HTLV-1 Proviral Load after Two Months' Treatment with Anti-CCR4 Monoclonal Antibody Mogamulizumab Predicts a Molecular Response to Disease and Durable Clinical Remission in Leukaemic Subtypes of Adult T-Cell Leukaemia/Lymphoma

Background: Adult T-cell leukaemia/lymphoma (ATL) is an aggressive T-cell malignancy caused by HTLV-1 infection. Typically patients treated with chemotherapy show transient remissions followed by early relapse. There is a need for markers that can predict the response to therapy early in treatment in order to stratify patients who should be considered early for alternative therapies including allogeneic transplantation. The 0761-009 study was a multi-centre, open-label, randomised study of anti-CCR4 monoclonal antibody mogamulizumab (KW-0761) or investigator's choice of chemotherapy in subjects with previously treated, relapsed or refractory ATL. Here we report longitudinal data on proviral load (PVL) and clonal abundance from 4 subjects treated in the UK with leukaemic-type disease (3 chronic, 1 acute subtype).

Methods: qPCR was undertaken to measure PVL and high-throughput sequencing (HTS) to quantify the abundance of the malignant clone, using our previously reported protocols.

Results: Two patients with chronic ATL were randomised to receive mogamulizumab and two patients (one chronic, one acute subtype) to receive chemotherapy (gemcitabine/oxaliplatin). The chemotherapy-treated patients both progressed early and therefore crossed over to mogamulizumab treatment. The three patients with chronic subtype ATL achieved a haematological remission before the end of cycle 1 (28 days therapy), and the patient with acute subtype achieved remission by the end of cycle 2. The median PVL of the 4 patients was 61.4% on trial entry, rising to 88.1% prior to first mogamulizumab infusion. By the end of cycle 2, the median PVL fell to 0.43% (range 0.11 - 28%). The median number of mogamulizumab cycles was 16 (range 2-30). All 3 patients with chronic ATL achieved a PVL <1% prior to the end of cycle 2 and were discontinued due to toxicity: one patient discontinued after cycle 2 (pneumonitis) but remained in clinical remission for 14 months off all treatment. One patient discontinued treatment after 16 cycles with a skin rash, initially considered to be cutaneous ATL, whic was treated topically with steroids, and remains in leukaemic remission two years off all treatment. Finally, one patient discontinued in July 2016 after 31 cycles of treatment, due to skin hypersensitivity, and remains in remission. The patient with acute subtype showed a normalisation of the lymphocyte count within a few weeks and had a normal PETCT scan, but did not show a significant reduction in PVL (44% baseline, 41% at end of Cycle 2). This patient continued in leukaemic remission for 16 months but developed an intraparotid localised nodal progression whilst on mogamulizumab treatment. HTS showed that long-term clinical responses were also associated with a 4-5 log reduction in the absolute abundance of the malignant clone.

Conclusions: Three patients with unfavorable chronic type of ATL achieved durable clinical remissions following treatment with mogamulizumab, and first experienced rapid normalisation of the lymphocyte count (2-4 weeks) followed by a ~ 2 log reduction in the PVL (to <1%) before the end of the 2nd cycle. In the chronic patients who tolerated treatment for long enough to achieve 4-5 log reductions in the absolute abundance of the malignant clone, durable remissions were maintained off

therapy. HTS can be used to quantify molecular responses to treatment and in this small group of patients achieving a proviral load <1% and a 4-5 log reduction in the absolute abundance of the malignant clone was associated with long-term clinical remissions.