent PROfound trial, randomized comparison of olaparib and ongoing androgen receptor inhibition among patients with relevant homologous recombination repair mutations, with a focus on BRCA1, BRCA2, and ATM. So you can see here that both imaging-based progression-free survival and overall survival improved for patients receiving olaparib if they had these relevant mutations.

While olaparib has been well studied and rucaparib is already approved, talazoparib is actually the most efficient PARP inhibitor at trapping PARP1 and PARP2 on DNA single-strand breaks. Additionally, it is currently approved to treat patients with breast cancer, and so the TALAPRO-1 study assesses its role in prostate cancer.

This is an international, open-label, phase II single-arm trial, looking at patients with DNA repair deficiencies, or homologous recombination repair deficiencies, and mCRPC. So patients had to have progressive mCRPC defined by at least three increasing PSA levels or radiographic progression defined by RECIST or Prostate Cancer Working Group 3 definitions. Additionally, they had to have an ECOG performance status of 0-2; have ongoing androgen deprivation or bilateral orchiectomy; have castration levels of testosterone; a life expectancy of at least six months; as well as notably 1-2 prior lines of taxane-based chemotherapy, including docetaxel; as well as at least one novel hormonal agent, including abiraterone or enzalutamide.

Potentially eligible patients were then screened using the FoundationOne CDx next-generation sequencing to identify relevant mutations. Initially, the study was divided into two cohorts, cohorts A and B as you see here, however, from February 2018 onwards, patients had to have alterations in one of a panel of 11 genes — demonstrated on the right side of the screen — to be eligible for inclusion. The authors further assessed whether these mutations are germline or somatic based on a comparison of tumor and matched saliva DNA sequencing.

Patients were given talazoparib one milligram daily, with dose adjustment for both renal insufficiencies, as well as grade 3-4 adverse events. Treatment was continued until progression, toxicity, investigator decision, withdrawal of consent, or death.

Patients were assessed with CT and MRI as well as bone scan during the screening process, as well as every 8 weeks of the first 24, then every 12 weeks thereafter. Soft tissue responses were assessed according to RECIST criteria, and bony progression according to the Prostate Cancer Working Group 3. Laboratory testing was similarly performed at the time of screening, every 2 weeks until week 9, every 4 weeks until week 25, then every 12 weeks thereafter. CTC counts were also assessed, and this was performed using the CELLSEARCH platform.

The authors' primary outcome was confirmed objective response rate by blinded review, as well as by investigator assessment or a host of secondary outcomes listed here, as well as exploratory analyses conducted according to potential biomarkers.

The authors planned a sample size of 100 patients, which would show that the objective response rate was at least 23% with a lower bound of the confidence interval higher than 15.2%. They planned a number of interim safety and anti-tumor analyses, however, they only reported the final analysis.

The analytic population comprised patients with measurable soft tissue disease at the time of screening, alterations in one of the 11 predefined genes, who had received at least one dose of study therapy. Post hoc analyses according to alteration group were performed, and the authors also analyzed the objective response rate according to single and co-occurring alterations. Further subgroup analyses assessed the effect of baseline demographic characteristics, and the authors finally considered CTC conversion as a decrease in the number of CTCs from greater than or equal to 5 to less than 5 per 7.5-milliliter sample of blood.

Now, I'm going to pass it over to Zach to walk us through the results of TALAPRO-1.

Zachary Klaassen: Thanks Chris. So this is the trial profile, and much like the PROfound trial, many patients had to be screened in order to enroll. A subset of patients, and we can see that here, 128 patients were enrolled, ultimately 127 patients received at least one dose of the study drug, making up the safety population, and 104 patients had HRR deficient measurable disease, which made up the anti-tumor activity population.

So this is the table 1 baseline patient characteristics in the safety population and the anti-tumor activity population. It's a busy slide, so we will focus on several of the columns here. So in the first column, to the right of the variables, is the anti-tumor activity population. We can see here that the median age was 69 years. The majority of patients were white, at 87%. The majority of patients had a PSA over 100, and I'll point out here that in the overall population, this was a 103.8 median PSA. However, we can see for some of these other more rare alterations, the PSA was 178.5 for ATM and 153.5 for others. Looking at the most common Gleason group score, it was Gleason group 4 and 5, at 61%. The initial M stage at diagnosis was 39% for M0, 45% for M1, and 13% for MX.

In terms of disease site, visceral versus non-visceral, 68% of patients had non-visceral metastases compared to 32% that had visceral. The majority of these patients were ECOG 0 and 1, however, of note, there was 9% of patients that were ECOG performance status 2. In terms of previous taxane use, 51% of patients had docetaxel only compared to 48% that had docetaxel and cabazitaxel, and with regards to previous novel hormone therapy, 35% had abiraterone only, 36% had enzalutamide only, and 27% had abiraterone and enzalutamide.

This table looks at the anti-tumor activity assessments overall and by the HRR gene alteration group, and we will focus the majority of our time on this table looking at the top half. So this is broken down by BRCA1/BRCA2, BRCA2 on its own, PALB2, ATM, and others. Looking at confirmed complete response, 11% for BRCA2, compared to 6% for ATM, 0% for PALB2, and 0% for the other alterations. Confirmed partial response was 36% for BRCA1 or BRCA2, 25% for PALB2, and 6% for ATM.  Interestingly, stable disease of any duration was 34% for BRCA1 or 2, 50% for PALB2, and even ATM at 35%, and 36% for others. So some of these traditionally less responsive gene alterations did have stable disease of any duration while on treatment.

