Phase II Trial of Olaparib in Patients With Metastatic Urothelial Cancer Harboring DNA Damage Response Gene Alterations.

PURPOSE: Poly (ADP-ribose) polymerase (PARP) inhibitors have demonstrated clinical benefit for patients with solid tumors bearing germline or somatic alterations in DNA damage response (DDR) genes. Somatic alterations in DDR genes are common in advanced urothelial cancer, raising the possibility that PARP inhibition may confer therapeutic benefit in a molecularly selected subgroup of patients with metastatic urothelial cancer (mUC). METHODS: This single-arm, open-label, multi-institutional, investigator-initiated phase II study evaluated the antitumor activity of olaparib 300 mg twice a day in participants with mUC harboring somatic DDR alterations. Patients had progressed despite previous platinum-based chemotherapy, or were cisplatin-ineligible, and harbored somatic alterations in at least one of a prespecified list of DDR genes. The primary end point was objective response rate; secondary end points were safety, progression-free survival (PFS), and overall survival (OS). RESULTS: Overall, 19 patients with mUC were enrolled and received olaparib; the trial closed early before slow accrual. The median age was 66 years (range, 45-82). Nine patients (47.4%) had received previous cisplatin chemotherapy. Ten patients (52.6%) had alterations in homologous recombination (HR) genes: eight patients (42.1%) had pathogenic BRCA2 mutations and two patients carried alterations in other HR genes. No patients achieved a partial response although six patients achieved stable disease lasting 2.13-16.1 months (median, 7.69). The median PFS was 1.9 months (range, 0.8-16.1), and the median OS was 9.5 months (range, 1.5-22.1). CONCLUSION: Single-agent olaparib showed limited antitumor activity in patients with mUC and DDR alterations, which may be related to poorly characterized functional implications of particular DDR alterations and/or cross-resistance with platinum-based chemotherapy in a disease where such therapy represents standard first-line treatment.

Spectrum of FGFR2/3 Alterations in Cell-Free DNA of Patients with Advanced Urothelial Carcinoma.

Detecting genomic alterations (GAs) in advanced urothelial carcinoma (aUC) can expand treatment options by identifying candidates for targeted therapies. Erdafitinib is FDA-approved for patients with platinum-refractory aUC with activating mutation or fusion in FGFR2/3. We explored the prevalence and spectrum of FGFR2/3 GAs identified with plasma cfDNA NGS testing (Guardant360) in 997 patients with aUC. FGFR2/3 GAs were detected in 201 patients (20%) with characterized activating GAs in 141 (14%). Our results indicate the Guardant360-based FGFR2/3 GA detection rate is similar to those described from previous studies employing tumor tissue testing, suggesting that plasma-based cfDNA NGS may non-invasively identify candidates for anti-FGFR targeted therapies.

Urothelial cancer harbours EGFR and HER2 amplifications and exon 20 insertions.

OBJECTIVE: To review the genomic landscape of advanced urothelial carcinoma (UC) to assess the frequencies of EGFR and ERBB2 (HER2) alterations. MATERIALS AND METHODS: Tumour specimens from 3753 patients with advanced UC were assayed with hybrid capture-based comprehensive genomic profiling of 180-395 genes. Tumour mutational burden (TMB) was assessed on 0.8 or 1.1 Mb of DNA, and is reported as mutations per megabase. RESULTS: In 3753 cases of UC, EGFR alterations were detected in 4.1% (154) and were most commonly amplifications (64%; 99/154), while exon 20 insertions (EGFRexon20ins ) were the second most common alteration (18%; 27/154). Alterations in ERBB2 were observed in 15% (552/3753) of cases and, similarly, ERBB2 amplification was the most commonly observed alteration (278/552; 50%); ERBB2exon20ins occurred in 3.6% (20/552) of cases. EGFRexon20ins and ERBB2exon20ins occurred in younger patients (median age 62 vs 69 years, P = 2.6E-2 and 60 vs 68 years, P = 7.8E-4), and these cases had significantly lower TMB (median 3.6 vs 7.2, P = 2.7E-4 and 2.5 vs 10, P = 1.2E-7) and less frequent TP53 alterations (3.7% vs 83%, P = 4.3E-14 and 20% vs 68%, P = 9.8E-4) compared to cases with other EGFR or ERBB2 alterations. CONCLUSION: EGFR and ERBB2 alterations occur in 4% and 15% of UC, respectively. EGFRexon20ins and ERBB2exon20ins were present in 0.7% and 0.5% of UC overall and collectively define a small, but distinct, subset of UC with infrequent co-occurrence of other drivers and low TMB. Given recent promising clinical studies of inhibitors with activity against exon 20 insertions in non-small cell lung cancer, consideration should be given to developing a trial inclusive of patients with UC harbouring these alterations.

