Operative intervention for recurrent adrenocortical cancer
Benzon M. Dy, MD,a Kevin B. Wise, MD,a Melanie L. Richards, MD,a William F. Young, Jr, MD,b Clive S. Grant, MD,a Keith C. Bible, MD, PhD,” Jordan Rosedahl, BS,d William S. Harmsen, MS,d David R. Farley, MD,a and Geoffrey B. Thompson, MD,ª Rochester, MN
Introduction. Adrenocortical cancer (ACC) recurs despite apparent complete resection. We examined the survival and palliative benefit of resection for recurrent ACC.
Methods. A review of all patients undergoing operation for ACC between 1980 and 2010 at our institution was performed in which we compared resection with nonoperative therapy.
Results. Overall, 164 patients underwent operation for ACC, 125 of whom underwent a complete resection (RO). Recurrence occurred in 93 RO patients (median, 15 months; range, 1.5-150 months). Symptoms at recurrence were present in 71% (66/93), including pain (34%) and hormone excess (43%). There were 67 patients who underwent reoperation for recurrence. Forty-eight of 67 patients underwent R0 resection for recurrence. Operative patients had a greater overall operative versus nonoperative management or no therapy (65 months vs 6 months, P < . 01). Median survival for nonoperatively managed patients (226 days) and those undergoing no therapy (179 days) was less than for debulking (1,272 days, P = . 002). RO for recurrence (P = . 005) and a disease-free interval >6 months (P <. 001) were associated with survival after operation, whereas original tumor size (P = . 47), grade (P = . 8), and stage (P = . 23) were not. Pain and hormonal symptoms improved in 84% of operative patients versus 29% of nonoperatively managed patients (P = . 005). Debulking had similar symptomatic improvement to R0 resection (P = .52).
Conclusion. Patients with recurrent ACC can benefit from operative intervention with improvement in survival and symptoms. Patients with a disease-free interval >6 months and complete resection are likely to benefit from resection of the recurrence, but the near universal improvement in symptoms may expand the criteria for operation in recurrent ACC. (Surgery 2013;154:1292-9.)
From the Division of Endocrine Surgery, Department of Gastroenterologic and General Surgery,” Department of Endocrinology, Department of Medical Oncology,” and Department of Statistics,ª Mayo Clinic, Rochester, MN
ADRENOCORTICAL CARCINOMA (ACC) is a rare malig- nancy with an estimated incidence of 1-2 per million people.1 The initial presentation ranges from incidental discovery to symptoms of hormone excess, mass effect, or pain. Most series report that greater than one-half of patients with ACC have functional tumors with cortisol or mixed hormonal secretion.2 The majority of patients present with advanced neoplasms that have invaded adjacent or- gans or metastatic disease and may not be candi- dates for complete resection.3 In patients with localized disease, operative resection remains the
Accepted for publication June 25, 2013.
Reprint requests: Geoffrey B. Thompson, MD, Section Head, Di- vision of Endocrine Surgery, Department of Gastroenterologic and General Surgery, Mayo Clinic, 200 First Street SW, Roches- ter, MN 55905. E-mail: thompson.geoffrey@mayo.edu.
0039-6060/$ - see front matter
@ 2013 Mosby, Inc. All rights reserved.
http://dx.doi.org/10.1016/j.surg.2013.06.033
mainstay of therapy. Patients with early-stage tu- mors who undergo a complete resection have a 40% 5-year survival rate whereas those with resid- ual disease fare poorly.4 Despite apparent com- plete microscopic operative resection, ACC recurs both locally or with distant metastases up to 50% of patients.4
Currently, patients with resected stage IV ACC have few options for adjuvant chemotherapy. The mainstay has been mitotane, which currently is the only Federal Drug Administration-approved medication for adjuvant treatment of ACC. This drug is often poorly tolerated and may be most effective in patients who achieve a therapeutic range to yield a survival benefit.5 Response to sys- temic therapy may aid in predicting favorable out- comes in patients but does not provide an option for patients with chemoresistant disease.6 Recently, Fassnacht et al7 published the results of two different leading multimodality therapies, which both included mitotane for ACC in a randomized,
controlled, multicenter trial. Their study included patients who initially presented with stage IV dis- ease. In patients who underwent treatment, the median progression-free survival was 5 months and 2.1 months in each arm, indicated poor out- comes in response to salvage systemic therapies. Despite this aggressive, multimodality treatment, the control of disease and survival for stage IV ACC continues to be challenging. Other treatment options include radiofrequency ablation or radia- tion therapy for peritoneal, bone, and visceral me- tastases, but the benefits from these approaches remain unknown.
