Neoadjuvant Strategies for Pancreatic Cancer
Neoadjuvant Strategies for Pancreatic Cancer
We have made much progress over the past 30 years in the surgical management of pancreatic cancer, and perioperative mortality rates are low in centers with experience in the treatment of this disease. However, surgical resection is clearly limited in achieving local and systemic control of pancreatic cancer, and chemoradiation will likely become a part of any successful pancreatic cancer treatment program.
Dr. Evans and colleagues provide an excellent summary of the current state of neoadjuvant therapy for pancreatic cancer. The authors address the lack of uniformity in surgical technique, resectability definitions, pathologic specimen evaluation, and preoperative assessment of locoregional tumor spread. They emphasize that these concepts must be standardized in order to allow definitive conclusions to be drawn from study protocols.
Dr. Evans et al favor high-quality contrast-enhanced helical computed tomography (CT) to assess pancreatic tumors for resectability and extent of disease. The use of CT scans for the evaluation of pancreatic cancer is well established, and many large centers use CT to ascertain the resectability of pancreatic tumors. With the advent of helical CT scans, the use of other studies such as endoscopic retrograde cholangiography and visceral angiograms has declined. Recently, however, there have been several articles evaluating the role of endoscopic ultrasound in the pretreatment evaluation of pancreatic tumors.[1-6]
Endoscopic Ultrasound vs Computed Tomography
In a retrospective review of 89 patients whose TNM classification was determined by endoscopic ultrasound, Ahmad et al argued that endoscopic ultrasound is inaccurate for the preoperative evaluation of pancreatic cancer. They found that the overall accuracy of endoscopic ultrasound for tumor and node staging was 69% and 54%, respectively, and that only 46% of patients predicted to be resectable by endoscopic ultrasound were actually resectable at the time of exploration. Interestingly, the same group later found that endoscopic ultrasound was associated with advantages over angiography in evaluating vascular invasion by tumors prior to exploration.
On the other hand, several groups favor the use of endoscopic ultrasound.[1,2,5] Gress et al performed CT and endoscopic ultrasound preoperatively on 151 patients, 81 of whom had undergone surgical exploration. They found that the accuracy of endoscopic ultrasound for tumor staging was 92%, 85%, and 93% for T1, T2, and T3 tumors, respectively, vs 65%, 67%, and 38% with CT. Node-staging accuracy with endoscopic ultrasound was 72% for N0 and N1 levels, compared to 52% and 100% with CT. Moreover, CT failed to detect a mass in 26% of patients found to have a tumor at exploration.
Erickson et al also found that endoscopic ultrasound enabled the detection of tumors not seen with CT in 34% of patients. Gress et al concluded that endoscopic ultrasound was 93% accurate in predicting resectability (vs 60% accuracy for CT) and noted that mis-staging by endoscopic ultrasound occurred in the learning phase in seven of nine cases.
Our experience supports the greater accuracy of preoperative endoscopic ultrasound, compared to CT, in evaluating locoregional disease. Unfortunately, endoscopic ultrasound is highly operator-dependent, and many institutions do not have clinicians with the expertise needed to take advantage of its capabilities. For both this reason and the fact that CT enables the diagnosis of liver metastasis, CT remains the study of choice for the evaluation of pancreatic cancer. As experience with endoscopic ultrasound increases, however, it will likely gain prominence in the evaluation of pancreatic cancer patients prior to their enrollment in study protocols.
Other Staging Modalities
Operative staging of tumors felt to be unresectable is also highly variable and dependent on the surgeon’s experience. Several centers have reported on the ability to resect tumors that originally were felt to be unresectable at institutions with limited experience in pancreatic tumor management. It is often impossible to determine whether a tumor is resectable until after it has been fully mobilized. Thus, surgical staging can be flawed, and endoscopic ultrasound offers the possibility of improving the assessment of locoregional disease.
Positron-emission tomography is another modality likely to play an important role in pretreatment assessment of pancreatic cancer. Positron emission tomography has been found to be more sensitive than CT in detecting primary tumors and diagnosing liver lesions thought to represent metastasis without biopsy.
Biliary Obstruction and Preoperative Stenting
Dr. Evans and coworkers discuss the treatment of biliary obstruction in patients undergoing preoperative chemoradiation. They review data that argue against preoperative stenting and note the association of that procedure with sepsis and mortality. However, they also cite data showing that the only morbidity associated with preoperative stenting is wound infection. We feel that preoperative stenting is unnecessary in most patients with biliary obstruction if they are not going to receive neoadjuvant therapy. Nevertheless, we do not believe that it adds significantly to operative mortality and morbidity and agree with its use in neoadjuvant protocols.