Circulating Tumor DNA Alterations in Advanced Urothelial Carcinoma and Association with Clinical Outcomes: A Pilot Study.

Cell-free circulating DNA (cfDNA) can be used for noninvasive profiling of tumor genomic aberrations. We hypothesized that molecular alterations may inform prognostication in advanced urothelial carcinoma (aUC). We evaluated 124 aUC patients who underwent cfDNA analysis using a 73-gene sequencing panel (Guardant360). The association of molecular alterations and clinical factors with overall survival (OS) and failure-free-survival (FFS) was evaluated using the Kaplan-Meier method and Cox proportional-hazards regression. The median age was 72yr, and 65 patients (52.4%) received prior therapy with platinum, 21 (17.1%) with a taxane, and ten (8.1%) with a PD-1/PD-L1 inhibitor. At least one genomic alteration was detected in 112 patients (90.3%). The median number of alterations per sample was four (range 0-80). Commonly altered genes included TP53 (54.8%), PIK3CA (24.2%), ARID1A (22.6%), ERBB2 (19.4%), EGFR (16.1%), NF1 (13.7%), RB1 (12.9%), FGFR3 (11.3%), BRAF (10.5%), BRCA1 (10.5%), and RAF1 (8.9%). BRCA1 and RAF1 alterations were associated with worse OS (hazard ratio [HR] 2.48; p=0.07; HR 4.87; p=0.007) and FFS (HR 2.35; p=0.016; HR 2.40; p=0.047). Poor Eastern Cooperative Oncology Group performance status and the presence of visceral metastasis were associated with shorter OS; genomic evolution was observed. In conclusion, cfDNA molecular alterations were detected in most aUC patients. BRCA1 and RAF1 alterations were negatively prognostic, supporting further evaluation of DNA damage response and RAF kinase inhibitors. PATIENT SUMMARY: Noninvasive testing of cell-free circulating DNA in advanced urothelial carcinoma identifies clinically relevant molecular aberrations. Alterations in BRCA1 and RAF1 genes appear to be negatively associated with clinical outcomes, supporting further study of DNA damage response and RAF kinase inhibitors in selected patients.

Characterization of metastatic urothelial carcinoma via comprehensive genomic profiling of circulating tumor DNA.

BACKGROUND: Biomarker-guided clinical trials are increasingly common in metastatic urothelial carcinoma (mUC), yet patients for whom contemporary tumor tissue is not available are not eligible. Technological advancements in sequencing have made cell-free circulating DNA (cfDNA) next-generation sequencing (NGS) readily available in the clinic. The objective of the current study was to determine whether the genomic profile of mUC detected by NGS of cfDNA is similar to historical tumor tissue NGS studies. A secondary objective was to determine whether the frequency of genomic alterations (GAs) differed between lower tract mUC (mLTUC) and upper tract mUC (mUTUC). METHODS: Patients from 13 academic medical centers in the United States who had a diagnosis of mUC between 2014 and 2017 and for whom cfDNA NGS results were available were included. cfDNA profiling was performed using a commercially available platform (Guardant360) targeting 73 genes. RESULTS: Of 369 patients with mUC, 294 were diagnosed with mLTUC and 75 with mUTUC. A total of 2130 GAs were identified in the overall mUC cohort: 1610 and 520, respectively, in the mLTUC and mUTUC cohorts. In the mLTUC cohort, frequently observed GAs were similar between cfDNA NGS and historical tumor tissue studies, including tumor protein p53 (TP53) (P = 1.000 and .115, respectively), AT-rich interaction domain 1A (ARID1A) (P = .058 and .058, respectively), phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA) (P = .058 and .067, respectively), erb-b2 receptor tyrosine kinase 2 (ERBB2) (P = .565 and .074, respectively), and fibroblast growth factor receptor 3 (FGFR3) (P = .164 and .014, respectively). No significant difference was observed with regard to the frequency of GAs between patients with mLTUC and mUTUC. CONCLUSIONS: Among patients with mUC for whom no tumor tissue was available, cfDNA NGS was able to identify a similar profile of GAs for biomarker-driven clinical trials compared with tumor tissue. Despite the more aggressive clinical course, cases of mUTUC demonstrated a circulating tumor DNA genomic landscape that was similar to that of mLTUC. Cancer 2018;124:2115-24. © 2018 American Cancer Society.