Pulmonary toxicity associated with increased 6-OH-paclitaxel exposure in pancreatic cancer patients treated with nab-paclitaxel, gemcitabine and cobicistat: Results of the IntenSify trial

Background: CYP3A5 expression is a basal and acquired mechanism of resistance of pancreatic ductal adenocarcinoma cells. Inhibition or knockdown of CYP3A5 restores the cells' sensitivity to paclitaxel in vitro . Hence, the addition of a CYP3A5 inhibitor to a paclitaxel-based chemotherapy regimen may re-sensitize resistant clones and confer clinical benefit. Since systemic paclitaxel exposure is determined by CYP3A and CYP2C8 activity, we anticipated the necessity to adapt the dose to account for a pharmacokinetic drug drug interaction. Methods: We conducted a single-center phase I 3+3 dose escalation trial exploring the safety, tolerability and pharmacokinetics of the novel combination of nab-paclitaxel, gemcitabine, and the CYP inhibitor cobicistat in patients with metastatic pancreatic cancer (NCT05494866). We planned with three dose levels (DL) of nab-paclitaxel (DL1: 75 m/m², DL2: 100 mg/m², DL3: 125 mg/m² d1, d8, d15, Q4W i.v.). Gemcitabine (1000 mg/m² d1, d8, d15, Q4W i.v.) and cobicistat doses (continuous 150 mg p.o. qd starting with C1D11) were the same in all DL. We collected serial plasma samples on day 1 and day 15 of cycle 1 to measure drug and metabolite plasma concentrations through UPLC MS/MS. We measured radiological response to treatment every 6-12 weeks according to RECIST 1.1. Patients were followed up for overall survival (OS). Results: We enrolled six patients in DL1. One patient was hospitalized after 6 weeks of treatment for grade 4 acute respiratory distress syndrome associated with LDH increase and lung ground-glass opacity in the CT scan. Two more asymptomatic cases of ground glass opacity associated with LDH increase were observed. We terminated the trial for safety concerns, even though none of the events formally fulfilled DLT criteria. The pharmacokinetic profile showed a minor increase in paclitaxel exposure (AUC-ratio [90% confidence interval]: 1.24 [1.06-1.45]), a moderate increase of 6-OH-paclitaxel (AUC ratio: 2.47 [1.24-4.92]), and a minor increase in gemcitabine AUC). Median progression free survival was 1.89 months and median OS was 4.54 months. Conclusion : Overall, the combination of gemcitabine and nab-paclitaxel with continuous dosing of cobicistat is not safe at the lowest explored dose. We observed pulmonary toxicity with ground glass opacity and LDH increased associated with ~2.5-fold increased 6-OH-paclitaxel levels. Higher metabolite levels were associated with increased severity of findings. Patient outcome did not warrant continuation of the trial. The results of this trial will inform future strategies for targeting CYP3A5 in pancreatic cancer. Potential approaches may be intermittent CYP3A inhibition, use of a specific CYP3A5 inhibitors, or local target inhibition.