Preoperative Therapy Issues
A consensus on the use of postoperative chemoradiation for patients with resectable pancreatic cancer has not been established, but Dr. Evans et al discuss several retrospective and prospective studies that support this approach. Because of the morbidity and prolonged recovery associated with pancreatic resection, they argue that delivery of postoperative therapy is often delayed. Thus, the authors suggest several potential benefits to the use of neoadjuvant therapy: earlier initiation of multimodality therapy, avoidance of unnecessary exploration in patients noted to have metastatic disease at restaging prior to surgery, and a decreased incidence of pancreaticojejunal anastomotic leaks. (This last point may relate to radiation-induced pancreatic fibrosis allowing better suture hold during reconstruction.)
However, they do not comment on the effects of neoadjuvant therapy on other operative characteristics, such as blood loss, operative time, postoperative hospital stay, and perioperative morbidity. This information would be of interest to the surgeon contemplating neoadjuvant therapy for his or her patients. Unfortunately, there are no randomized studies comparing neoadjuvant vs adjuvant therapy for pancreatic cancer.
Dr. Evans and colleagues present their data on patients given preoperative chemoradiation (5.5 weeks of radiation combined with continuous-infusion fluorouracil [5-FU]) and note an isolated local or peritoneal recurrence rate of 11%, as compared with 50% to 80% in previous reports of surgical resection alone. However, 79% of their patients had distant metastasis, most commonly in the liver. Because gastrointestinal toxicity necessitates hospital admission in one-third of patients given this regimen, the authors also discuss the data on rapid-fractionation (or short-course) preoperative chemoradiation. They note a 10% locoregional recurrence and 25-month median survival, similar to that seen with standard-fraction chemoradiation. These data support the use of short-course chemoradiation over the standard rapid-fraction technique. Unfortunately, metastatic disease control is limited with either strategy.
The authors also address the use of gemcitabine (Gemzar)-based therapy, alone and in combination with radiation. However, due to the significant toxic effects associated with its use, the authors feel that this drug should be considered investigational when combined with radiation. On the other hand, Klein et al compared the use of gemcitabine to treatment with 5-FU and leucovorin in patients with advanced pancreatic cancer, without concomitant radiation. They found a significantly higher clinical benefit, but no increase in survival in the gemcitabine arm, with mild toxicity in both arms.
Significant strides have been made in the management of locoregional pancreatic cancer with the use of multimodality therapy. The article by Dr. Evans et al provides an excellent review of the current approaches to this disease. Future treatment regimens will need to account for the metastatic potential of this cancer with agents that are more effective systemically. In the meantime, earlier diagnosis of the disease using molecular markers and sensitive imaging techniques such as endoscopic ultrasound or positron-emission tomography may enable intervention before the disease has progressed beyond the locoregional stage.
1. Tierney WM, Fendrick AM, Hirth RA, et al: The clinical and economic impact of alternative staging strategies for adenocarcinoma of the pancreas. Am J Gastroenterol 95:1708-1713, 2000.
2. Gress FG, Hawes RH, Savides TJ, et al: Role of EUS in the preoperative staging of pancreatic cancer: A large single-center experience. Gastrointest Endosc 50:786-791, 1999.
3. Ahmad NA, Lewis JD, Ginsberg GG, et al: EUS in preoperative staging of pancreatic cancer. Gastrointest Endosc 52:463-468, 2000.
4. Ahmad NA, Kochman ML, Lewis JD, et al: Endosonography is superior to angiography in the preoperative assessment of vascular involvement among patients with pancreatic carcinoma. J Clin Gastroenterol 32:54-58, 2001.
5. Erickson RA, Garza AA: Impact of endoscopic ultrasound on the management and outcome of pancreatic carcinoma. Am J Gastroenterol 95:2248-2254, 2000.
6. Mertz HR, Sechopoulos P, Delbeke D, et al: EUS, PET, and CT scanning for evaluation of pancreatic adenocarcinoma. Gastrointest Endosc 52:367-371, 2000.
7. Klein B, Sadikov E, Mishaeli M, et al: Comparison of 5-FU and leucovorin to gemcitabine in the treatment of pancreatic cancer. Oncol Rep 7:875-877, 2000.
8. Myung SJ, Kim MH, Kim YS, et al: Telomerase activity in pure pancreatic juice for the diagnosis of pancreatic cancer may be complementary to K-ras mutation. Gastrointest Endosc 51:708-713, 2000.