A more personalized approach to the treatment of genitourinary cancers was discussed during the Clinical Science Symposium, “Evolving Role of DNA Repair in Genitourinary Cancer Management.” Three abstracts presented during the session evaluated the role of DNA repair genes in the treatment of metastatic prostate cancer and urothelial carcinoma.
During the first presentation, Peter Nelson, MD, of Fred Hutchinson Cancer Research Center, discussed the results of an analysis of inherited mutations in DNA repair genes in men with metastatic castration-resistant prostate cancer (Abstract 5009), which found that 11.8% of unselected men with prostate cancer had deleterious DNA repair gene mutations.
According to Dr. Nelson, the prevalence of germline DNA repair gene mutations in men with localized prostate cancer for family predisposition is insufficient to warrant routine testing. However, the frequency of these mutations in metastatic prostate cancer had not been established.
For the study, Dr. Nelson and colleagues recruited 692 men with metastatic prostate cancer from eight institutions. All participants were unselected for a family history of cancer. The researchers isolated germline DNA from blood, saliva, or tissue uninvolved with cancer and used multiplex sequencing assays to assess germline mutations in 20 DNA repair genes associated with autosomal dominant cancer predisposition syndromes.
The analysis identified 84 presumed deleterious germline DNA repair gene mutations in 82 men (11.8%). The mutations were found in 16 genes including BRCA2 (44%), ATM (13.5%), CHEK2 (11.0%), BRCA1 (7.0%), RAD51D (4.0%), and PALB2 (4.0%).
Among both men with and without DNA repair gene mutations about 22% had a first-degree relative with prostate cancer. However, 71% of men with the mutations had a first-degree relative with cancers other than prostate cancer compared with 50% of the men without mutations (OR=2.4; 95% CI[1.4-4.3]; p = 0.001).
Dr. Nelson and colleagues compared the frequency found in their study with data on localized prostate cancer taken from The Cancer Genome Atlas (TCGA), which sequenced tumors from approximately 500 men. The frequency of germline DNA repair gene mutations in metastatic prostate cancer significantly exceeded the prevalence of 4.6% found in TCGA (p < 0.001), including men with high-risk disease.
“We would argue that all men with metastatic prostate cancer should have germline testing,” Dr. Nelson told the ASCO Daily News. “The figure of 11% warrants it. There are emerging therapies that would specifically be used in men with DNA repair defects and that is platinum chemotherapy, which is used in ovarian cancer with DNA repair defects and also PARP inhibitors.”
Although PARP inhibitors are not yet approved for men with prostate cancer and germline DNA repair aberrations, a recent study showed that men with DNA repair defects had very good partial responses to the PARP inhibitor olaparib and those without DNA repair defect had minimal response.
In addition, these men with mutations could also be sentinels for their family, Dr. Nelson said.
“If a man is found to have DNA repair defect in the germline, then their family members, sisters, brothers, daughters, or sons are at risk for a malignancy and should be tested,” he said.
In a discussion of these results, Judy Ellen Garber, MD, MPH, of the Dana-Farber Cancer Institute, said that not looking in prostate cancer for these mutations has certainly been understandable, but it is past time to recognize that these mutations are present.
Dr. Maha Hussain
Co-Targeting PARP and AR
Maha Hussain, MD, FACP, FASCO, presented the results of a biomarker-based trial in men with metastatic castration-resistant prostate cancer designed to test whether the addition of a PARP-1 targeted therapy was superior to hormone therapy alone and whether ETS gene fusion was a predictive biomarker for response (Abstract 5010).
The trial included 159 patients who were stratified by ETS status and randomly assigned to abiraterone plus prednisone with or without the PARP-1 inhibitor veliparib. There were 148 patients evaluable for response. Overall, the best overall PSA response rate was 64% for abiraterone plus prednisone compared with 71% for patients given the addition of veliparib.
“We designed the study with the objective of declaring success if there was a 20% improvement in PSA response rate overall, but when we look at the overall response rate we did not achieve that 20% difference between arms,” Dr. Hussain told the ASCO Daily News.
There was also no significant difference in the median progression-free survival (PFS), which was 8.8 months for abiraterone/prednisone and 11.0 months with the addition of veliparib (p = 0.87). No differences in response rate or progression-free survival were found by ETS status.
