|Year : 2016 | Volume
| Issue : 1 | Page : 417-421
Radiological tumor density and lymph node size correlate with survival in resectable adenocarcinoma of the pancreatic head: A retrospective cohort study
Soumil J Vyas1, Yogesh S Puri1, Biku J John1, Dominic Yu2, Jennifer Watkins3, Charles Imber1, Guiseppe Fusai1, Shankar Arjun1, Dinesh Sharma1, Brian R Davidson1, Massimo Malago1, Sakhawat Rahman1
1 Division of HPB and Liver Transplant Surgery, Royal Free Hospital and Medical School, London, NW3 2QG, UK
2 Division of Diagnostic and Interventional Radiology, Royal Free Hospital and Medical School, London, NW3 2QG, UK
3 Division of Histopathology, Royal Free Hospital and Medical School, London, NW3 2QG, UK
|Date of Web Publication||13-Apr-2016|
Soumil J Vyas
Division of HPB and Liver Transplant Surgery, Royal Free Hospital, Pond Street, London, NW3 2QG
Source of Support: None, Conflict of Interest: None
Introduction: Tumors within the pancreatic head show a variable density and enhancement on computerized tomography (CT). The relationship between the radiological appearance of pancreatic adenocarcinoma on CT and survival remains unclear. The aim of this study was to evaluate the relationship between the tumor density on CT and survival. We also evaluated the correlation between lymph node (LN) size and overall survival in patients undergoing pancreaticoduodenectomy for head of pancreas adenocarcinoma.
Materials and Methods: Case records of patients undergoing pancreaticoduodenectomy for the adenocarcinoma of pancreas head, between 2005 and 2009, were evaluated. CT was interpreted to document tumor density – Hounsfield unit (HU) and LN size of enlarged LNs. Histology was analyzed to review tumor differentiation and LN status. Survival was correlated with LN size and tumor density (HU).
Results: Increasing tumor density was significantly associated with an adverse outcome (P = 0.042, hazard ratio [HR] 1.034, 1.002–1.067 95% confidence interval [95% CI]). Patients with well-differentiated tumors had significantly lower tumor density as compared to moderately differentiated tumors (39.00 ± 26.00 vs. 71.31 ± 21.03 HU, P = 0.005). LN size more than 1 cm irrespective of LN status strongly correlated with the survival and was found to be an important prognostic factor (19.37 ± 2.71 months vs. 27.44 ± 2.74 months; P = 0.025; HR 2.70; 1.09–6.68 95% CI).
Conclusion: Increasing pancreatic tumor density and the lymph nodal size of more than 1 cm are strong predictors of unfavorable overall survival for resectable adenocarcinoma of the pancreatic head. Further studies are required to identify the value of these proposed prognostic factors.
Keywords: Adenocarcinoma of the head of pancreas, Hounsfield units, survival, tumor density
|How to cite this article:|
Vyas SJ, Puri YS, John BJ, Yu D, Watkins J, Imber C, Fusai G, Arjun S, Sharma D, Davidson BR, Malago M, Rahman S. Radiological tumor density and lymph node size correlate with survival in resectable adenocarcinoma of the pancreatic head: A retrospective cohort study. J Can Res Ther 2016;12:417-21
|How to cite this URL:|
Vyas SJ, Puri YS, John BJ, Yu D, Watkins J, Imber C, Fusai G, Arjun S, Sharma D, Davidson BR, Malago M, Rahman S. Radiological tumor density and lymph node size correlate with survival in resectable adenocarcinoma of the pancreatic head: A retrospective cohort study. J Can Res Ther [serial online] 2016 [cited 2019 Nov 15];12:417-21. Available from: http://www.cancerjournal.net/text.asp?2016/12/1/417/171358
| > Introduction|| |
Pancreatic ductal adenocarcinoma is the lethal and aggressive tumor with only 10–15% of the patients being eligible for the curative resection at presentation.,, The most reported tumor related prognostic factors in ductal adenocarcinoma of the head of the pancreas are tumor stage at presentation, tumor size, differentiation, lymph nodal status, perineural and the lymphovascular invasion, gross and microscopic venous invasion,, and completeness of the resection (R status).,, High resolution contrast-enhanced multidetector computerized topographic (MDCT) scanning with arterial, portal, and hepatic venous phase (MDCT) is currently the corner stone of the preoperative staging. However, the role of the CT scan to determine overall prognosis has remained poorly defined, not much data and the publications exist on the radiological tumor density and the lymph node (LN) size as possible prognostic factors for the overall survival in pancreatic adenocarcinoma. We hypothesize that the degree of the tumor differentiation would correlate with the tumor density and radiological tumor density and, therefore, can be predictive of the outcomes, hence, we evaluated tumor and pancreatic density as possible prognostic factors.
