Borderline and locally advanced pancreatic cancer : redefining the biological and technical profile of the disease

Abstract

Surgery is the only treatment modality that provides a chance for long-term survival in pancreatic cancer (PC). Thus, the current classification systems for PC are technically skewed to predicted the probability for surgical resection, and not adapted for tumor biology. As more potent oncologic therapy steps forward, questions arise whether more aggressive surgery is motivated and how to select the better surgical candidates based of predicted tumor behavior. Also, new tumor-specific treatments should be sought to overcome the aggressive PC biology.

Paper I investigated the short and long-term outcome in a series of pancreatectomy with venous resection (VR). VR can be carried out safely, with low morbidity attributable to the vascular reconstruction itself. No factors were associated with severe morbidity. VRs brought similar survival benefit for resectable, borderline (BRPC) or locally advanced (LAPC) or type of periampullary tumor. Factors pointing shorter survival were attributable to tumor biology and patients’ characteristics (elevated CA19-9 and ASA score) and not technical in nature. Paper II investigated the role of surgery after neoadjuvant treatment (NAT) for BRPC and LAPC. Surgical resection could be carried out safe, despite that vascular procedures were most often required. Surgery significantly improved survival both after FOLFIRINOX and other combination chemotherapy, even for higher levels of preoperative CA19-9. Even significant dose reductions of FOLFIRINOX did not impair the prognosis. There was no difference in survival between BRPC and LAPC patients, whether resected or not, and the recurrence pattern was similar - with distant metastases in all and few local recurrences.

Paper III looked at the impact on survival of biologic prognostic factors potentially available preoperatively (mGPS, CA19-9, para-aortic lymph node, PALN, status) in resected patients with resectable, BRPC, and LAPC. All factors could much better discriminate differences in survival than the resectable, BRPC, and LAPC, including inside each category. Positive PALN had strongest negative impact on survival; their presence was significantly associated with elevated preoperative CA19-9, particularly in LAPC patients after NAT. Paper IV found that tumor-infiltrating lymphocytes (TILs) can be isolated from PC in sufficient amount required for adoptive transfer therapy. TILs showed phenotype that can expand upon stimulation, home, and recognize tumor-associated antigens and autologous tumor cells, and induce autologous tumor-cell killing in culture.

In conclusion, new classification of PC is needed that better reflects the chance for survival. Biological factors should be integrated to successfully guide treatment, even if would still have leading role. Possibilities open to target specific tumor biology by adoptive TIL transfer.