In patients with resected pancreatic cancer, adjuvant cisplatin, 5-FU, and interferon chemoradiation produces a median survival of 27 months, according to initial results of the ACOSOG Z05031 trial. However, nearly all patients experience grade 3 or 4 toxicities.
In patients with resected pancreatic cancer, adjuvant cisplatin, 5-FU, and interferon chemoradiation produces a median survival of 27 months, according to initial results of the ACOSOG Z05031 trial. However, nearly all patients experience grade 3 or 4 toxicities.
The phase II multicenter trial enrolled patients who had undergone complete resection of pancreatic cancer confined to the head of the gland, said lead author Vincent J. Picozzi, Jr., MD. During cycle one, the patients received conventionally fractionated radiation therapy with 5-FU, cisplatin, and interferon alfa; during cycles two and three, they received 5-FU only.
He noted that nonstandard radiation fields were used. “Both the clinical and planned treatment volumes were designed to be approximately 40% smaller than those conventionally used, such as in RTOG 9704, in an attempt to minimize combined-modality toxicity,” he explained (ASCO 2008 abstract 4505).
The study was stopped prematurely because an interim analysis showed that 96% of patients experienced grade 3 or 4 toxicity, Dr. Picozzi said, noting that most of the events fell in the grade 3 category. Some 30% to 40% of patients each had grade 3-4 anorexia, nausea, and neutropenia; 20% to 30% each had grade 3-4 dehydration, diarrhea, vomiting, and stomatitis; and 12% had grade 3-4 thrombocytopenia. However, there were no deaths due to toxicity, he added.
Of the 89 patients enrolled in the study, 83% had T3 tumors, 73% had nodal involvement, 25% had positive margins, and, of particular note, 46% had poorly differentiated tumors-an unusually high value for this population, according to Dr. Picozzi. Of the 80 evaluable patients, 80% completed chemoradiation therapy and 56% initiated all therapy cycles. The median follow up was 28 months.
Median survival was 27.1 months and the two-year rate of survival was 58%, which compared favorably with values from other trials such as GITSG (21 months, 43%) and RTOG 9704 (20.6 months, 41%). “It is important to note that the ACOSOG trial represents the third reported trial using cisplatin, 5-FU, and interferon chemoradiation in the adjuvant treatment of resected pancreas cancer,” he said, with a total of 174 patients treated and two-year rates of survival consistently exceeding 50%.
Median disease-free survival was 14.1 months, according to Dr. Picozzi. In terms of patterns of recurrence, 47% of patients had a local recurrence and 35% of patients had a distant one. Both margin status and tumor grade significantly predicted overall survival. “Of note was that many prognostic variables commonly reported in other studies, including T stage, tumor size, and nodal status, were not prognostically significant,” he pointed out. The ability to receive therapy may also be prognostic, he noted, and in fact median survival in the subset of patients who initiated all therapy cycles was six months longer than that in the patients overall (33.4 vs 27.1 months).
“This study achieved its initial primary endpoint of a 20-month overall survivorship of greater than 65% in a patient population that could be characterized as pathologically nonfavorable for this disease type and stage,” Dr. Picozzi concluded. “Toxicity with the present regimen was a significant but potentially manageable concern in this multi-institutional setting.” He noted that the investigators are conducting additional analysis and follow-up planning.
Commentary
“My kudos go to Dr. Picozzi and his colleagues because this was a multicenter trial with a reasonable sample size, it required restaging postoperatively, and survival results are encouraging,” said Robert A. Wolff, MD, of M.D. Anderson. However, the regimen had high toxicity, even though most patients were treated at large university centers, and almost half of patients had local failures, similar to that seen with adjuvant chemotherapy alone.
“This is a complex regimen, and I don’t believe it’s easily applied in the community setting, and I wonder if this can accrue patients to a large phase III trial,” he said. Dr. Wolff recommended conducting a trial to try to determine the relative contributions of the chemotherapy and radiation therapy components of the regimen, noting that the CONKO-003 trial has found 5-FU with platinum to be active in gemcitabine (Gemzar)-refractory pancreatic cancer.