Phase I trial of fludarabine, bortezomib and rituximab for relapsed and refractory indolent and mantle cell non-Hodgkin lymphoma

Study design

This phase I trial was designed to evaluate the maximum tolerated dose (MTD) and dose limiting toxicities (DLT) of bortezomib in combination with rituximab and fludarabine. Secondary objectives included overall response and clinical benefit rates. The protocol was approved by the institutional review board at University Hospitals Case Medical Center and written informed consent was obtained before enrollment.

Patient selection

Patients aged 18 years or older with a confirmed indolent NHL (including grade 1 or 2 follicular lymphoma, marginal zone lymphoma, lymphoplasmacytic lymphoma and small lymphocytic/chronic lymphocytic leukaemia) who had relapsed or had been refractory (defined as no response, or progression within 6 months of completing therapy) to at least one standard therapy were candidates for this study. Also included were patients with relapsed or refractory mantle cell lymphoma, aside from those with the blastic variant. Patients were required to have an Eastern Cooperative Group performance status of 0–2 and to have measurable disease. Baseline laboratory parameters included an absolute neutrophil count (ANC) ≥ 1·5 × 10⁹/l, platelet count ≥ 75 × 10⁹/l, and adequate renal and hepatic function.

Patients were excluded for the following conditions: radiation, chemotherapy, or monoclonal antibody therapy within the previous 4 weeks or radioimmunotherapy within the previous 3 months; previous bortezomib therapy; a history of uncontrolled orthostatic hypotension; grade 2 or higher neuropathy; prior allogeneic transplant; central nervous system lymphoma; serious infection, including human immunodeficiency virus disease; pregnancy or breast feeding.

Treatment schedule

Bortezomib was supplied by the Division of Cancer Treatment and Diagnosis (DCTD), National Cancer Institute (NCI) as a sterile, lyophilized powder in vials with mannitol, which was reconstituted with normal saline to a drug concentration of 1 mg/ml. It was administered by intravenous push over 5 s on days 1, 4, 8 and 11 followed by a 10-d rest period. Fludarabine was handled as directed by the package insert and material safety data sheet and administered at 25 mg/m² intravenously as a 30 min infusion daily for 3 or 5 d of each 21 d cycle. Rituximab was handled as directed by the package insert and material safety data sheet and administered at 375 mg/m² on day 1 of each cycle per a standardized infusion protocol. Treatment was repeated every 3 weeks until disease progression, unacceptable toxicity or to a maximum of eight cycles. Haematopoetic growth factors were not allowed during cycle one and were administered at the treating physician’s discretion thereafter.

Dose escalation is depicted in Table I. Cohorts of 3–6 patients were treated at each dose level. For dose escalation to proceed, three patients had to complete cycle 1 without a DLT. If one DLT was observed, then an additional three patients were accrued, and further escalation occurred if no additional DLTs were seen. If DLT was observed in 2 of 3–6 patients enrolled at a specific dose level during cycle 1, that step was considered the dose-limiting level. A total of six patients were treated at the dose below the DLT level, thus establishing the MTD.

Response and Toxicity Criteria

Toxicities were assessed using the NCI Common Toxicity Criteria version 3.0 (http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/ctcaev3.pdf). DLT was defined as any grade 3 or higher treatment-related non-haematological toxicity and grade 4 neutropenia or thrombocytopenia occurring within the first cycle. Grade 4 lymphopenia and anaemia were not considered DLTs. The MTD was defined as one dose level below the level determined to be the DLT.

Assessment of response was performed after every two cycles of treatment. Response was defined using the International Workshop NHL criteria (Cheson et al, 1999). For patients with chronic lymphocytic leukaemia, the NCI Working Group guidelines were used to determine a response (Cheson et al, 1996). Time to disease progression was calculated as the number of months from the initiation of therapy to the first documentation of disease progression, death regardless of cause or change in therapy, whichever occurred first.

Patient characteristics

Twenty-four patients were enrolled onto the study between July 2003 and July 2008. Patient characteristics are listed in Table II. The mean age was 62 (range, 36–87) years and 13 patients were >60 years old. All patients were assessable for toxicity. Patients received an average of 3·5 cycles (range, 1–8) and 20 patients who received two or more cycles were assessable for clinical response. The mean number of prior chemotherapy regimens was 2 (range, 1–7). Twenty-three patients had previously received rituximab or ibritumomab tiuxetan, eight patients had previously received fludarabine. Twelve patients were refractory to their most recent treatment. Of these, 10 patients were refractory to rituximab or ibritumomab tiuxetan and 2 to fludarabine. Five patients had failed an autologous stem cell transplant.