Despite rare reports of complete responses to adjuvant chemotherapy, complete resection of primary disease and metastases has been consid- ered to be the only potentially curative option for ACC. Predictors of survival after operation for recurrence and metastatic disease have not been firmly established nor are outcomes fully known. Several studies have, nevertheless, suggested increased survival after metastasectomy in patients with isolated liver or lung metastases8,9 In addition, patients are often symptomatic with recurrent disease stemming from pain or stigmata of hormone-producing neoplasms and may benefit from palliative resection. In this context, we comprehensively reviewed both short- and long- term outcomes from both a palliative and survival perspective for surgical resection in ACC patients with locally recurrent and distant metastatic disease at the Mayo Clinic during a recent 30-year period.
METHODS
Records of patients who underwent surgery for ACC between 1980 and 2010 at Mayo Clinic, Rochester, were reviewed retrospectively. Patients who underwent a first metastasectomy or resection of locoregional recurrence were examined for both disease-free interval (DFI), overall survival (OS), and palliation of symptoms, including pain and symptoms stemming from hormonal excess. DFI was defined as time from resection of index ACC to detection of first recurrence on imaging; OS was defined as the time of first recurrence after resection of the index tumor to either time of last follow-up or to time of death. Metastasectomy was defined as operative resection from sites separate from the original tumor bed, such as the liver, lung, bone, or brain whereas intra-abdominal dis- ease adjacent to the tumor bed was considered locoregional recurrence. R0 resection was defined as complete resection of disease with microscopi- cally negative margins. Debulking of tumor involved resection of tumor with gross disease
remaining after operative intervention. The pres- ence of pain and symptoms of hormonal excess were screened during preoperative consultation and overactive secretion was confirmed with biochemical testing. Improvement in symptoms was assessed at postoperative visits. Improvement in hormonal symptom improvement was assessed with biochemical evaluation if the patient was not on glucocorticoids- or mineralocorticoid- replacement therapy postoperatively.
Those who underwent operation for recurrent and metastatic disease were compared with pa- tients with recurrent disease who had no opera- tive intervention and had either systemic therapy or no therapy at all for recurrent disease. Patients were included in this study if they had their primary operation performed at our institution or at an outside institution if pathology was available for review. Patients were excluded from statistical analysis if they did not have an R0 resection performed at initial operation, if they underwent neoadjuvant chemotherapy prior to resection of their initial tumor, or if they declined to participate in research studies. Each patient underwent a comprehensive preoperative evalua- tion by a multidisciplinary team involving endo- crinologists, medical oncologists, and surgeons. Staging was evaluated according the current American Joint Committee on Cancer 7th edition classification system. Follow-up was performed at regular intervals postoperatively with computed tomography scans as the main modality for documentation of recurrence with the use of positron emission tomography imaging with 18F- fluorodeoxyglucose at the discretion of providers. Institutional review board approval was obtained for this study.
Statistical analysis was performed with SAS software (version 9.2; SAS Inc, Cary, NC). Survival curves were performed using the Kaplan Meier Method. Pearson x2 or Fisher exact tests were used for comparison of nominal data for baseline char- acteristics. Ordinal data were compared with the Spearman rank-order correlation coefficient.
RESULTS
We identified 164 patients who underwent surgery for ACC at Mayo Clinic, Rochester, Minne- sota, between 1980 through 2010. There were 102 patients who had their primary operation per- formed at Mayo Clinic whereas 62 patients had their initial resection performed elsewhere and referred to our institution for recurrence or metastatic disease. Of these 164 patients, 125 had an R0 resection (Fig 1).
Overview of patients with adrenocortical cancer operatively managed from 1980-2010.
