Paper Details

ABSTRACT

Background: Hairy cell leukemia (HCL) is a B-cell malignancy presenting with cytopenias and splenomegaly. Malignant cells strongly express B-cell antigens and also CD25, CD103, CD125 CD11c, and are driven by the BRAF V600E mutation. While HCL responds well to purine analogs, variant HCL (HCLv), which lacks CD25, CD125 and BRAF mutation, responds poorly to purine analogs even in 1^st line and overall survival is limited. Patients with unmutated (um) IGHV4-34+ HCL are clinically like HCLv even if phenotypically consistent with classic HCL. Inhibitors targeting BRAF +/- MEK are highly active in HCL but less effective after patients relapse or progress on treatment, and those with HCLv are ineligible. To target BCL-2 in HCL, venetoclax was tested and complete remissions (CRs) were reported in 2 of 6 patients in a small study, in addition to a case report of one CR. To determine the potential for venetoclax in targeting BCL-2 in both HCL and HCLv, we performed a phase 2 trial including patients with either diagnosis.

Patients and Methods: A phase 2 trial of venetoclax was planned for 17 patients. The primary endpoint was by 32 weeks to exceed the 24.3% overall response rate (ORR) reported for the BTK inhibitor ibrutinib, specifically to rule out an unacceptably low ORR of 15% in favor of an improved ORR of 40%, with the goal to achieve at least 5 (29.4%) of 17 responses. To be eligible, patients required both prior purine analog and BRAF therapy (if V600E+) and need for more treatment. To avoid tumor lysis syndrome (TLS), venetoclax began at 20 mg/day orally and increased weekly to 50, 100, 200, and 400, so 400 mg would begin C2D1, the first day of the 2^nd 28-day cycle. Patients were restaged including bone marrow and imaging at 12 (C4D1) and 32 (C9D1) weeks, and venetoclax was continued through 19 cycles or until progression/toxicity. After 19 cycles, patients could choose to continue venetoclax off protocol. CR required absence of morphologic evidence of HCL/HCLv in bone marrow, as well as resolution of HCL/HCLv-related cytopenias. MRD was assessed by bone marrow biopsy immunohistochemistry (IHC) and flow cytometry of bone marrow aspirate and blood flow cytometry; MRD-free meant negative by all tests. Partial response required the same as CR except for only >50% improvement in HCL/HCLv marrow infiltration.

Results: Sixteen of 17 allowable patients were enrolled, including 15 males and 1 female, with 13 HCL and 3 HCLv. Of the 13 HCL, 2 had umIGHV4-34 with MAP2K1 mutations and 2 had non-V600E BRAF mutations. Of the 16 total patients, 7 were not yet evaluable for response, but of 9 evaluable patients, 5 (56%) achieved CR, meeting the primary endpoint. All 3 patients with HCLv had CR, including 1 with a 43x71 mm celiac HCLv mass who became MRD-free, and 2 with TP53 mutations and 22.7-23.7 cm spleens associated with baseline transfusion dependence. The 3 HCLv patients each achieved CR by the C4D1 restaging. The other 2 patients with HCL achieving CR, one with BRAF V600E+ HCL and the other with N486_P490 BRAF deletion, had each progressed after responding to BRAF-MEK therapy, and had baseline grade 4 neutropenia and thrombocytopenia. While the patient with N486_P490 BRAF deletion also responded early, the patient with BRAF V600E+ HCL achieved CR later by C9D1, with marrow HCL infiltration by IHC decreasing from 90% at C1D1 to 60% at C4D1, 13% at C9D1, and 2% at C20D1. No responding patients have relapsed, with relapse-free survival 5-20 (median 11) months. Venetoclax-related adverse events documented in >1 of 13 evaluable patients included (grade 1-4, grade 3-4) neutropenia (38%, 23%), leukopenia (38%, 15%), lymphopenia (23%, 23%), and thrombocytopenia (15%, 8%). One patient died of disseminated fungal infection unrelated to venetoclax 5 months after stopping venetoclax and 4 months after initiating BRAF therapy. We found that dose reductions and interruptions were not required when sufficient filgrastim was administered for neutropenia, and eventually cytopenias improved.

Conclusion: Despite limited follow-up, venetoclax is active in phase 2 testing in patients with HCL and HCLv, meeting its primary endpoint. Venetoclax 400 mg/day is well-tolerated without evidence of TLS, and hematologic toxicity can generally be managed without interruptions and dose-reductions. While HCL responses are achieved slowly, it is possible that patients with HCLv and high-risk forms of HCL may respond more quickly.