Publication
Article
Author(s):
Erdafitinib reduced the risk of death by 36% vs investigator’s choice of chemotherapy in patients with FGFR2/3-altered metastatic urothelial cancer who were previously treated with anti–PD-1 therapy, according to findings from the phase 3 THOR trial.
Erdafitinib (Balversa) reduced the risk of death by 36% vs investigator’s choice of chemotherapy in patients with FGFR2/3-altered metastatic urothelial cancer who were previously treated with anti–PD-1 therapy, according to findings from the phase 3 THOR trial (NCT03390504). The results were presented in a late-breaking abstract session during the 2023 ASCO Annual Meeting by Yohann Loriot, MD, PhD, who states that the targeted therapy should be considered the new standard of care for this patient population.1
At a median follow-up of 15.9 months, the median overall survival (OS) was 12.1 months with erdafitinib (n = 136) vs 7.8 months with investigator’s choice chemotherapy (n = 130; HR, 0.64; 95% CI, 0.47-0.88; P = .005). Loriot explained that based on the interim analysis, the independent monitoring committee recommended to stop the study, unblind the data, and cross over patients from chemotherapy to erdafitinib.
The median progression-free survival (PFS) was 5.6 months with erdafitinib vs 2.7 months with chemotherapy, translating to a reduction in the risk of disease progression of death of 42% (HR, 0.58; 95% CI, 0.44-0.78; P = .0002). The overall response rate (ORR) was 45.6% with a complete response (CR) rate of 6.6% and a partial response (PR) rate of 39.0% in the erdafitinib arm. The ORR in the chemotherapy arm was 11.5% with 1 CR (0.8%) and a PR rate of 10.8%.
“The phase 3 THOR study supports the clinical efficacy of erdafitinib as the standard of care option for patients with metastatic urothelial cancer with FGFR alteration after immune checkpoint inhibitor treatment,” Loriot, deputy head of Therapeutic Innovation and Early Trials Department at Gustave Roussy Institute, INSERM 981, University Paris-Saclay in France said in a presentation of the data.
“The OS benefit of erdafitinib supports molecular testing of FGFR in all patients with metastatic urothelial carcinoma,” he added.
Daniel P. Petrylak, MD, echoed that sentiment in a discussion of the abstract, citing recent findings that showed only 41% of patients between 2019 and 2021 underwent testing for FGFR alterations, with only 30% of patients proceeding to erdafitinib, which he described as a disappointing number.2 “The most important take home message of the day is that all patients with metastatic urothelial carcinoma need to be tested for FGFR3 or FGFR2 alterations,” said Petrylak, a professor of medicine (medical oncology) and of urology and coleader of Cancer Signaling Networks at Yale Cancer Center in New Haven, Connecticut.
Erdafitinib received accelerated approval from the FDA on April 12, 2019, for the treatment of patients with metastatic or locally advanced urothelial carcinoma with susceptible FGFR3 or FGFR2 genetic alterations with disease progression on or following at least 1 platinum-based chemotherapy.3 The approval was based on results of phase 2 Study BLC2001 (NCT02365597); THOR was the planned study to confer benefit for regular approval.4
THOR is a randomized, open-label, multicenter study, open to patients with stage IV metastatic or unresectable urothelial cancer. Patients were screened for selected FGFR alterations and were assigned to 2 cohorts based on prior anti–PD-1 treatment. Those with prior anti–PD-1 therapy (cohort 1 for this analysis) were randomly assigned to erdafitinib or investigator’s choice of chemotherapy. Erdafitinib was given at a dose of 8 mg once daily, with uptitration to 9 mg once daily if serum phosphate level is 9.0 mg/dL or less and there are no associated adverse events are observed at day 14. Chemotherapy was docetaxel 75 mg/m2 or vinflunine 320 mg/m2 every 3 weeks.
Treatment was continued until disease progression, intolerable toxicity, or withdrawal of consent. Assessments via CT or MRI were performed every 6 weeks for 6 months, every 12 weeks for the following 6 months, then as clinically indicated. The primary end point was OS, with secondary end points of PFS, ORR, and safety.
In the prior anti–PD-1 cohort, 33.1% of patients in the erdafitinib arm had received 1 line of prior systemic therapy, with 24.3% having received chemotherapy plus anti–PD-1 treatment, and 8.1% receiving anti–PD-1 treatment alone. In the chemotherapy arm, 25.4% of patients received 1 line of prior systemic therapy with 11.5% receiving combination treatment and 12.3% receiving an anti–PD-1 inhibitor alone.