| > Materials and Methods|| |
All the patients undergoing pancreaticoduodenectomy (both pylorus preserving and classical Whipple's) in Tertiary Referral Centre between 2005 and 2009 were evaluated. Data collected were preoperative CT scans, complete histopathology including tumor size, tumor differentiation, lymph nodal status, number of LNs resected and number of LNs positive, lymphovascular and perineural invasion, and tumor differentiation. All histopathology slides were independently reviewed and reported as per existing guidelines. All CT scans were undertaken with following protocol, the patients were scanned 40–50 s after initiation of the contrast material injection in what is called pancreatic parenchymal phase. This demonstrated the enhancement patterns in the tumor and also showed accurately the peripancreatic arterial and the venous structure. The protocol 4 mm × 2.5 mm collimation with a reconstructed slice thickness of 5 mm was followed in all CT scans. 100 ml of Ultravist 300 was given as a contrast agent and the scan delay of 40 s was obtained. Coronal reformatted images (3 mm reconstruction slice complete) were created. Lymph nodal status was also evaluated according to set CT guidelines. CT scans were interpreted by the single consultant radiologist to specifically determine the density of the tumor and that of the adjacent pancreas in Hounsfield unit (HU).
Overall survivals were calculated on Kaplan–Meier curves. All the patients underwent standardized pancreaticoduodenectomy (pylorus preserving or the classical Kausch–Whipple procedure) performed by the same team of consultant surgeons. Data were analyzed using IBM SPSS V20 [IBM SPSS Inc.].
LN size of 1 cm and more was considered cut off. This was based on the fact that in most series of LNs <1 cm could be easily identified on the CT scans. The patients undergoing total pancreatectomy and pancreaticoduodenectomy for the ampullary and the bile duct cancers were excluded.
| > Observations and Results|| |
With respect to our observations on tumor density, complete radiological and histopathological data were available on 25 patients. 12/25 patients were deceased at the time of analysis and the meantime to mortality was 17.9 ± 8.25 months (range being 7 to 38 months). The mean time to follow-up was 17.6 ± 10.1 months [Table 1]. We correlated tumor density as an independent variable against survival. Tumor density as measured in the HUs was correlated against the histopathological grade of the tumor. This was, therefore, an attempt to identify if preoperative determination of the tumor density on CT scan (predictive of tumor grade) can be an additional prognostic marker in pancreatic adenocarcinoma [Table 2].
By far and large, the nontumor pancreas had greater density on the CT scan as compared to pancreatic tumor (93.8 ± 25.2 HU vs. 61.2 ± 24.2 HU). The patients with well-differentiated tumors had significantly lower tumor density as compared to moderately differentiated tumors (39.00 ± 26.00 HU vs. 71.13 ± 31.03 HU, P = 0.05), and although well-differentiated tumors had a lower tumor density compared to poorly differentiated tumors, it did not reach the statistical significance (39 ± 26 HU vs. 61.83 ± 13.35 HU, P = 0.19) [Table 2]. Patients with a lower tumor density on CT scan had a better outcome as compared to the patients with a higher tumor density. Hence, based on these findings, it appeared that the tumor density could be used as an important prognostic marker in pancreatic adenocarcinoma and a further prospective research in this direction is warranted. Complete radiological and the histological data for evaluation of the LNs as a prognostic marker was available in 35 patients and 1 out of 35 patients did not have any radiologically detectable LNs. 15 out of 35 patients had radiologically significant nodes that is more than 1 cm either at locoregional (n = 33) or at distant sites, (n = 2). 13 out of 35 patients were alive at the time of analysis. The mean follow-up period was 21 ± 17.5 months (4–101 months). The mean time to mortality was 19.3 ± 10 months. There did not appear to be any influence of the radiological or pathological tumor size on survival (P = 0.392) [Figure 1]. We also found that the LN size of more than 1 cm irrespective of lymph nodal status strongly correlated with survival and it was found to be an important prognostic factor (19.37 ± 2.71 months vs. 27.44 ± 2.74 months, hazard ratio [HR] =2.70, 1.09–6.68 95% confidence interval [95% CI], P = 0.025) [Figure 2]. As would be expected, histologically positive LNs adversely affected the survival (32.62 ± 0.8 vs. 21.3 ± 2.1 months, HR of 5.5 [1.2–25.3, 95% CI], P = 0.01) [Figure 3]. Thus, LNs size of more than 1 cm did not necessarily mean they were metastatic. The LN size of more than 1 cm may be due to inflammation and these reactive LNs by themselves may be poor prognostic factor in the resectable adenocarcinoma of the pancreatic head. The patients with a lower tumor density on CT scan had a better outcome as compared to the patients with a higher tumor density. We also found improved survival in the patients with well-differentiated tumors as compared to poorly differentiated lesions.