I've highlighted in the box the objective response rate, which is probably the most important aspect of this table. For BRCA1 or BRCA2, it was 46%, for BRCA2 on its own, it was 46%, for PALB2 it was 25%, ATM was 12%, and for other, was 0%. Finally, at the bottom of the slide, the median radiographic progression-free survival was 11.2 months for BRCA1 or 2, 5.6 months for PALB2, 3.5 for ATM, and 1.8 months for the other alterations.

This is a subgroup analysis of the objective response and PSA response rate, with the PSA response rate on the right and the objective response in the middle. We can see that, especially for objective response rate, this is a pretty consistent benefit across several subgroups, with a bit of a less benefit for PSA responses. You can see here by the forest plot.

In the next couple of slides, we will look at several waterfall plots, and this one is the best change from baseline in the sum of diameters for target lesions by blinded independent central review. You can see the overall reduction in tumor burden was 80%, which is quite good. And specifically, the author notes that for BRCA1 and 2, it was a 90% change from baseline in terms of tumor diameter.

This is the best change from baseline in PSA assessment. An overall reduction in PSA was seen in 72% of patients, with BRCA1 and 2 having an 85% decrease in PSA from baseline. And looking briefly at the left side of this plot, in the patients that did not have a PSA response, the majority of these were either ATM or the other recombinant gene alterations.

This waterfall plot looks at the best change from baseline in CTC count. Overall, there was an 82% reduction in CTC count, with an impressive 93% specific to those with BRCA1 or 2 mutations. And again, looking at the left of the slide, the patients that did not have a decrease in CTC count were primarily the other alterations in the green columns.

So this is a swimmer plot looking at the time to response by BICR in patients with confirmed complete response or partial response, which was 31 patients. We can see the median time to objective response was 3.4 months, with an IQR of 1.8 to 5.4. The median duration of response was 12.8 months, with a 95% confidence interval of 6.5 to not evaluable.

These Kaplan-Meier curves look at the radiographic progression-free survival and overall survival. So on the left is radiographic progression-free survival. You can see that the orange line is all of the patients grouped together, the blue line is BRCA1 or BRCA2, and the green line is ATM. We can see that — not surprisingly — BRCA1 and 2 had a radiographic progression-free survival median of 11.2 months, compared to the overall group at 5.6 months, and the ATM group of 3.5 months. Looking at overall survival, again, the overall group is in gray, the non-BRACA1 and 2 are in green, and the BRCA1 and 2 is in blue. The BRACA1/2, again, with a longer median overall survival of 24 months, compared to the non-BRACA1/2 of 11.5 months, and the overall group of 16.4 months.

This is quite a large table of all-cause treatment-emergent events in the safety population. Focusing on the grade 4, very few grade 4 events that we can see. Any hematological grade 4 event was present in 4% of patients, as well as 3% of patients, having thrombocytopenia. Looking at the grade 3 events, any grade 3 event was seen in 45% of patients. Most commonly, we see that this was pulmonary embolism at 5% of patients, as well as any hematological grade 3 event at 32%, most notably anemia at 31%, neutropenia at 8%, and thrombocytopenia at 6%.

So several discussion points from the TALAPRO-1 trial. This trial showed that talazoparib had a robust anti-tumor activity that was most notable and durable against tumors with BRCA1 or BRCA2 gene alterations. Also, this confirmed that there were responses in patients with alterations in PALB2 and ATM alone. Additionally, there was stable disease of any duration observed in patients with alterations in PALB2, ATM, and the other rare HHR genes, such as ATR, CHECK2, FANCA, MLH1, MRE11A, NBN, and RAD51C.

Importantly, the results in TALAPRO-1, which showed an objective response rate in BRCA1/2 of 46%, are similar to the other PARP inhibitor trials, which Chris discussed at the beginning of the slides. If we look at PROfound, in the olaparib arm, the objective response rate in BRCA1/2 or ATM was 33.3%. In the TRITON2 study, which looked at rucaparib, the objective response rate in BRCA1 and 2 was 43.5%, and in ATM was 10.5%. Finally, in the GALAHAD study, which looked at niraparib, the objective response rate in BRCA1/2 was 41.4%. So we can see that the objective response rate in TALAPRO-1 of 46% is quite consistent with these other studies.

So in conclusion, the benefit-risk profile in TALAPRO-1 suggests that talazoparib may provide effective therapy for advanced metastatic castrate-resistant prostate cancer with DDR alterations either directly or indirectly associated with HRR. There is evidence of durable anti-tumor activity, with a favorable benefit-risk profile in men who have been pretreated with novel hormone therapy and taxane chemotherapy. Finally, these results support the further assessment of talazoparib, which is in the ongoing phase III TALAPRO-2 trial, which is looking at talazoparib plus enzalutamide in the first-line setting of MCRPC plus or minus HRR gene alterations; as well as in the recruiting phase III TALAPRO-3 trial, which is looking at talazoparib plus enzalutamide in metastatic hormone-sensitive prostate cancer patients with HRR gene alterations.

Thank you very much. We hope you enjoyed this UroToday Journal Club discussion on the recently published TALAPRO-1 study.