Which patients with resectable pancreatic cancer truly benefit from oncological resection: is it destiny or biology?

Abstract

Pancreatic cancer has a dismal prognosis. A technically perfect surgical operation may still not provide a survival advantage for patients with technically resectable pancreatic cancer. Appropriate selection of patients for surgical resections is an imminent issue. Recent studies have provided an important clue on what serum biomarkers may be used to select out the patients who would unlikely benefit from the surgical resection.

Keywords:

• pancreatic cancer
• surgery
• serum biomarker

Pancreatic cancer, a devastating disease with incidence rate nearly equaled its mortality rate, ranks fourth among the most common deaths due to cancer in the United States.¹ It occurs frequently worldwide as well and is a one of the most common causes of cancer-associated mortality in developed countries. Unfortunately, the treatment strategies against this dismal disease provide miniscule improvements in survival, with an average survival of 5% of patients at 5 y after diagnosis of pancreatic adenocarcinoma.²

Among the preferred treatment options, including surgery, chemotherapy, radiotherapy, or an integrated multidisciplinary treatment, performing an oncological surgical resection of the primary tumor in its early stages (stage I or II) has been considered the only means of obtaining a long-term survival. For example, patients who underwent a “true” R0 resection exhibited a significantly longer median survival (35 months) compared with patients with a close surgical margin (16 months). These data indicated an obvious survival benefit from a resection with clean or wide negative margins.³ However, even in patients undergoing a potentially curative resection, recurrence and cancer associated mortality is common. Following the completion of 2 randomized controlled studies for the surgical resection of pancreatic cancer,^4,5 one overarching question emerged: why is the median survival of pancreatic cancer patients who have undergone a radical resection still measured in months? Based on the high rate of metastatic recurrence in these studies, one could conceive that there is a subgroup of patients who do not benefit from a surgical resection as a result of micrometastatic disease. However, preoperative radiographic studies or even the careful intraoperative assessment would not be able to identify this patient subgroup. Therefore, biomarkers that can be used preoperatively to predict the outcome of surgical resection of pancreatic cancer are highly demanded.

Notably, Dr. Xianjun Yu's group at Fudan University Pancreatic Cancer Institute has set the foot to search for these biomarkers preoperatively.^6-8 In the first study, they demonstrated that serum biomarker CA125 has a superior value in predicting the resectability of pancreatic cancer compared with CA19–9 and other tumor markers. This finding was true even for cases whose resectability was misjudged by preoperative CTs. The group suggested that an aberrantly high pre-operative level of CA125 indicates a poor outcome.⁶

In the second study, Xu and co-workers from Dr. Yu's group analyzed the metabolic tumor burden as measured by ¹⁸F-FDG PET/CT using 2 crucial volumetric parameters including metabolic tumor volume (MTV) and total lesion glycolysis (TLG). They used these measurements to predict overall survival (OS) and recurrence-free survival (RFS) of patients with pancreatic ductal adenocarcinoma who underwent radical pancreatectomy, and examined pathological tumor size, baseline serum CA19–9 level, and SUVmax. Remarkably, the results suggested that MTV and TLG are both independent factors for predicting the prognosis of post-operative pancreatic cancer patients. Larger MTV and TLG values were significantly associated with poorer OS and RFS; and both parameters were better predictors of prognosis than serum CA19–9 levels, SUVmax, and pathological tumor size. Therefore, Dr. Yu's group suggested further exploring the values of using MTV and TLG to identify the patients who have a poor prognosis following the surgical resection.⁷

In the third paper, based on their prior findings, Liu and colleagues from Dr. Yu's group conducted a retrospective analysis of ∼1,000 patients with pancreatic cancer, including resected, locally advanced, and metastatic pancreatic adenocarcinoma, to uncover a preoperative profile of serum markers that can predict the outcome following surgical resection.⁸ Dr. Yu's group selected to test 8 serum tumor markers that were most commonly used in gastrointestinal cancers. The study did confirm that the preoperative CA19–9 level was an independent factor for predicting OS by showing that increased preoperative CA19–9 level is associated with poor survival in patients with resected pancreatic cancer. More importantly, the study validated that CEA and CA125 in preoperative serum were also prognostic factors, respectively, for resectable pancreatic cancer. Therefore, they tested the prognostic value of the combination of elevated CEA, elevated CA125 and CA19–9 ≥ 1 ,000 U/mL in preoperative sera. They found that the presence of a preoperative serum CEA + /CA125 + /CA19–9 ≥ 1 ,000 U/mL signature may be an indicator of the existence of micrometastases at the time of surgical resection and contribute to the poor outcomes following surgical resection as patients with this signature exhibited significantly higher rates of DPC4 loss and S100A2 overexpression in their resected pancreatic cancer. Therefore, Dr. Yu's group urged the use of appropriate biomarkers to preoperatively select out patients who would not benefit from the surgery.

Obviously, patients who demonstrated the CEA+/CA125+/CA19–9≥1,000 U/mL signature in their preoperative sera and who had a median survival of 5.1 to 7.0 months⁸ did not benefit from pancreatectomy. Moreover, with such short survival periods, these patients had already spent a minimum of 2 months to completely recover from the surgery. On another hand, they may benefit from upfront systemic chemotherapy in its contemporary forms, which have been shown to prolong the median survival of metastatic pancreatic cancer to 8.5 or 11.1 months.^9,10 Nevertheless, although increasing attention has been paid to the use of preoperative chemotherapy, there is a lack of appropriate strategies to select the patients with a technically resectable disease for neoadjuvant chemotherapy. Thus, Dr. Yu's group proposed to further explore the value of using the CEA+/CA125+/CA19–9≥1 ,000 U/mL signature to select the patients for the upfront surgery in prospective studies.

For resectable pancreatic cancer, achieving a negative margin, performing a rational lymphadenectomy, and making every effort to increase the safety of the surgical procedure are all crucial.^11-13 However, a technically perfect surgical operation may still not provide a survival advantage for patients with technically resectable tumors. Therefore, appropriate selection of patients for surgical resections is an imminent issue. The studies that were published by Dr. Yu's group have significantly contributed to the identification of prognostic indicators preoperatively to predict the outcome of surgical resection of pancreatic cancer and also provided an important clue on what serum biomarkers may be used to select out the patients who would unlikely benefit from the surgical resection. Appropriately individualizing patients for different treatment modalities such as upfront surgery or neoadjuvant therapy followed by surgery will potentially maximize the survival benefits of each treatment modality, and will thus exert a profound influence on the future pancreatic cancer patient care.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.