| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, USA
Correspondence: Address correspondence and reprint requests to: David Shibata, MD, Division of Gastrointestinal Oncology, Department of Interdisciplinary Oncology, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, Florida 33612, USA; E-mail: shibatad{at}moffitt.usf.edu
| ABSTRACT |
|---|
|
|
|---|
Methods: We retrospectively reviewed 70 patients who underwent cytoreductive surgery for neuroendocrine hepatic metastases between 1996 and 2005. Twenty-two patients had pre and post-operative CgA and/or 5HIAA levels measured. Reduction of biomarkers following cytoreduction was correlated with patient symptoms and progression of disease following surgery.
Results: Our study consisted of 14 males and 8 females with a mean age of 55 (±12 years). Median follow-up was 18 months (range 564 months). Six patients (26.1%) had complete (R0) cytoreduction, while 4 (17.4%) and 13 (56.5%) had microscopic (R1) and gross (R2) disease remaining. All patients reported improvements in their symptoms, with 12 (54.5%) reporting complete resolution (CR) and 10 (45.5%) reporting partial resolution (PR). Reduction of CgA of
80% was highly predictive of complete resolution of symptoms (P = 0.007) and stabilization of disease (P = 0.034). Reduction of 5HIAA levels of
80% (or normalization) was predictive of symptomatic relief, but not progression of disease (P = 0.026 and P = 0.725). Five of six patients who had R0 resections had CR and were free of disease at last follow-up (median 24.5 months, range: 1148, P = 0.002).
Conclusions: We conclude that
80% reduction in CgA level following cytoreductive surgery for carcinoid tumors is predictive of subsequent symptom relief and disease control. Substantial reduction in CgA is associated with improved patient outcomes, even after incomplete cytoreduction.
Key Words: Carcinoid Chromogrannin A 5HIAA Serotonin Hepatic metastases Debulking
| INTRODUCTION |
|---|
|
|
|---|
Neuroendocrine tumors represent a broad range of cancers, which share their origins in the primitive neural crest. Production of bioactive substances is a shared feature of all functional neuroendocrine tumors, and their classification depends on their location and secretory behavior.1 Tumors that retain their ability to secrete bioactive amines at or above normal physiologic levels, which lead to symptomatic manifestations, are termed "functional" while tumors that do not produce clinical symptoms are termed "nonfunctional".2 The term "carcinoid" has been credited to Oberndorfer who recognized the slow, but progressive nature of these tumors nearly one hundred years ago.3 Current nomenclature limits the term carcinoid to those neuroendocrine tumors that secrete serotonin.
In contrast to other carcinomas, cytoreductive surgery for metastatic neuroendocrine tumors has been shown to provide symptomatic relief as well as improved overall survival.46 The development of metastases to the liver is a common finding in patients with carcinoid tumors, and in fact, up to 50% of patients with midgut primaries will have evidence of hepatic metastases at the time of their diagnosis.7 Aggressive surgical cytoreduction of hepatic metastases has been shown to provide significant benefit over medical therapy in the management of carcinoid syndrome and overall survival.4,6,8
The secretory nature of carcinoid tumors allows for quantification of tumor burden through the evaluation of biomarker levels. The product of serotonin metabolism, 5-hydroxyindole acetic acid (5HIAA), has been the most commonly used marker and has been shown to correlate with disease burden as well as progression.913 Although highly specific, the sensitivity of 5HIAA in identifying carcinoid progression has been limited. Chromogranin A (CgA), a glycoprotein associated with the secretory granules of amine producing cells, has more recently been used as a marker of disease progression in patients with carcinoid cancer. CgA has superior sensitivity to 5HIAA in identification of disease progression, and high plasma levels correlate with poor prognosis.10,1315 There have been no studies to date that describe the role of biomarkers following cytoreductive hepatic surgery for metastatic carcinoid cancer.
| PATIENTS AND METHODS |
|---|
|
|
|---|
All patients underwent cytoreductive hepatic resection and/or RFA. Most often, these techniques were used in combination to achieve a goal of at least 90% tumor volume reduction. Complete resection was performed whenever possible. Resection or ablation was considered complete (R0) if no evidence of gross disease was present at the termination of the surgery and no evidence of remaining microscopic disease was identified on final pathology. Resections in which all gross tumor was removed or ablated, but whose final pathology revealed positive microscopic margins were considered R1, while cytoreduction with known gross tumor left behind was considered R2.
Symptomatic response to surgery was considered partial (PR) if number of bowel movements and flushing episodes were reduced by at least 50% compared to pre-operative levels. Response was considered complete (CR) if the patient had resolution of symptoms.