164 patients
39 R1 or greater
125 Ro resection
93 recurrent ACC
32 No recurrence
26 Non-operative management
67 Surgery
48 Rg resection
19 R1 or greater
Of the 125 patients who had a primary R0 resection, recurrence occurred in 93 (74%) at a median time of 15 months (range, 1.5- 150 months) from the time of surgery. There were 67 of these patients (72%) with recurrence who underwent 115 operative procedures, whereas 26 patients had nonoperative therapy or no intervention. Demographics are listed on Table I. Of note, patients with nonoperative therapy were more likely to have distant metastases in the lung and less frequent intra-abdominal recurrence. A greater proportion of patients treated nonopera- tively had a greater original stage tumors and a lesser DFI. The identified sites of recurrent disease for those who underwent operative resection of local disease were locoregional (n = 39), liver (n = 20), lung (n = 14), and other sites (n = 13) (Tables II and III). The median number of opera- tions performed was 2 (range, 1-12), and the median follow-up time was 48 months (range, 2- 239 months).
Survival was calculated for patients with recur- rence undergoing operative intervention versus nonoperative intervention or no intervention. OS at 1, 2, and 5 years for patients undergoing resection of metastases was 82%, 67%, and 30% compared with 26%, 13%, and 0%, in those undergoing adjuvant therapy (respectively) versus 30%, 10%, and 0% among those subjected to best supportive care alone (P < . 0001; Fig 2). When a subset analysis was performed between those un- dergoing systemic therapy versus best medical management in the nonoperatively managed groups, there was no difference in their overall characteristics and median survival.
Symptoms at recurrence were present in 71% (66/93) of patients, including pain in 32 of 93 (34%) and hormone excess in 40 of 93 (43%) patients. Cortisol hypersecretion was most com- mon with 20 of 93 patients symptomatic at recur- rence followed by aldosterone excess in 6 patients, androgens in 2 patients, and mixed hormonal secretion in 12 patients.
Patients who did not achieve an R0 resection for recurrent disease had a lesser OS than patients who had a complete resection of their recurrence and/or metastases (Fig 3); in contrast, patients who underwent operative intervention for debulk- ing had an improved median survival of 3.5 years compared with patients who underwent adjuvant treatment without operation or supportive care alone (P= . 002). Patients who underwent adjuvant treatment alone or no therapy had similar out- comes (median = 7.4 months vs 6.0 months, P = . 43).
Complete resection of recurrence (P = . 005) and a DFI greater than 6 months (P <. 001) were both independently associated with improved sur- vival among patients undergoing operative inter- vention. The location of recurrence was not predictive of OS nor was the number of proce- dures performed for recurrence. Other factors examined included original size (P = . 47), grade (P =. 8), and stage (P= . 23) of tumor. These factors were also not predictive of improved outcomes af- ter operative intervention. Patient sex was statisti- cally significant predictor of survival as women had an improved OS (Table IV).
Pain and symptoms associated with hormone excess improved in 84% of patients undergoing
| Surgery, 67 patients | Nonsurgery, 26 patients | P value | |
|---|---|---|---|
| Site of recurrence | |||
| Abdomen | 39 | 6 | .003 |
| Liver | 14 | 6 | .79 |
| Lung | 21 | 17 | .004 |
| Brain | 2 | 0 | 1 |
| Bone | 10 | 3 | 1 |
| Number of sites of recurrence | .92 | ||
| 1 | 52 (77%) | 20 (77%) | |
| 2 | 12 (18%) | 6 (23%) | |
| 3 | 2 (3%) | ||
| 4 | 1 (2%) | ||
| Size of original tumor (median) | 11 | 13.5 | .19 |
| Grade of original tumor (median, | .34 | ||
| range) | |||
| 1 | 1 | 3 | |
| 2 | 10 | 3 | |
| 3 | 9 | 5 | |
| 4 | 8 | 4 | |
| Stage of original ACC (median, range) | .02 | ||
| 1 | 1 (2%) | 0 | |
| 2 | 46 (69%) | 10 (38%) | |
| 3 | 3 (4%) | 2 (8%) | |
| 4 | 17 (25%) | 14 (54%) | |
| Disease-free interval | .005 | ||
| <6 months | 13 (19%) | 13 (50%) | |
| >6 months | 54 (81%) | 13 (50%) | |
| Sex | .17 | ||
| Men | 35 (52%) | 9 (35%) | |
| Women | 32 (48%) | 17 (65%) |
ACC, Adrenocortical carcinoma.
| Location of recurrence | No. patients | Median survival, y |
|---|---|---|
| Abdomen | 39 | 7.8 |
| Liver | 14 | 5.8 |
| Lung | 20 | 6.2 |
| Brain | 2 | 1.3 |
| Bone | 11 | 3.7 |
ACC, Adrenocortical carcinoma.
