MIAMI BEACH – Survival for patients with hilar cholangiocarcinoma was similar between those who underwent transplantation and those who underwent resection, but neoadjuvant therapy may give transplant strategy the edge, findings of a study and meta-analysis suggest.
“Neoadjuvant chemoradiation therapy is clearly a factor that affects patient survival, and may be the only reason the patients who received transplantation had better overall survival than [did those who had] resection,” Michele Gage, MD, a general surgeon at Johns Hopkins Medicine in Baltimore, said at the annual meeting of the Americas Hepato-Pancreato-Biliary Association.
“We found survival in the group that received neoadjuvant chemoradiation therapy and transplant had a statistically significantly better outcome compared to a control group of resection,” Dr. Gage said. More importantly, patients who received a transplant without neoadjuvant therapy had a statistically significant worse outcome than patients who got resection alone.”
The investigators noted that patient selection for neoadjuvant therapy might also be a factor contributing to superior overall survival. In a multicenter study of 147 patients undergoing liver transplantation for hilar cholangiocarcinoma, a subgroup of patients who met the selection criteria of the Mayo Clinic protocol but had not undergone neoadjuvant therapy had a 59% 5-year survival rate ().
Study discussant Maria B. Majella Doyle, MD, a general surgeon at Washington University in St. Louis, agreed that patient selection for transplantation is a likely factor.
Dr. Doyle then asked Dr. Gage how she accounts for the heterogeneity among studies performed over a 20-year period.
“That is why we did subgroup analysis of neoadjuvant versus no neoadjuvant therapy,” Dr. Gage replied.
In the future, an intent-to-treat analysis might be more accurate, Dr. Majella Doyle said, because more patients are placed on a liver transplant list than typically have the procedure.
Dr. Gage noted that 28%-48% of patients started on neoadjuvant therapy in the two studies that offered both neoadjuvant therapy and a transplant in the meta-analysis never made it to transplantation. When they were included, overall survival dropped to approximately 35% in one study and 44% in the other.
In the primary meta-analysis (before the subanalysis looking at neoadjuvant therapy), 398 patients underwent resection and another 200 underwent liver transplantation between 1996 and 2106. Patient demographics were similar between groups, including more men than women, except the average age in the resection group was older, Dr. Gage said.
Overall survival favored the transplant group at each time point: 78% versus 70% with resection at 1 year; 56% versus 42% at 3 years; and 46% versus 29% at 5 years. The odds ratios, respectively, were 1.27, 1.49 and 1.83, but the findings were not statistically significant at a 95% confidence interval.
Margin involvement was 9% in the transplant patients versus 32% in the resection patients, Dr. Gage said. The best chance of cure is R0 resection, but half of patients with hilar cholangiocarcinoma, the most common cancer of the biliary tract, are unresectable, she added.
Six of the nine studies in the meta-analysis reported margin status. Of the 344 patients in these studies, 79% achieved R0 status overall.
“The goal of treatment is R0 resection,” Dr. Gage said in response to a question about when neoadjuvant therapy is warranted. “In the patients who are resectable, I think the correct answer would be to proceed with resection. However, for those patients who are borderline resectable, it would be reasonable to consider neoadjuvant therapy.”
“One of the major things that is undervalued is neoadjuvant therapy allows better patient selection,” said session moderator Eric Jensen, MD, FACS, of University of Minnesota Health in Minneapolis. “When you say an obviously resectable tumor, when you look at the data – we’re wrong 30% of the time. So I’m in favor of neoadjuvant therapy for everybody, but that is just my bias.”
The small number of studies is a limitation of the study, Dr. Gage said. Also, all the studies were nonrandomized and retrospective, and some research spanned many years, which could introduce bias because of changes in practice over time, she added.
Based on their findings, the investigators proposed that future studies explore routine administration of neoadjuvant therapy prior to resection.
Dr. Gage and Dr. Majella Doyle had no relevant financial disclosures.