Progression of disease was evaluated based on follow-up computed tomography scanning (CT) and/ or Octreoscan. All patients undergoing cytoreduction had follow-up CT scans, typically three months after surgery and every 6 months after that. Scans were done sooner or at more frequent intervals for any patient with return of symptoms or elevation of CgA or 5HIAA levels. Patients were considered to have no evidence of disease (NED) after undergoing R0 or R1 resection or ablation if they had no radiographic findings of disease (CT or Octreoscan) and were symptom free at the time of follow-up. Stable disease (SD) indicates patients who had R2 cyoreduction with known tumor remaining, but on follow-up, had no evidence of disease progression when compared to the immediate post-operative evaluations. Progression of disease (PD) refers to patients who had worsening of symptoms and/or evidence of disease progression on radiographic evaluation. Interpretion of radiographic response was based on current response evaluation criteria in solid tumors (RE-CIST).16
Pre and post-operative CgA and 5HIAA levels were compared and percent reduction was calculated. The percent reduction in biomarker levels (both CgA and 5HIAA) were then correlated with improvement in patient symptoms and stabilization (or progression) of disease. Fisher Exact test was used to correlate outcomes with biomarker reduction. P-value of < 0.05 was considered statistically significant.
| RESULTS |
|---|
|
|
|---|
80% was highly predictive of complete resolution of symptoms (P = 0.007) and stabilization of disease (P = 0.034), even in patients with gross residual tumor (Fig. 1b, c
80% reduction in CgA had evidence of disease progression at last contact (Fig. 1c
|
80% reduction (or normalization) of 5HIAA levels. Of these, 8 patients had complete resolution of symptoms and 3 patients had partial symptoms relief. One patient with 5HIAA reduction of < 80% had PD at last contact, while 3 patients with
80% reduction had developed PD. Reduction of 5HIAA of
80% (or normalization) was predictive of symptomatic relief (P = 0.026), but not progression of disease (P = 0.725).
Six of twenty-two patients had R0 cytoreduction with curative intent. Five of these patients (83.3%) had complete resolution of symptoms and were free of disease at last follow-up (median 24.5 months, range 1148 months). R0 or R1 resection was highly predictive of complete symptomatic relief (Fig. 1e
) and control of disease progression (P = 0.038 and P = 0.003, respectively) (Fig. 1f
). We had one patient who developed an early hepatic recurrence despite what was thought to be an R0 resection. Of note, this patients CgA level dropped by only 63.6%. The remaining patients who had R0 resection were all NED at last follow-up.
| DISCUSSION |
|---|
|
|
|---|
80% correlates directly with improvement in symptoms and control of disease progression (Fig. 1b, c
80% reduction or normalization of 5HIAA was highly predictive of symptomatic relief, but did not reach statistical significance in its ability to predict control of disease progression (Fig. 1dWe evaluated only those patients with hepatic metastases from functional carcinoid tumors. In addition, we included only those patients who had elevated biomarkers despite somatostatin therapy and who had both pre and post-operative biomarker levels drawn within 3 months of surgery. Many patients had levels drawn prior to initiation of somatostatin therapy and then underwent cytoreduction without repeat levels being measured. We felt that inclusion of these patients would have allowed for bias because the drop in biomarker level could not be attributed specifically to our surgical intervention but may have been caused by the medical therapy. Although limited by the small cohort of patients meeting our inclusion criteria, to our knowledge, this study represents the first report that shows a predictive value of post-operative biomarker data.
Cytoreductive surgery for metastatic neuroendocrine tumors is becoming more widely used as outcomes with hepatic surgery continue to improve and less invasive techniques (RFA) have progressed. Acknowledging the benefits of surgical cytoreduction, we attempted to identify predictive factors which could be used to evaluate the efficacy of surgical therapy in the immediate post-operative period. Currently, surgical interventions are evaluated with subjective symptomatic measures. Following cytoreduction, management is guided by a watchful waiting approach, with treatment based on progression of disease. Proactive management is impossible since there is currently no way to predict outcomes. We hypothesized that biomarkers (CgA and 5HIAA) could be used as objective measures to evaluate the efficacy of therapy and potentially guide patient care following cytoreduction. Those patients who do not demonstrate at least 80% reduction in CgA or 5HIAA may benefit from more aggressive follow-up with early intervention in the case of recurrence or progression of disease. We have validated a strong correlation between reduction of these levels and patient outcomes.
