Surgery in adrenocortical carcinoma: Importance of national cooperation and centralized surgery
Ilse G. C. Hermsen, MD,a Thomas M. A. Kerkhofs, MD,a Gijsbert den Butter, MD, PhD,b Job Kievit, MD, PhD,“ Casper H. J. van Eijck, MD, PhD,d Els J. M. Nieveen van Dijkum, MD, PhD,e and Harm R. Haak, MD, PhD,a on behalf of Dutch Adrenal Network™, Veldhoven/Eindhoven, Leiden, Rotterdam, and Amsterdam, the Netherlands
Background. The low incidence rate of adrenocortical carcinoma (ACC) requires a multidisciplinary approach in which surgery plays an essential role because complete resection of the primary tumor is the only chance of cure. To improve patient care, insight into operative results within the ACC population is essential. In 2007, a Dutch Adrenal Network Registry was created covering care and outcome of patients treated for ACC in the Netherlands since 1965. Using this database, we performed a study (1) to gain insight into surgical performance in the Netherlands and (2) to compare operative data with international literature.
Methods. Data on patients treated from 1965 until January 2008 were studied. The following data were collected: age, gender, functionality of the tumor, stage at diagnosis, operative procedure, completeness of surgery, disease recurrence, adjuvant mitotane therapy, and recurrence-free and overall survival (OS). Results. A total of 175 patients were studied, of whom 149 underwent surgery. Patients with complete resection had significantly longer OS times than patients with incomplete resection (P = . 010). Patients operated on in a Dutch Adrenal Network center had significantly longer duration of OS in both univariate (P = . 011) and multivariate analysis (P = . 014). A significantly greater OS was observed for operated stage IV patients compared with nonoperated patients (P = . 002).
Conclusion. Our data suggest the relevance of national cooperation and centralized surgery in ACC. For selected patients with stage IV disease, surgery can be beneficial in extending survival. On the basis of the retrospective analysis, operative ACC in the Netherlands can and will be improved. (Surgery 2012;152:50-6.)
From the Departments of Internal Medicine” and Surgery,b Máxima Medical Centre, Veldhoven/Eindhoven, the Netherlands; Department of Surgery,” Leiden University Medical Centre, Leiden, the Netherlands; Department of Surgery,ª Erasmus Medical Centre, Rotterdam, the Netherlands; and Department of Surgery,e Academic Medical Centre, University of Amsterdam, Amsterdam, the Netherlands
ADRENOCORTICAL CARCINOMA (ACC) is a rare disease, with approximately 1-2 new patients diagnosed per
“The other doctors and participating centers of the Dutch Adre- nal Network: E. M. W. Eekhof, MD, PhD, VU University Medical Centre, Amsterdam; B. Havekes, MD, PhD, Maastricht Univer- sity Medical Centre; W. W. de Herder, MD, PhD, and R. A. Feel- ders, MD, PhD, Erasmus Medical Centre, Rotterdam; H. Timmers, MD, PhD, University Medical Centre Nijmegen; M. Kerstens, MD, PhD, University Medical Centre Groningen; J. W. A. Smit, MD, PhD, and A. J. Gelderblom, MD, PhD, Leiden University Medical Centre, Leiden; J. H. de Vries, MD, PhD, Ac- ademic Medical Centre, Amsterdam; and P. M. Zelissen, MD, PhD, University Medical Centre, Utrecht.
Accepted for publication February 9, 2012.
Reprint requests: Ilse G. C. Hermsen, MD, Department of Inter- nal Medicine, Máxima Medical Centre, PO Box 90052, 5600 PD Eindhoven, the Netherlands. E-mail: I.Hermsen@mmc.nl.
0039-6060/$ - see front matter @ 2012 Published by Mosby, Inc.
doi:10.1016/j.surg.2012.02.005
million in population a year.1 The rarity of the dis- ease and its dismal prognosis require a multidisci- plinary approach to each patient to improve patient survival. The role of surgery is essential be- cause complete resection of the tumor without the tumor rupturing is the only chance of cure.2-4 In the Netherlands, most patients initially present to their general practitioner with various symptoms. In our experience three routes of subsequent di- rection are most common: (1) the patient is re- ferred to an university hospital for further work- up and treatment; (2) the patient is referred to a local hospital, is diagnosed with an ACC, and is then referred to a university hospital for surgery and further treatment; (3) the patient is referred to a local hospital and receives all treatment, in- cluding surgery, there. This diversity leads to the distribution of patients with ACC across many hos- pitals, with a subsequent diluting of expertise.
To improve care for patients with ACC and to stimulate scientific research and (inter) national trial participation, the Dutch Adrenal Network (DAN) was set up in 2004. Before 2004, the individual university centers acted as referral cen- ters for difficult and rare diseases, such as ACC. With the foundation of the DAN (consisting of eight university hospitals and one coordinating center), forces were combined and care for pa- tients with ACC was organized.
