CHICAGO – Adding the kinase inhibitor ruxolitinib to capecitabine improved overall survival of a subset of patients with metastatic pancreatic cancer, compared with capecitabine and placebo.
In a randomized phase II trial in 127 patients with metastatic pancreatic ductal adenocarcinoma, the second-line therapy combination of ruxolitinib (Jakafi) and capecitabine (Xeloda) was not associated with better outcomes than capecitabine and placebo in the overall population.
But among 60 patients with high levels of C-reactive protein indicative of systemic inflammation, adding ruxolitinib improved median overall survival to 83 days, compared with 55 days for controls, reported Dr. Herbert I. Hurwitz, professor of medicine in the division of oncology at Duke University, Durham, N.C.
Dr. Herbert Hurwitz
"These data support the role of inflammation and the JAK-STAT pathway in particular for patients with pancreatic cancer," Dr. Hurwitz said at the annual meeting of the American Society of Clinical Oncology.
Ruxolitinib is an inhibitor of the Janus kinases (JAK) 1 and 2, which have been shown to mediate cytokine signaling through activation of STAT transcription factors. In preclinical models, pro-inflammatory cytokine signaling has been shown to contribute to pancreatic cancer initiation and progression.
Systemic inflammation is also associated with weight loss, decreased muscle mass, and poor performance status, all of which can shorten survival in patients with advanced pancreatic cancer.
The investigators hypothesized that ruxolitinib could improve survival of advanced pancreatic cancer patients when added to standard chemotherapy by dampening inflammation and thereby reducing cachexia and its related effects on patients’ overall health.
In tumor xenograft models of pancreatic cancer, the combination of ruxolitinib and capecitabine showed antitumor activity, prompting investigators to evaluate the drugs in human clinical trials.
Recap of RECAP
The phase II RECAP study (A Randomized Phase II Study of Ruxolitinib Efficacy and Safety in Combination With Capecitabine for Subjects With Recurrent or Treatment Refractory Metastatic Pancreatic Cancer) enrolled 127 patients with histologically confirmed metastatic pancreatic ductal adenocarcinoma and a Karnofsky performance score of 60 or greater who had disease progression on gemcitabine (Gemzar).
The patients were randomized to capecitabine 1,000 mg/m2 twice daily for 14 days, plus either ruxolitinib 15 mg twice daily for 21 days, or placebo.
The median overall survival was 136.5 days for patients treated with the combination, and 129.5 days for those treated with capecitabine/placebo. The hazard ratio (HR) was 0.79, but the difference was not statistically significant (P = .25, above the prespecified 2-sided P value of .20).
However, when the authors performed the prespecified analysis of patients with C-reactive protein (CRP) levels above 13 mg/L, they saw a significant difference: 83.0 days for patients on the combination, vs. 55.0 days for capecitabine/placebo (HR, 0.47; 2-sided P = .01).
The 6-month overall survival rate for patients treated with the combination was 42%, compared with 11% for patients treated with capecitabine/placebo.
In a multivariate analysis of patients with high CRP levels – controlling for age, serum lactate dehydrogenase and albumin levels, liver and lung metastases, performance score, prior erlotinib (Tarceva), prior radiation or surgery, and sex – they found that the association of the combination therapy with better overall survival remained (adjusted HR, 0.50; 2-sided P = .037).
In addition, an analysis of survival according to modified Glasgow Prognostic Score (mGPS), a combination of CRP and albumin measures, showed that patients with a higher score indicative of higher degrees of systemic inflammation had better survival when treated with ruxolitinib than with placebo.
The median progression-free survival (PFS), a secondary endpoint, was in the overall population (intention to treat) 51.0 days for combination-treated patients, compared with 46.9 days for those treated with capecitabine/placebo (HR, 0.75; 2-sided P = .14). In the prespecified population with a high CRP level, the respective PFS was 48.0, vs. 41.5 days (HR, 0.62; 2-sided P = .10).
Grade 3 or 4 adverse events occurred in 74.6% of patients on the combination and 81.7% of those on capecitabine/placebo. More patients discontinued the study drug because of adverse events in the placebo group (12 patients, vs. 7 in the ruxolitinib group).
Adverse events were generally lower in frequency among patients treated with the combination, except for pulmonary embolism (a common side effect of JAK/STAT inhibitors), which occurred in seven patients on the combination, compared with three on placebo, and anemia (9 vs. 1 patient, respectively).
Invited discussant Dr. Andrew Ko of the Helen Diller Family Comprehensive Cancer Center at the University of California, San Francisco, applauded the investigators for their "smart trial design and drug development strategy."
He said that the preplanned analysis of a subgroup of interest allowed investigators to better plan for patient selection in a phase III trial, called JANUS.