Many authors have reported the benefits of cytoreductive surgery in terms of symptom improvement, disease progression, and overall survival.4,8 Hepatic artery embolization is also widely used for advanced disease although surgical resection/ablation appears to compare favorably to this method.17,18 In a previous study, Osborne et al. reviewed 120 patients who had either HAE or surgical cytoreduction for hepatic metastases from neuroendocrine tumors at our institution between 1996 and 2005, and found a significant improvement in symptoms and overall survival in the operative cohort.19 Initial symptomatic relief was similar between groups; however, duration of relief was significantly longer in the surgical group [mean 33.3 months versus 20.3 months (P < 0.001)]. Perhaps most importantly, mean survival for patients undergoing operative cytoreduction was almost doubled (40.3 months versus 21.3 months) when compared to embolization therapy (P < 0.001). These findings compare favorably to other reports and clearly demonstrate that surgical cytoreduction is an effective means to achieve both palliation and long-term survival for patients with hepatic metastases from neuroendocrine tumors.
Other investigators have identified the utility of biomarkers as prognostic indicators in the management of patients with neuroendocrine tumors.10 Multiple studies have shown that increased levels of 24-hour urinary 5HIAA correlate with decreased survival and severity of symptoms. In 1991, Agronovitch et al. described the utility of 5HIAA in identifying patients with poor prognosis. Those with elevated levels of 5HIAA had decreased survival compared to those who had normal levels.12 Similarly, 5HIAA levels of > 300 micromoles/day and > 500 micromoles/day have been identified as indicators of severity of carcinoid symptoms as well as poor outcomes.9,13
Chromogranin A is an acidic glycoprotein found in the secretory vesicles of endocrine and neuroendocrine cells. Although CgA has no inherent bioactive function, it has been found to be an excellent marker for tumor burden and disease progression. Sensitivity and specificity of 93% and 88% respectively have been reported.20 Many authors have shown superiority of CgA over other biomarkers in the management of patients with carcinoid tumors.21 Similar to studies of 5HIAA, CgA has been shown to be a useful prognostic indicator, with elevated serum levels harboring poor prognosis.2224 Others have confirmed that elevated levels of CgA and 5HIAA are independent predictors of survival in patients with carcinoid tumors.13
We have found that percent reduction in biomarker levels following cytoreduction correlates with relief of symptoms and control of disease progression in patients with hepatic metastases from functional carcinoid tumors. A significant number of patients with neuroendocrine tumors are asymptomatic despite metastatic disease and progression is noted only by elevated biomarker levels. In cases such as these, it may be particularly important to have accurate means to evaluate surgical intervention. We have shown that CgA and 5HIAA can be used to accurately assess the efficacy of cytoreductive surgery. In patients who do not achieve at least 80% reduction of biomarker levels, consideration may be given to more aggressive follow-up with early surgical intervention if progression or recurrence is identified. For hepatic metastases, staged procedures may be necessary to preserve hepatic function while achieving optimal debulking.
Evaluation of optimal management of carcinoid and other neuroendocrine tumors is made difficult by their tendency towards slow progression and prolonged survival even in the face of metastatic disease.25,26 While these issues may be reassuring to patients, they make interrogation of treatment modalities difficult. As larger groups of patients are studied, it is becoming apparent that surgical therapy for metastatic disease is perhaps the most effective means of achieving disease control and potential cure. As noted by us and others, patients who are able to have complete R0 resection have significantly improved outcomes compared to those who have R1 or R2 debulking.2729 Patients who do not demonstrate reduction of biomarkers to the level expected for an R0 resection should be followed closely, as they are likely more prone to develop early recurrences that may be amenable to repeat therapy. We used 80% reduction as our cutoff for statistical analysis because it was slightly less than the mean reduction for our study group and clearly an obtainable objective. Rather than a hard index for cytoreduction, this level may be considered as a guide. Patients who achieve this level certainly have a tendency towards improved outcomes, while patients who do not, seem to do more poorly.
This study reveals that improved patient outcomes correlate with the extent of biomarker reduction following cytoreductive surgery for hepatic metastases from carcinoid tumors. We have found that CgA may be used as an indicator of the efficacy of surgical cytoreduction and as a predictor of patient outcomes. While 5HIAA was an excellent predictor of symptomatic relief, we did not find reduction in 5HIAA levels to be predictive of disease control. Biomarkers should be an integral part of pre and post-operative evaluation for patients with carcinoid and other neuroendocrine tumors. Consideration should be given to more aggressive follow-up and potentially further surgical intervention in those patients who do not demonstrate expected reduction in biomarker levels following cytoreductive surgery.
| FOOTNOTES |
|---|
Received for publication April 11, 2006. Accepted for publication June 13, 2006.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. C. Yao, A. T. Phan, D. Z. Chang, R. A. Wolff, K. Hess, S. Gupta, C. Jacobs, J. E. Mares, A. N. Landgraf, A. Rashid, et al. Efficacy of RAD001 (Everolimus) and Octreotide LAR in Advanced Low- to Intermediate-Grade Neuroendocrine Tumors: Results of a Phase II Study J. Clin. Oncol., September 10, 2008; 26(26): 4311 - 4318. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |