BY ROBERT H. CARLSON
BARCELONA, Spain – Three abstracts selected for presentation in a session on pancreatic cancer here at the European Society for Medical Oncology (ESMO) World Congress on Gastrointestinal Cancer each featured something unusual – positive news.
“That isn't something we hear often in pancreatic cancer,” said the Discussant for all three papers, Jordan Berlin, MD, Professor of Cancer Research and Clinical Director of the Gastrointestinal Oncology Program at Vanderbilt University Medical Center.
That may have stretched the point a bit, as two of the trials were not positive in their primary endpoints, but he found value in those anyway.
The trials were:
· Napoli-1, a positive trial of MM-398, a liposomal irinotecan, in second-line therapy;
· the RECAP trial, in which ruxolitinib did not improve overall survival but which Berlin said had a “possible positive twist” in showing an association between inflammation and outcome; and
· a biomarker analysis of the MPACT trial, negative for finding SPARC prognostic, although Berlin said it was still a positive trial in that it identified performance status as a predictive marker.
Nanotherapeutic Activity in Napoli-1
MM-398 is a “nanotherapeutic,” consisting of irinotecan encapsulated in a liposomal sphere. Andrea Wang-Gillam, MD, PhD, Assistant Professor in the Division of Oncology, Section of Medical Oncology, at Washington University in St. Louis, who presented the results of the randomized Phase III Napoli-1 trial, explained that this delivery system allows for longer drug exposure in the circulation and more accumulation of the drug and its active metabolite, SN38, at the tumor site.
First author was Daniel Von Hoff, MD, Physician-in-Chief and Director of Translational Research at TGen, Translational Genomics Research Institute, in Phoenix. Napoli-1 compared the use of MM-398 (151 patients) versus fluorouracil/leucovorin (149 patients) versus MM-398 plus 5-FU/leucovorin (117 patients). All 417 patients had metastatic pancreatic cancer and had received prior gemcitabine-based chemotherapy.
The primary endpoint of overall survival was met for the MM-398-5FU/leucovorin combination, although not for single-agent MM-398.
Median survival times were 6.1 months for the combination, 4.9 months for MM-398 alone, and 4.2 months for 5FU/leucovorin.
“Median survivals are important here,” Berlin said in his discussion. He compared the median survival of patients in Napoli-1 with those in two Phase II trials with FOLFIRI (leucovorin, 5-FU, and irinotecan) as second-line treatment for metastatic pancreatic cancer: GISCAD had an 8% rate and a median survival of 5.0 months (Zaniboni A et al: Cancer Chemother Pharmacol 2012; 69:1641-1645); and a French trial in patients resistant to gemcitabine had a median survival of 6.6 months (Neuzillet C et al: World J Gastroenterol 2012; 18:4533-4541).
He cautioned that comparing uncontrolled Phase II trials with a Phase III trial has to be done with caution, because there tend to be decreases in efficacy going from Phase II to III--“as we all know.”
Nevertheless, he said Napoli-1 is certainly a positive trial and a viable option, presuming regulatory approval.
“But oxaliplatin-5FU looks about as good,” he said. “There is clear benefit to the patients [in Napoli-1], but it is not clear if this is better than what we already have with FOLFIRI or FOLFOX.
“In the long run, we have yet to find something other than chemotherapy to treat this disease.”
Combination therapy led to more gastrointestinal side effects than standard treatment alone in Napoli-1. Diarrhea occurred in 12.8, 21.1, and 4.5 percent of the combination trial arm, MM-398-alone arm, and 5FU/leucovorin arm, respectively. Vomiting occurred in 11.1, 13.6, and 3.0 percent, respectively; and fatigue occurred in 13.7, 6.1, and 3.7 percent, respectively.
Side effects may have been “manageable” in Napoli-1, Berlin said, but he added that “manageable does not mean tolerable; it means you have to change the regimen or dose modify when you get some serious side effects--The definitions of Grade 3 fatigue or Grade 3 diarrhea all sound pretty bad to me.”