|Figure 1: Kaplan–Meier curve showing survival plot for overall survival in relation to tumor size|
Click here to view
|Figure 2: Kaplan–Meier curve showing influence of radiological lymph node size on survival|
Click here to view
|Figure 3: Kaplan–Meier curve showing influence of histological lymph node status on survival|
Click here to view
| > Discussion|| |
Prognostic factors in pancreatic adenocarcinoma are patient-related and tumor-related. For the purpose of the study, we are discussing only tumor-related factors. Conventionally, tumor-related factors are tumor stage at presentation, tumor size, differentiation, lymph nodal status, perineural and lymphovascular invasion, gross and microscopic venous invasion, and completeness of resection (R status).,,, Apart from these, an elevated C-reactive protein (CRP) and low albumin levels are associated with systemic inflammatory response related to the tumor that have also found to be associated with a poor survival., Elevated CRP has been shown to be a disease-independent prognostic factor in pancreatic ductal adenocarcinoma. An increase in CRP has been shown to be a predictor of poor prognosis and also predicts the recurrences and the poor survival. In the present study, LNs of more than 1 cm in size irrespective of histopathological status were associated with a poor outcome. Tumor-associated inflammation is associated with poor survival. This along with systemic inflammation is the cause of the elevated CRP. Thus, the systemic inflammatory response syndrome preoperatively indicated by high CRP is an adverse prognostic factor. The presence of previous intervention which includes endoscopy and stenting may lead to the development of inflammatory and the reactive LNs which may be an adverse prognostic factor. We did not carry out any separate analysis to check if there were nodes on the initial CT scan and on the scans done following intervention such as endoscopic stenting. However, this may be the focus of the future studies on the subject. More recently, the LN ratio that is ratio of the metastatic to the examined LNs has been found to be an independent and reliable prognostic indicator of the survival in resected pancreatic ductal adenocarcinoma.,, Our data with respect to tumor size in the present study were at variance with that suggested by Fortner et al., wherein the patients with tumor size of more than 2.5 cm were found to have a higher degree of involvement and worse survivals as compared to the patients with tumors <2.5 cm. In our study, the presence of high tumor density had an inverse relation with a degree of differentiation of the tumor. Therefore, with an increasing tumor density on the CT scanning, the histological grade of the tumor was worse. Hence, poorly differentiated tumors would be expected to have a higher tumor density and this would, therefore, be a poor prognostic factor. Thus, as we found that the increasing density of the tumor (39.00 ± 26 HU) for well differentiated tumors had a negative impact on the survival; it can be used as an important prognostic factor when assessing the patients for the prognosis. In a recent publication, increased pancreatic steatosis as measured in HU on CT scan was found to be associated with the abbreviated survival and was a negative prognostic factor. In similar concept, we found an increased tumor density of the pancreas on CT scan to have a negative correlation with overall survival. We did not find any literature specifically commenting on the pancreatic tumor density as measured on the CT scan as a prognostic indicator. In our knowledge, this may be the first reported series of pancreatic tumor density as a prognostic marker for a resectable pancreatic adenocarcinoma. Pancreatic stroma had been reported to be a prognostic indicator predicting the survival. Increasing amount of stroma had been correlated with the worse median survivals. Erkan et al. in their study defined the activated stroma index (ASI) and proposed that the increasing ASI, reflective of increasing amount of stroma was associated with a poor survival. ASI has been identified as a novel prognostic marker in the pancreatic adenocarcinoma. This is also reflective of the increasing tumor desmoplasia, which may account for increased tumor density on CT scan. It is possible that the increasing pancreatic tumor stroma and desmoplasia may be associated with increasing tumor density on CT scan. This corroborates the finding in our current study that the increased pancreatic tumor density is associated with an unfavorable prognosis. Smith et al. in meta-analysis showed that expression of the vascular endothelial growth factor (VEGF) has been a marker of adverse and a poor prognosis in the pancreatic cancer. It may be possible that increasing VEGF may account for the increased radiological tumor density. Another important area of investigation may be a relationship between the tumor grade and tumor density on CT scan. It is well-known that the survival is poor among high grade tumors and tumors with a poor differentiation. Wasif et al. in their publication suggested that the tumor grade and the differentiation be included in American Joint Committee on Cancer staging system. The relationship between the tumor grade and the tumor density should be looked at specifically with respect to further evaluating tumor density as a prognostic marker. Wasif et al. in their review also stated that a high tumor grade itself is influenced by the tumor size of more than 2 cm, positive LNs, and peripancreatic invasion. In fact, it is well-known that the patients with these characteristics in their primary tumor, positive LNs, and peripancreatic nodes would have had a higher tumor grade and more likely to be poorly differentiated as compared to a well-differentiated tumors. Hence, if the tumor grade can be predicted by the tumor density on a CT scan, identification of the tumor density as a factor can help stage and prognosticate the cancer. In one of the largest data on pancreaticoduodenectomies for the cancer, increasing tumor grade was identified as a negative prognostic factor with a HR of 1.6. However, not much data have been published with respect to looking at the association between the tumor grade and the tumor density on the CT scanning and in fact based on the tumor grade, Adsay et al. have devised a scoring system for pancreatic cancer similar to the prostate cancer which incorporates tumor grade and data., Tumor enhancement on MDCT has correlated with the amount of angiogenesis and it is inversely related to the tumor fibrosis. The tumor density in our study had a correlation with a degree of differentiation of the tumor. We found that the greater the tumor density on CT scan, more dedifferentiated the tumor was. We found that the patients with well-differentiated tumors had a significantly lower tumor density as compared to those with moderately differentiated tumors (39 ± 26.00 vs. 71.13 ± 21.03, P = 0.05). However, paradoxically, although well-differentiated tumors had a lower tumor density compared to poorly differentiated tumors which had a higher tumor density, they did not reach statistical significance (P = 0.19). The CT enhancement of the pancreatic tumor and pathological grading was evaluated by Wang et al. They, however, found that the well-differentiated tumors have slightly higher enhancement of the tumor as compared to the normal pancreatic parenchyma, while moderately differentiated tumors enhance less. Poorly differentiated tumors were likely to present as low density tumors with areas of necrosis. Hashimoto et al. in an interesting study used CT enhancement of the pancreas to predict the degree of pancreatic fibrosis. They found a positive correlation between the degree of pancreatic fibrosis as evaluated on the CT scan and pancreatic anastomotic leak. In a similar study, Hattori et al. have demonstrated that degree of pancreatic fibrosis and tumor angiogenesis on histology correlated with the tumor enhancement on the CT scanning. These factors are relevant in the degree of differentiation of the tumor as well and hence we feel that the degree of differentiation of the tumor can be predicted on the basis of CT enhancement as well. D'Onofrio et al. performed a contrast-enhanced ultrasound (CEUS) in the patients with pancreatic adenocarcinoma and found a positive correlation between CEUS and tumor differentiation and vascularity. Therefore, they concluded that CEUS findings would correlate with survivals based on tumor grade and may be in future used to provide neoadjuvant chemotherapy in selected patients. In a similar manner, evaluating the degree of pancreatic tumor density may help in identifying the grade of the tumor as we have proposed in this study and selecting appropriate neoadjuvant treatment in the selected patients.
The present study is obviously limited due to its retrospective nature and sample size. More prospective studies will be required to determine the precise value of LN status and tumor density as the prognostic factors for resectable pancreatic cancer. The presence of LN size and tumor density as such may not dictate clinical management but in the individual patients, this may help to prognosticate the condition.
| > Conclusion|| |
Increasing tumor density on CT scan and lymph nodal size of more than 1 cm irrespective of lymph nodal status are adverse prognostic markers of survival in adenocarcinoma of the pancreatic head. These should be added to the list of other more conventional indicators determining the prognosis in resected patients. More prospective studies would be required to determine the precise influence of these in determining the long-term survival.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Jamieson NB, Denley SM, Logue J, MacKenzie DJ, Foulis AK, Dickson EJ, et al.
A prospective comparison of the prognostic value of tumor- and patient-related factors in patients undergoing potentially curative surgery for pancreatic ductal adenocarcinoma. Ann Surg Oncol 2011;18:2318-28.
Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al.
Cancer statistics, 2008. CA Cancer J Clin 2008;58:71-96.
Sener SF, Fremgen A, Menck HR, Winchester DP. Pancreatic cancer: A report of treatment and survival trends for 100,313 patients diagnosed from 1985-1995, using the National Cancer Database. J Am Coll Surg 1999;189:1-7.
Sohn TA, Yeo CJ, Cameron JL, Koniaris L, Kaushal S, Abrams RA, et al.
Resected adenocarcinoma of the pancreas-616 patients: Results, outcomes, and prognostic indicators. J Gastrointest Surg 2000;4:567-79.
van Roest MH, Gouw AS, Peeters PM, Porte RJ, Slooff MJ, Fidler V, et al.