operative intervention compared with 29% of nonoperatively managed patients (P = . 005). When examining patients who underwent adju- vant therapy compared with no therapy at all, we found no difference in improvement of symptoms. Overall, debulking was associated with similar symptomatic improvement to that seen in patients after an R0 resection (P = .52) with an improve- ment in 80% of operatively managed patients.
| Number of sites of recurrence | No. of patients | Median survival, y |
|---|---|---|
| 1 | 52 | 5.8 |
| 2 | 12 | 4.2 |
| 3 | 2 | 1.8 |
| 4 | 1 | 2.4 |
ACC, Adrenocortical carcinoma.
DISCUSSION
Operative intervention traditionally has been considered to be contraindicated in patients with recurrent and metastatic ACC despite the poor outcomes attained among patients receiving salvage systemic therapy in the absence of opera- tive exploration for potential resection. No pro- spective, randomized trial has been performed to evaluate a survival benefit in response to adjuvant
Survival According to Intervention
1.0
Kaplan-Meier Estimate
-55 pts
0.8
-45 pts
0.6
-24 pts
Surgery 67 patients
0.4
0.2
No intervention 9 patients
Non-operative
0.0
17
patients
intervention
0
1
2
3
4
5
6
7
8
9
10
Time (Years)
Survival by Resection Type
1.0
-43 patients
Kaplan-Meier Estimate
P = 0.005
0.8
-30 patients
0.6
RO Resection
48
patients
0.4
-10
patients
L
0.2
0.0
Non-RO Resection 19 patients
0
1
2
3
4 5 6 7 8 9
10
Time (Years)
mitotane therapy with retrospective cohort studies producing conflicting results.5,10,11 As knowledge of tumor biology has increased over time, criteria such as progression-free survival and resectable recurrence have emerged as possible indications for a meaningful intervention in patients with recurrent disease. This aggressive approach has been the practice at our institution as well as at many others for several years, but data supporting this approach are sparse.
DFI between initial resection and first disease recurrence was found to be a predictor of out- comes in patients undergoing operative interven- tion for favoring those who had a DFI greater than 6 months. Recently, Erdogan et al12 published the collective experience from the German
Adrenocortical Carcinoma Registry and found that a time to first recurrence greater than 12 months was predictive of improved outcomes af- ter metastasectomy compared with time to first recurrence less than 12 months. This difference may be explained by their inclusion of patients who did not achieve an R0 resection and had pro- gression of disease rather than a recurrence after R0 resection. This may be expected to lead to early failures among patients who undergo operation because a complete resection is not achieved or the tumor capsule had been violated during pri- mary intervention.
In this present study, stage, grade, and size of the original neoplasm were not found to be pre- dictive of successful outcomes in patients with
| Patient characteristic | Hazard ratio | P value |
|---|---|---|
| Site of recurrence | .8 | |
| Abdomen | 0.822 | |
| Bone | 1.94 | |
| Brain | 4.9 | |
| Liver | 1.17 | |
| Lung | 1.11 | |
| Number of sites | .1 | |
| 1 | Reference | |
| 2 | 1.3 | |
| 3 | 11.24 | |
| 4 | 3.93 | |
| Size of tumor | 1 | .47 |
| Grade of tumor | .8 | |
| 1 | Reference | |
| 2 | 1.63 | |
| 3 | 1.26 | |
| 4 | 1.2 | |
| Stage of ACC | .23 | |
| 2 | Reference | |
| 3 | 0.822 | |
| 4 | 1.2 | |
| Disease free >6 mos | .001 | |
| Yes | Reference | |
| No | 3.54 | |
| R0 resection | .005 | |
| Yes | Reference | |
| No | 2.61 | |
| Sex | .04 | |
| Female | Reference | |
| Male | 2.02 |
ACC, Adrenocortical carcinoma.
recurrence. Stage was not an independent factor for poor outcomes even in patients who presented initially with stage IV disease, including invasion into the adrenal vein, renal vein, or inferior vena cava. This finding may initially seem contradictory to an earlier report from our group published by Kendrick et al,4 where stage of the original tumor was found to be an independent predictor of sur- vival when comparing stage I/II with stage III and IV tumors (American Joint Committee on Cancer, 5th Edition). This apparent contradiction may be the result of selection bias as in the context of the present study because we examined patients with earlier-stage tumors that have an underlying aggressive pathology, whereas those who did not recur were excluded from our study population. In addition, patients with stage III tumors may be understaged because lymph nodes are not
consistently reported in examination of tumor specimens.
Similarly, our colleagues’ previous report iden- tified adjuvant therapy with mitotane as a favorable predictor of outcomes, but adjuvant therapy was not found to improve survival in this study. We suspect that our study group may represent pa- tients who were resistant to systemic therapy or those intolerant to systemic treatment. This possi- bility is perhaps not surprising because the efficacy of adjuvant therapy with mitotane is widely debated, with results ranging from a threefold decrease in recurrence in a multi-institutional European trial to no difference at all reported from MD Anderson. 13,14
It is important to note that the median number of reoperations performed in our study was 2. One patient underwent 12 procedures with a survival of 70 months from the initial operation. In our experience, there does not seem to be an absolute limitation to the number of operative re- exploration procedures; rather, the indication to undergo an operation is limited by the preopera- tive assessment of attaining an R0 resection, the time to detection of recurrent disease, and the overall functional status of the patient. When extending survival data from 5 years to 10 years, there is a decline in OS, suggesting that a cure is possible albeit rare in ACC. Although OS appears improved with operative intervention, it is pres- ently unclear which patients with recurrent and metastatic disease will benefit most, saving for apparent improved outcomes with greater initial DFI.
It is worth mentioning that patients who under- went debulking procedures (R2) fared better (me- dian survival = 3.5 years) than patients undergoing nonoperative treatment or supportive care after recurrence. Although R2 resections have previ- ously been reported to have no survival benefit, our results indicate an improvement over adjuvant therapy alone or supportive therapy alone.8 It is unclear whether the better survival associated with debulking is associated directly with a decrease in tumor volume from operative interven- tion or instead a favorable tumor biology with an indolent disease course or some selection bias.
An inherent limitation of this study is its retrospective design and inherent selection bias. Our multidisciplinary approach has been to eval- uate patients on an individual basis, combining factors including DFI, response to therapy, extent of recurrence, and patient’s Eastern Cooperative Oncology Group performance status in evaluating the appropriateness of operative intervention. The
disparity between those offered an operation versus managed nonoperatively is evident in the demographics and original tumor characteristics of the operatively managed patients compared with the nonoperative group. The nonoperatively managed patients have a greater proportion of greater-stage tumors, a lesser DFI, and more distant metastases. In addition, the number of sites of recurrence or location of recurrence is difficult to compare between patients who underwent resection and who did not. In general, those offered resection were more likely to have a lesser burden of disease or have disease confined to a single area within the site of recurrence or metas- tasis. Patients with evidence of carcinomatosis on preoperative imaging were excluded from opera- tive intervention.
In an attempt to select patients who will have a favorable outcome postoperatively, our approach frequently includes the use of the Berruti protocol, which uses mitotane, etoposide, doxorubicin, and cisplatin with reimaging and evaluation performed after 2 or 3 months of therapy.15 Those patients with responsive or stable disease with no indication of additional sites of disease are considered for operative intervention. The favorable outcomes achieved in our study do advocate for an aggressive operative approach in appropriately selected pa- tients when favorable tumor biology is probable. The rarity of ACC leads to difficulty in studying outcomes prospectively and highlights the need for collaboration among large tertiary cancer insti- tutions. In our study, selection bias is an inherent weakness of this retrospective analysis and prompts us to continue to advocate for a multicenter data- base or ideally prospective examination of re- sponses to future treatments.
Unique to our analyses is the evaluation of the effects of operative metastasectomy on symptoms of patients with recurrent disease. Palliation of symptoms may provide an additional indication for resection of recurrence or metastasectomy in patients with recurrent ACC. Pain is a known consequence of patients with recurrent disease, but ACC also is associated frequently with the additional factor of hormonal excess that can often be debilitating for patients. Several medica- tions exist that can ameliorate symptoms, such as metyrapone, ketoconazole, mifepristone, and mi- totane, but the side effect profiles are often limiting.16 Mitotane also has a narrow therapeutic index, and many patients have difficulty with toler- ance. In our study, 84% of patients undergoing re- operation, had improvement in symptoms of either pain and/or from those secondary to
hormonal excess, which is markedly better than observed in nonoperatively managed patients. In addition, both patients who underwent R0 and de- bulking procedures had similar subjective improve- ments in symptoms (P = . 52), which may suggest that symptom palliation may represent an impor- tant indication for operative exploration even when a curative resection is not deemed possible.