Toxicity

Haematological and non-haematological toxicities are depicted in Tables III and IV. No DLTs were observed on dose steps 1–4. With the addition of rituximab at dose level 5, two patients experienced DLTs including neutropenia and thrombocytopenia in one patient as well as neutropenia and neuropathy in another. After independent review and consultation with the Cancer Therapy Evaluation Program, dose steps 5a and 5b were added in which doses of fludarabine and bortezomib were attenuated. Nine additional patients were treated, 3 at level 5a and 6 at 5b. One additional DLT, grade 3 dyspnea, was observed at dose 5b.

Myelosuppression was the most common toxicity, with neutropenia being dose limiting. Grade 3/4 neutropenia and thrombocytopenia were observed in 24% and 17% of cycles, respectively, and appeared to be cumulative in nature. Six patients received filgrastim or pegylated filgrastim support after demonstrating prolonged neutropenia during the first cycle of therapy, four of whom were treated on rituximab-containing dose levels. Five patients were removed from the study due to prolonged myelosuppression, preventing initiation of the subsequent cycle within the protocol-specified 5-week time period. As expected, all patients developed grade 3/4 lymphopenia.

The major non-haematological toxicities were neurological and infectious in aetiology. The peripheral neuropathy was predominately sensory in nature, more prone to develop in patients with previous sensory neuropathies and also appeared to be cumulative. Grade 2 or 3 neuropathy developed in nine patients, six of which had received prior vincristine. This contributed to the decision to discontinue protocol treatment in six cases. It was partially reversible with dose reduction or discontinuation of therapy. One patient treated at dose level 5 experienced an increase in his pre-existing grade 1 peripheral neuropathy to grade 3 during cycle 1. This decreased to baseline with a dose reduction of bortezomib. Two additional patients experienced grade 3 orthostatic hypotension possibly due to drug-induced autonomic neuropathy. Both patients’ symptoms resolved with therapy cessation.

The overall incidence of grade 3 infection was 6% (5 of 86 cycles). These included one oropharyngeal infection, one cellulitis, one periodontal abscess and two episodes of culture-negative neutropenic fever. In two instances, patients were removed from the trial following a >5-week period without initiation of the next course of treatment. Other grade 3 toxicites included grade 3 dyspnea potentially related to bortezomib in a patient treated at dose level 5b, grade 3 diarrhoea in two patients, and one episode each of fatigue, hypoglycaemia, and an infusion-related rituximab reaction. One patient experienced grade 3 pain related to progressive disease, not believed to be neurological in aetiology. The only grade 4 non-haematological toxicities included five episodes of hyperglycaemia, all occurring in a single patient with diabetes mellitus after receiving glucocorticoids as part of her anti-emetic regime.

Dose modifications or delays in therapy occurred in 34 of the 86 cycles (40%). Bortezomib was dose reduced in 16 cycles and individual doses were held 26 times for a delivered dose intensity of 88%. In addition, there were a total of 17 weekly delays in therapy. The most common reasons for interruptions in therapy were neutropenia or thrombocytopenia (55% of cases) as well as peripheral neuropathy (32%).

Clinical responses

Objective responses were observed in 11 patients (Tables V and VI) for an overall response rate of 45% (95% confidence interval [CI]: 0·28 – 0·65) with a clinical benefit rate (complete remissions + partial remissions + stable disease) of 71% (17/24) with 95% CI: 0·51–0·86. Two of the three patients achieving a complete response have no signs of relapsed disease at 1 and 3 years, respectively, despite both having been heavily pre-treated including receiving high-dose chemotherapy with autologous stem cell rescue. Six of the responding patients went on to receive additional therapy. One patient underwent high-dose chemotherapy and autologous stem transplantation and another patient received an umbilical cord blood transplant. Both remain in remission for >5 years. Two patients received additional immunochemotherapy, including fludarabine + rituximab and rituximab + EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin), respectively, and two patients received further rituximab given in maintenance fashion. Six patients were classified as having stable disease with a median time of 10 months (range 4–30+). Of the 12 patients with disease refractory to their most recent regimen, one complete response, four partial responses, and three patients with stable disease (of 4, 5 and 6 months, respectively) were observed.