Among patients who received 2 prior lines of systemic therapy in the erdafinib arm (66.2%), the first line of therapy was chemotherapy alone in 56.6%, chemotherapy plus anti–PD-1 in 4.4%, and other in 5.1%. These rates in patients who received chemotherapy (74.6%) were 58.5%, 7.7%. and 8.5%, respectively. Second line of therapy for patients in the erdafitinib and chemotherapy arm were anti–PD-1 alone (55.9% vs 58.5%), chemotherapy alone (7.4% vs 10.8%), and other (2.9% vs 5.4%).
Other baseline characteristics between the arms were well balanced. The median age in the erdafitinib arm was 66 years (range, 32-85), most were men (70.6%), White (59.6%), and had the presence of visceral metastases (74.3%) with 22.8% being in the liver. Nearly all patients had an ECOG performance status of 0 or 1 (91.2%).
Primary tumors were present in the upper tract for 30.1% of patients in the erdafitinib arm and 36.9% of patients in the chemotherapy arm. PD-L1 low expression (combined positive score < 10) was detected in 92.7% of patients in the erdafitinib arm (among 96 evaluable patients) and 86.1% of patients in the chemotherapy arm (among 79 evaluable patients).
In terms of FGFR alterations, at baseline these were reported as mutations in 79.4% and 82.3%, fusions in 18.4% and 14.6%, and mutations and fusions in 1.5% and 2.3% of patients in the erdafitinib and chemotherapy arms, respectively.
Results were consistent across all subgroups and those with translocations and upper tract primary tumor location showing deriving the most benefit. Patients with FGFR translocations had a median OS of 16.4 months with erdafitinib vs 8.0 months with chemotherapy (HR, 0.49; 95% CI, 0.23-1.03). Those with upper tract tumors had a median OS of 23.3 months vs 7.2 months with erdafitinib and chemotherapy, respectively (HR, 0.34; 95% CI, 0.18-0.64).1 In a question-and-answer session, Loriot noted that thedifference noticed in patients with upper tract location vs the bladder may be attributed to mutations and biologic components, but no conclusive data were available at this time.
In terms of safety, in the erdafitinib safety cohort (n = 135), 45.9% of patients had at least 1 treatment-related adverse effect (TRAE) of grade 3 or 4. The most common TRAEs of any-grade or grade 3/4 in this arm were hyperphosphatemia (78.5% and 5.2%), diarrhea (54.8% and 3.0%), stomatitis (45.9% and 8.1%), dry mouth (38.5% and 0%), palmar-plantar erythrodysesthesia (30.4% and 9.6%), and onycholysis (23.0% and 5.9%).
In the chemotherapy arm (n = 112), 46.4% of patients experienced a TRAE of grade 3 or 4. The most common any-grade and grade 3/4 events were anemia (27.7% and 6.3%), alopecia (21.4% and 0%), nausea (19.6% and 1.8%), neutropenia (18.8% and 13.4%), leukopenia (11.6% and 8.0%), and febrile neutropenia (8.0% and 8.9%).1
Treatment discontinuation rates were 8.1% and 13.4% in the erdafitinib and chemotherapy arms, respectively. Serious AEs were reported in 13.3% of patients in the erdafitinib arm and 1 treatment-related death occurred and was cited by investigators as being sudden death. AEs in the erdafitinib arm were mostly manageable with dose modifications and supportive care. In the chemotherapy arm, 24.1% of patients had serious AEs and 6 treatment-related deaths occurred and were because of febrile bone marrow aplasia (n = 2), febrile neutropenia (n = 1), septic shock (n = 2), and atypical pneumonia (n = 1).1
AEs of interest included nail disorders, skin disorders, eye disorders (excluding central serous retinopathy), and central serous retinopathy alone. These occurred as any-grade events in the erdafitinib arm at rates of 66.7%, 54.8%, 42.2%, and 17.0%, respectively. Grade 3/4 events were reported in 11.1%, 11.9%, 2.2%, and 2.2%, respectively. None of these events were reported at grade 3/4 in the chemotherapy arm but occurred at grade 1/2 as follows: nail disorders (5.4%), skin disorders (12.5%), eye disorders (5.4%).
Petrylak noted that future studies are needed to develop rational sequencing of eradfitnib, enfortumab vedotin-ejfv (Padcev), and sacituzumab govitecan-hziy (Trodelvy) in the third-line setting, but for now decisions should be made based on toxicity.