REFERENCES
1. Golden SH, Robinson KA, Saldanha I, Anton B, Ladenson PW. Clinical review: prevalence and incidence of endocrine and metabolic disorders in the United States: a comprehen- sive review. J Clin Endocrinol Metab 2009;94:1853-78.
2. Icard P, Goudet P, Charpenay C, Andreassian B, Carnaille B, Chapuis Y, et al. Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French Associa- tion of Endocrine Surgeons study group. World J Surg 2001; 25:891-7.
3. Dackiw AP, Lee JE, Gagel RF, Evans DB. Adrenal cortical car- cinoma. World J Surg 2001;25:914-26.
4. Kendrick ML, Lloyd R, Erickson L, Farley DR, Grant CS, Thompson GB, et al. Adrenocortical carcinoma: surgical progress or status quo? Arch Surg 2001;136:543-9.
5. Luton JP, Cerdas S, Billaud L, Thomas G, Guilhaume B, Ber- tagna X, et al. Clinical features of adrenocortical carci- noma, prognostic factors, and the effect of mitotane therapy. N Engl J Med 1990;322:1195-201.
6. Gonzalez RJ, Tamm EP, Ng C, Phan AT, Vassilopoulou-Sellin R, Perrier ND, et al. Response to mitotane predicts outcome in patients with recurrent adrenal cortical carcinoma. Sur- gery 2007;142:867-75.
7. Fassnacht M, Terzolo M, Allolio B, Baudin E, Haak H, Ber- ruti A, et al. Combination chemotherapy in advanced adre- nocortical carcinoma. N Engl J Med 2012;366:2189-97.
8. Gaujoux S, Al-Ahmadie H, Allen PJ, Gonen M, Shia J, D’An- gelica M, et al. Resection of adrenocortical carcinoma liver metastasis: is it justified? Ann Surg Oncol 2012;19:2643-51.
9. op den Winkel J, Pfannschmidt J, Muley T, Grunewald C, Dienemann H, Fassnacht M, et al. Metastatic adrenocortical carcinoma: results of 56 pulmonary metastasectomies in 24 patients. Ann Thorac Surg 2011;92:1965-70.
10. Vassilopoulou-Sellin R, Guinee VF, Klein MJ, Taylor SH, Hess KR, Schultz PN, et al. Impact of adjuvant mitotane on the clinical course of patients with adrenocortical can- cer. Cancer 1993;71:3119-23.
11. van Slooten H, Moolenaar AJ, van Seters AP, Smeenk D. The treatment of adrenocortical carcinoma with o, p’- DDD: prognostic implications of serum level monitoring. Eur J Cancer Clin Oncol 1984;20:47-53.
12. Erdogan I, Deutschbein T, Jurowich C, Kroiss M, Ronchi C, Quinkler M, et al. The role of surgery in the management of recurrent adrenocortical carcinoma. J Clin Endocrinol Metab 2013;98:181-91.
13. Grubbs EG, Callender GG, Xing Y, Perrier ND, Evans DB, Phan AT, et al. Recurrence of adrenal cortical carcinoma following resection: surgery alone can achieve results equal to surgery plus mitotane. Ann Surg Oncol 2010;17:263-70.
14. Terzolo M, Angeli A, Fassnacht M, Daffara F, Tauchmanova L, Conton PA, et al. Adjuvant mitotane treatment for adre- nocortical carcinoma. N Engl J Med 2007;356:2372-80.
15. Berruti A, Terzolo M, Sperone P, Pia A, Della Casa S, Gross DJ, et al. Etoposide, doxorubicin and cisplatin plus
mitotane in the treatment of advanced adrenocortical carci- noma: a large prospective phase II trial. Endocr Relat Can- cer 2005;12:657-66.
16. Williams RH, Larsen PR. Williams textbook of endocri- nology. 10th ed. Philadelphia (PA): Saunders; 2003.
DISCUSSION
Dr Barb Miller (Ann Arbor, MI): Your presentation closely parallels the study that recently came out in The Journal of Clinical Endocrinology and Metabolism from the German registry, whose authors also found that a 12-month interval to recurrence was predictive of improved survival with resection, as well as an R0 resec- tion, which again tells everyone how important that first operation is. I have two questions for you.