Ruxolitinib: Inflammation Influences Outcomes
In the placebo-controlled randomized Phase II RECAP study of capecitabine with or without ruxolitinib as second-line therapy for metastatic disease, the ruxolitinib combination showed clinical activity compared with capecitabine-placebo. Response rates were 7.8 percent for the combination versus zero percent for placebo.
Overall survival and progression-free survival “favored the combination therapy, but the effect size was small,” said Herbert I. Hurwitz, MD, Professor of Medicine at Duke University, who presented the data.
Importantly, though, evidence of overall survival and progression-free survival was observed in a subgroup of patients with elevated levels of the systemic inflammation marker C-reactive protein (greater than the group median of 13 mg/L). Overall survival rates at three and six months were 48 and 42 percent, respectively for the gemcitabine-ruxolitinib arm (31 patients), versus 29 and 11 percent with gemcitabine-placebo (29 patients).
Ruxolitinib, a JAK1/JAK2 inhibitor, inhibits STAT through JAK and also reduces inflammatory cytokines, a predictor of survival in pancreatic cancer, Berlin explained, noting that there have been consistent preclinical data showing that blocking the STAT pathway results in growth inhibition in pancreatic cancer.
The hypothesis around inflammation is a key point and well worth investigating further, he said. “The science that ruxolitinib reduces inflammation may have exposed a weakness in pancreas cancer. If we were to reduce inflammation, cancer-cachexia, or whatever kills the patient, it might prolong life and it might give us more of a chance to treat the cancer itself.”
SPARC Is Poor Prognostic Indicator
The Phase III MPACT trial (NEJM 2013; 369:1691-1703) with 861 patients showed superior overall survival rates for use of nab-paclitaxel plus gemcitabine versus gemcitabine alone as first-line treatment for patients with metastatic pancreatic cancer.
At this meeting, Manuel Hidalgo, MD, Vice Director of Translational Research at Centro Nacional de Investigaciones Oncológicas in Madrid, presented an analysis of SPARC (secreted protein acidic and rich in cysteine) expression in MPACT patients to determine if it might be a prognostic factor. Daniel Von Hoff was senior author.
Although high expression of SPARC has been associated with worse overall survival rates in patients with resectable pancreatic cancer, Hidalgo reported here that stromal SPARC was not prognostic of overall survival in MPACT, with a hazard ratio of 1.02 between patients with high- and low-SPARC scores.
Three significant predictors of overall survival were identified: treatment, Karnofsky performance status, and the presence of liver metastases.
Hidalgo concluded that while “nab-paclitaxel is currently a new standard of care that has demonstrated a clinical benefit across all patient subgroups,” treatment decisions in the metastatic setting should not be based on SPARC expression.
“Every way Dr. Hidalgo tested it, SPARC was not predictive,” Berlin said in his discussion. “So we do not have a biomarker for gemcitabine and nab-paclitaxel, and we can probably stop testing SPARC in advanced disease.”
But based on the CONKO-001 adjuvant therapy study, where stromal SPARC was prognostic for gemcitabine-treated patients (Oettle H et al: JAMA 2013;310:1473-1481), it would still be reasonable to test for SPARC in the adjuvant setting, he added.
On the one hand, Berlin said, it probably isn't necessary to continue testing SPARC in metastatic pancreatic cancer “because pancreatic cancer is a disease where we don't need to seek prognostic markers for poor outcomes.”
The money could be better spent elsewhere. “On the other hand, a predictive marker would be valuable, which is a reason to test further.”
‘Use Resources Wisely’
Berlin said there had been very few prospective collections of SPARC until MPACT, and while Hidalgo's report was negative in its primary endpoint, the study was valuable in that it collected “a large number of precious samples from patients with metastatic pancreatic cancer, which is difficult to do.
“A lot of people gave a lot to get this data,” Berlin concluded, speaking to all oncology researchers. “It is our responsibility not only to report it, but to use it wisely.”