A multi-centre phase I trial of the PARP inhibitor olaparib in patients with relapsed chronic lymphocytic leukaemia, T-prolymphocytic leukaemia or mantle cell lymphoma

Relapsed chronic lymphocytic leukaemia (CLL), mantle cell lymphoma (MCL) and T-prolymphocytic leukaemia (T-PLL) remain incurable despite the availability of novel agents. Genetic alterations in the ATM-TP53 DNA damage response (DDR) pathway represent an important mechanism of chemoresistance to conventional chemotherapeutic agents and also drive genomic instability.

The ataxia telangiectasia-mutated (ATM) protein plays a critical role in the DNA damage response to double strand breaks (DSBs) (Shiloh & Ziv, 2013). Poly (ADP-ribose) polymerase (PARP) plays a central role in single strand break (SSB) repair and when the activity of this enzyme is inhibited unrepaired SSB lesions are converted into DSBs during DNA replication. Tumour cells deficient in homologous recombination repair (HRR) proteins, such as BRCA1, BRCA2 or ATM, may develop lethal amounts of DNA damage when treated with PARP inhibitors. We have demonstrated the efficacy of PARP inhibition on the growth of ATM-defective CLL and MCL in vitro (Weston et al, 2010). Olaparib (Lynparza™; AstraZeneca, Cambridge, UK) is an oral PARP inhibitor licensed as monotherapy for the maintenance treatment of adult patients with platinum-sensitive relapsed BRCA-mutated (germline and/or somatic) high-grade serous epithelial ovarian, fallopian tube, or primary peritoneal cancer (Tutt et al, 2010; Kaufman et al, 2015; Mateo et al, 2015; Ledermann et al, 2016). Olaparib is well tolerated and demonstrates significant activity in combination with chemotherapy (Bang et al, 2015) but myelosuppression may be a potentially limiting factor.

We report the results of a phase I conventional dose escalation trial using a cumulative 3 + 3 design to assess safety and maximum tolerated dose (MTD) of the PARP-inhibitor olaparib in patients with relapsed CLL, T-PLL or MCL unsuitable for further conventional treatment. The initial 3 cohorts (9 patients) received the original capsule formulation of olaparib (Figure S1). During the trial AstraZeneca developed a tablet formulation to improve drug loading, bioavailability and reduce the number of tablets. Consequently, a further 2 cohorts received the new tablet formulation (6 patients). A total of 15 patients with relapsed CLL (n = 9), MCL (n = 4) or T-PLL (n = 2) were treated (Table S1). The median age of patients was 69 years (range 53–77) and the median number of previous lines of therapy was 3 (range 1–7). The median duration of olaparib treatment was 71 days with an interquartile range of 26–93 days (Table 1). Myelosuppression was the most common haematological grade 3–4 toxicity and was seen in eight patients. Overall, both formulations of olaparib were generally well tolerated with the most common adverse events (AEs) being anaemia (66%), thrombocytopenia (53%), fatigue (53%), nausea (33%) and neutropenia (33%). Grade ≥3 AEs were seen in 10 patients (66%), [anaemia (33%), thrombocytopenia (33%), neutropenia (20%)]. Of the 6 patients dosed at 200 mg bd (capsule), 3 patients experienced grade ≥3 AEs and all 3 patients who were dosed at 400 mg bd (capsule) experienced at least 1 grade ≥3 AE. For the tablet formulation of olaparib, 4 of the 6 patients dosed at 100 mg bd experienced grade ≥3 AEs (Table 2). As regards the development of dose-limiting toxicities (DLTs), 1 out of 6 patients receiving olaparib 200 mg bd capsules developed a DLT (grade 4 thrombocytopenia). All three patients who received the higher dose of 400 mg bd capsules developed DLTs that were possibly attributable to olaparib within 8 weeks of treatment initiation (Grade 3 maculo-papular rash, grade 4 anorexia/weight loss and grade 4 thrombocytopenia). The MTD for olaparib capsules was therefore defined as 200 mg bd using the 3 + 3 dose-escalation design. The tablet formulation of olaparib was introduced at a treatment dose of 100 mg bd and was administered to six patients. One patient from the initial cohort developed a fatal DLT which presented as an infective episode, renal failure (acute kidney injury) and bleeding with a high International Normalised Ratio on warfarin. No DLT was experienced in the subsequent cohort but one patient was not evaluable due to early disease progression. Unfortunately, recruitment ceased after this cohort (mainly due to the availability of Bruton tyrosine kinase inhibitor trials) and we were therefore unable to define an MTD for the tablet formulation. The tablet formulation dose of 300 mg bd is now used in most studies for monotherapy in solid tumours (Mateo et al, 2016). The median overall survival (OS) from the start of treatment for all 15 patients (9 deaths in trial period) was 129 days (Figure S2A). The median OS for patients treated with capsules (106 days) was not dissimilar to that for patients treated with tablets (129 days) (Figure S2B).

Specific primers for targeted deep sequencing of ATM (exons 4–65), SF3B1 (exons 13–16), TP53 (exons 4–10), BIRC3 (exons 2–9) and MYD88 (exon 5) were designed with the D3 Assay Design web-based tool (https://www.fluidigm.com/assays). Twelve patients (80%) had evidence of a mutation in at least one of the 6 well-established CLL ‘driver’ genes: ATM, TP53, BIRC3, SF3B1, NOTCH1 and MYD88. (Figure S3A). A further patient, Patient 13, presented with monoallelic ATM loss due to an 11q deletion (Figure S3A, Table 1). SF3B1 gene alterations have similar functional consequences to that of ATM loss (Te Raa et al, 2015), justifying our strategy to observe ATM and SF3B1 mutant tumours as a single group of nine patients (60%) (Figure S3A, Table 1). Duration of olaparib treatment ranged from 8 to 133 days with a median of 83 days in patients whose tumours harboured mutations within ATM or SF3B1 (‘mutated’) compared to 37·5 days in those lacking such alterations (‘unmutated’) (Figure S3B). Although not significant, a longer median survival time of 192 days was also seen in patients with a ‘mutated’ genotype compared to 89 days in the ‘unmutated’ group (Figure S3C). Therefore, aberrations in the ATM pathway may be associated with improved responses and OS with PARP inhibitor treatments even among heavily pre-treated and relapsed patients with CLL, MCL and T-PLL. Future studies would be needed to better define the optimal dosage in haematological tumours but this early data suggest that olaparib could have potential clinical utility in patients with a defective ATM pathway.

Author contributions

G.P., C.Y. and T.S. designed the research and G.P. was Chief Investigator for the trial. C.Y and D.S. conducted the statistical analyses. M.G., M.J.S.D., C.F., D.O., A.P., E.M., S.D., D.A, A.B., P.H., G.F, and S.R. were Principal Investigators for the trial. G.P., C.Y., C.O., D.S. and T.S. interpreted the results. G.P., C.Y., D.S., C.O., R.B., P.M. and T.S. were involved in writing the draft manuscript and approved the final manuscript.

Competing interests

The authors have no competing interests.