In your conclusion, you said these were highly selective patients. How do you select those patients? What were your criteria? Do you wait a certain amount of time to figure out who is going to blossom, or do you just go from the first scan where you have a recurrence? My second question is, in those patients who are being debulked for hormone excess, in neuroendocrine tumors we talk about a 90% rule. If you think you can get 90% out, go ahead and do it. But those are in very slow-growing tumors and very different from the aggressive biology of ACC. So in those patients that you debulk, is that a durable result? How long did that last? Did any of them come off their medication or did they still require medication for con- trol? And what was the morbidity rate and mortality rate from your surgery for those reoperations?
Dr Benzon M. Dy: To answer your first question, the se- lection process is a complex one. Our current evaluation process involves multiple disciplines, including endocri- nologists, surgical oncologists, and our thoracic, endo- crine and HPB surgeons. Our first priority is to make sure that we have an appropriate operative candidate that can withstand a reoperation. We also have to define our goals-Is it to have an RO curative resection or is it a palliative operation for someone with some symptoms such as pain or hormonal excess? If they have had a reasonable DFI and localized disease, then we consider it reasonable to go ahead and proceed with an operation. For patients who have an indeterminate recurrence, and we are not certain of the aggressiveness of the tumor biology, in select patients, we have used preoperative eval- uation such as the Berutti protocol to place patients on a chemotherapy regimen, including mitotane, etoposide, doxorubicin, and cisplatin, and then reassess them in three months with imaging. If they have had regression or disease stability, we will offer them a resection of their recurrence or metastasis. If they have developed new distant metastatic lesions or rapid growth of their locore- gional occurrence we would recommend definitive pallia- tive chemotherapy or best medical management at that point. So certainly, a complex workup.
In terms of morbidity and mortality, we did not look specifically at these as end points of our study. There was
limited mortality throughout the series and were used in statistical analysis. However, morbidity was not specif- ically quantified as part of the study.
For patients with preoperative symptoms, there does seem to be a durable cure in terms of their hormonal excess. Those that are debulked will often be able to come off their medications for at least six months to a year.
Dr Richard A. Hodin (Boston, MA): I think I under- stood you to say that the location of the recurrence didn’t matter. Can you specifically comment on patients who had a recurrence in the tumor bed? And also whether external beam radiation was used in any patients.
Dr Benzon M. Dy: We looked at whether patients had either locoregional recurrence versus distant metastases in either the liver, lung, bone, or brain, which were the most common sites of recurrence within our series. The majority of patients, or more than half of patients, had locoregional occurrence. They did not seem to fare any differently compared with patients who had a solitary lung metastasis, as long as we were able to attain an R0 resection.
Dr Richard A. Hodin (Boston, MA): Was radiation used?
Dr Benzon M. Dy: Radiation was used in fewer than five patients throughout the series and did not have an impact on decreasing recurrence. But it’s certainly a small number for us to examine from a statistical standpoint.
Dr Douglas B. Evans (Milwaukee, WI): Just to follow up on Dr Miller’s question, the patients who underwent operation, they received chemotherapy plus-minus mito- tane. Is that correct?
Dr Benzon M. Dy: That’s correct.
Dr Douglas B. Evans (Milwaukee, WI): I think it’s important to note that we should not just charge out and operate on someone with four metastases because they’re 6 months from their primary resection. And then, secondly, the difference between RO and R1, I’m confused as to why you made such a big point of this. How do your pathologists determine an R1, microscopi- cally positive margin? And why would you exclude those patients, since essentially, by definition, the surgeon per- formed a complete gross resection of the tumor? So theoretically, the operative procedure, in the eyes of the surgeon, would be no different between R0 and R1. And what did they actually ink in their specimen, since if you ink the entire specimen, probably everyone will have an R1 resection; right?
Dr Benzon M. Dy: We did not make a specific deter- mination for R1. Our differentiation was R0 versus de- bulking, R2. Interestingly, no patients in our review of the pathology had microscopically positive margins. However, or main goal was to differentiate between pa- tients who had gross disease behind (R2) with goals of palliation versus a complete resection.