Results of pancreaticoduodenectomy in patients with periampullary adenocarcinoma: Perineural growth more important prognostic factor than tumor localization. Ann Surg 2008;248:97-103.
Mortele KJ, Banks PA, Silverman SG. State-of-the-art imaging of acute pancreatitis. JBR-BTR 2003;86:193-208.
Pine JK, Fusai KG, Young R, Sharma D, Davidson BR, Menon KV, et al.
Serum C-reactive protein concentration and the prognosis of ductal adenocarcinoma of the head of pancreas. Eur J Surg Oncol 2009;35:605-10.
Jamieson NB, Glen P, McMillan DC, McKay CJ, Foulis AK, Carter R, et al.
Systemic inflammatory response predicts outcome in patients undergoing resection for ductal adenocarcinoma head of pancreas. Br J Cancer 2005;92:21-3.
La Torre M, Cavallini M, Ramacciato G, Cosenza G, Rossi Del Monte S, Nigri G, et al.
Role of the lymph node ratio in pancreatic ductal adenocarcinoma. Impact on patient stratification and prognosis. J Surg Oncol 2011;104:629-33.
Bhatti I, Peacock O, Awan AK, Semeraro D, Larvin M, Hall RI. Lymph node ratio versus number of affected lymph nodes as predictors of survival for resected pancreatic adenocarcinoma. World J Surg 2010;34:768-75.
Pawlik TM, Gleisner AL, Cameron JL, Winter JM, Assumpcao L, Lillemoe KD, et al.
Prognostic relevance of lymph node ratio following pancreaticoduodenectomy for pancreatic cancer. Surgery 2007;141:610-8.
Fortner JG, Klimstra DS, Senie RT, Maclean BJ. Tumor size is the primary prognosticator for pancreatic cancer after regional pancreatectomy. Ann Surg 1996;223:147-53.
Mathur A, Hernandez J, Shaheen F, Shroff M, Dahal S, Morton C, et al.
Preoperative computed tomography measurements of pancreatic steatosis and visceral fat: Prognostic markers for dissemination and lethality of pancreatic adenocarcinoma. HPB (Oxford) 2011;13:404-10.
Erkan M, Michalski CW, Rieder S, Reiser-Erkan C, Abiatari I, Kolb A, et al.
The activated stroma index is a novel and independent prognostic marker in pancreatic ductal adenocarcinoma. Clin Gastroenterol Hepatol 2008;6:1155-61.
Smith RA, Tang J, Tudur-Smith C, Neoptolemos JP, Ghaneh P. Meta-analysis of immunohistochemical prognostic markers in resected pancreatic cancer. Br J Cancer 2011;104:1440-51.
Wasif N, Ko CY, Farrell J, Wainberg Z, Hines OJ, Reber H, et al.
Impact of tumor grade on prognosis in pancreatic cancer: Should we include grade in AJCC staging? Ann Surg Oncol 2010;17:2312-20.
Winter JM, Cameron JL, Campbell KA, Arnold MA, Chang DC, Coleman J, et al.
1423 pancreaticoduodenectomies for pancreatic cancer: A single-institution experience. J Gastrointest Surg 2006;10:1199-210.
Adsay NV, Basturk O, Bonnett M, Kilinc N, Andea AA, Feng J, et al.
A proposal for a new and more practical grading scheme for pancreatic ductal adenocarcinoma. Am J Surg Pathol 2005;29:724-33.
Hattori Y, Gabata T, Matsui O, Mochizuki K, Kitagawa H, Kayahara M, et al.
Enhancement patterns of pancreatic adenocarcinoma on conventional dynamic multi-detector row CT: Correlation with angiogenesis and fibrosis. World J Gastroenterol 2009;15:3114-21.
Wang ZQ, Li JS, Lu GM, Zhang XH, Chen ZQ, Meng K. Correlation of CT enhancement, tumor angiogenesis and pathologic grading of pancreatic carcinoma. World J Gastroenterol 2003;9:2100-4.
Hashimoto Y, Sclabas GM, Takahashi N, Kirihara Y, Smyrk TC, Huebner M, et al.
Dual-phase computed tomography for assessment of pancreatic fibrosis and anastomotic failure risk following pancreatoduodenectomy. J Gastrointest Surg 2011;15:2193-204.
D'Onofrio M, Zamboni GA, Malagò R, Mantovani W, Principe F, Gallotti A, et al.
Resectable pancreatic adenocarcinoma: Is the enhancement pattern at contrast-enhanced ultrasonography a pre-operative prognostic factor? Ultrasound Med Biol 2009;35:1929-37.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2]