Dr. Hussain and colleagues also evaluated a subgroup of patients with available tissue (87 patients) for DNA damage repair mutations. Twenty-five percent of patients had a DNA repair gene deficiency; 59.1% had BRCA2, and 22.7% had ATM mutations.
The treatment was very well-tolerated overall in both arms. Hyperglycemia was the only high-grade adverse event that occurred in more than 5% of patients in any one arm.
Dr. Michael J. Morris
Avoiding Surgery in Urothelial Carcinoma
In the final abstract presentation of the session, Gopa Iyer, MD, of Memorial Sloan Kettering Cancer Center, presented results of a study that found an association between patients with urothelial carcinoma having specific DNA damage repair gene alterations and improved response to chemotherapy.
“In bladder cancer, especially muscle-invasive tumors, the primary standard of care treatment is to give chemotherapy typically with a cisplatin-based regimen and then have the patient undergo surgery to remove the bladder,” Dr. Iyer told the ASCO Daily News. “Sometimes when we do surgery to remove the bladder, there is no evidence of cancer in the bladder because the chemotherapy worked so well that it eradicated all remnants of disease in the bladder. Those are the patients who have the best survival.”
Dr. Iyer said that removal of the bladder has a major effect on quality of life for patients with urothelial carcinoma and that chemotherapy received prior to removal has potential for long-term toxicity.
“If we can prospectively identify patients with DNA damage response gene alterations within the tumor, we might be able to select those patients for chemotherapy but not have them undergo radical cystectomy,” Dr. Iyer said. “Maybe we can spare the bladder, which would significantly improve quality of life for most patients.”
In this study, Dr. Iyer and colleagues hypothesized that alterations in DNA damage repair genes would predict pathologic response to chemotherapy. The phase II trial included 46 patients treated with six cycles of dose-dense gemcitabine and cisplatin. After chemotherapy, all patients underwent cystectomy to assess pathologic response.
The pathologic response rate in the trial was 56.5% with 26 patients responding to treatment and 20 not responding. The pathologic complete response rate was 15%. Two patients who did not respond had node-positive disease at the time of cystectomy.
Exon capture sequencing of 341 to 410 genes was performed on pre-treatment transurethral resection specimens and matched germline DNA on 34 patients; 21 samples were taken from responders. Candidate DNA damage repair genes predicting for response to cisplatin-based chemotherapy included ERCC2, ERCC5, BRCA2, RAD51C, ATR, and RECQL4.
Fifty percent of samples harbored at least one gene alteration. Most of the alterations occurred in the homologous recombination pathway, but they also occurred in the nucleotide excision repair pathway. The frequency of those mutations within the nucleotide excision repair pathway was driven by hot spot mutations within ERCC2. Twenty-six percent (nine patients) had deleterious gene alterations, and eight of these nine patients were responders.
“Of interest is that the deleterious DNA damage repair gene alterations are not simply within the response patients, but are actually enriched within those patients who are complete responders or had carcinoma in situ, in other words, the maximal amount of pathologic downstaging to chemotherapy,” Dr. Iyer said.
The specificity of DNA damage repair gene alteration for response was 92% and the positive predictive value was 88%.
“Those patients with deleterious DNA gene alterations had either no evidence of cancer within their bladder when they had cystectomy or had superficial disease, both of which are surrogates of having better survival,” Dr. Iyer told the ASCO Daily News.
“Dr. Iyer's study clinically validates ERCC2 mutations as predictive for exceptional response to cisplatin-based chemotherapy in muscle-invasive bladder cancer,” abstract discussant Eliezer Mendel Van Allen, MD, of Dana-Farber Cancer Institute, said. “In so doing, this study highlights the increasing ability to blend precision medicine genomics approaches with conventional chemotherapy, which is very exciting and may have relevance beyond this clinical indication.”
Furthermore, Dr. Van Allen said, Dr. Iyer's study demonstrates the need to “dig deeper” into the impact of additional DNA repair mutations and their impact on chemosensitivity.
“These results also lay the foundation for offering select patients a bladder-sparing, chemotherapy alone approach to the management of their muscle-invasive bladder cancer,” he said.