The DAN has two main goals: (1) to retrospec- tively analyze the quality of care provided to patients with ACC in the Netherlands in the past and identify areas for potential improvement and (2) to prospectively formulate standards of care and monitor the quality and outcomes of the care provided. Prospective standards of care for pa- tients suspected of ACC include multidisciplinary assessment and planning, thorough hormonal work-up and imaging preoperatively, operation by an experienced surgeon, and standardized follow- up and adjuvant treatment in cooperation between endocrinologists and oncologists.
The present study retrospectively reports on the care and health outcomes of patients with ACC who were treated in the Netherlands until 2008, with the following objectives: (1) to gain insight into the relationship between surgical and nonsur- gical treatment on the one hand and health outcomes on the other, and (2) to compare our findings with international literature.
MATERIALS AND METHODS
All DAN centers were visited to gather patient data. These include all eight university hospitals (LUMC Leiden, Erasmus MC Rotterdam, AMC Amsterdam, VU Amsterdam, UMCG Groningen, UMC Utrecht, UMCN Nijmegen, and Maastricht UMC) and one dedicated nonuniversity hospital, the Máxima Medical Centre in Eindhoven/Veld- hoven. Both patients who were operated on in Network hospitals as well as patients who were initially operated in non-DAN hospitals and later directed to a DAN hospital were registered.
Patients and data. From 1965 until January 2008, adult patients (age >16 years) with histologically confirmed ACC from 1965 until 2008 were entered. Per patient, the following data were collected: age, gender, hormonal function of the tumor, stage at diagnosis (European Network for the Study of Adrenal Tumors staging (ENS@T staging; Table I),5 surgical procedure (open or laparoscopic adrenalec- tomy), completeness of surgery (complete resection [R0]; incomplete resection and/or tumor spill during surgery [R]]; macroscopically incomplete
resection [R2]; completeness unknown [Rx]), dis- ease recurrence (defined as recurrence of ACC in patients who had a complete or microscopically incomplete resection of stage I-III disease), disease recurrence (local/metastatic), adjuvant mitotane therapy (mitotane after complete resection), recurrence-free survival (RFS), overall survival (OS), and follow-up until 2010. Because a consensus does not exist regarding operation in patients with stage IV disease, patients who did not undergo an op- eration were also included.
Statistical analysis. Clinical characteristics of the patients were presented as mean with standard deviation (SD) or median with range as appropri- ate. Numbers and percentages were used for proportions in qualitative data. For continuous variables, t test was used. Proportions were com- pared with the use of x2 statistics.
OS and disease-free survival (DFS) were com- puted by use of the Kaplan Meier product limit method. Differences in survival were calculated with the log-rank test. Variables identified as sig- nificant at the P > .05 level on univariate analysis were subjected to multivariate analysis using Cox- regression. Cox-regression analysis was used to correct for age, sex, stage at diagnosis, hormonal function, side, stage at diagnosis, resection status, and at surgery in a DAN or local hospital. Mitotane therapy is considered adequate when plasma levels of 14 mg/L are reached.6 A P < . 05 is considered significant. Statistical analysis was performed using SPSS version 15.0.
RESULTS
Patient characteristics. A total of 175 patients (71 men, 104 women) were included in our anal- ysis. Median age at diagnosis was 49 years (range, 16-79). The majority of patients presented with a functional tumor (54%). Three patients presented with stage I disease, 43 patients with stage II disease, 54 (31%) with stage III disease, and 75 (42%) stage IV.
In all, 149 patients (85%) underwent surgery (Table II). The majority of these patients pre- sented with stage III or IV disease. Eighty-nine pa- tients (60%) were operated on in a DAN center; the remaining 60 patients (40%) were operated on in local hospitals. Surgery was not performed in 26 patients, 23 of whom presented with stage IV disease and 3 with stage III disease. These pa- tients were treated with mitotane, chemotherapy, radiotherapy, or a combination. Median follow-up was 25 months (range, 0-451).
Surgical and adjuvant treatment. Of the 100 patients with stage I-III disease, 97 received
| Adrenocortical carcinoma staging according to ENS@T | |
| T1 | Tumor ≤5 cm |
| T2 | Tumor >5 cm |
| T3 | Tumor infiltration in surrounding tissue, histologically proven |
| T4 | Tumor invasion in adjacent organs or venous tumor thrombus in vena cava or renal vein |
| N0 | Negative lymph nodes |
| N1 | Positive lymph nodes |
| M0 | No distant metastases |
| M1 | Distant metastases |
| Stage I | T1, N0, M0 |
| Stage II | T2, N0, M0 |
| Stage III | T1-2, N1, M0 OR T3-4, N0, M0 |
| Stage IV | T1-4, N0-1, M1 |
| Variable | Dutch Adrenal Network Center (n = 89) | Local (n = 60) P-value | |
|---|---|---|---|
| Gender, n (%) | |||
| Male | 37 (41.6) | 24 (40) | |
| Female | 52 (58.4) | 36 (60) | .848 |
| Age, yr | |||
| Median | 48.7 | 46.1 | |
| Range | 21-79 | 16-71 | .323 |
| Side, (%) | |||
| Right | 47 (52.8) | 26 (43.3) | |
| Left | 39 (43.8) | 34 (56.7) | .178 |
| Hormonal secretion, n (%) | |||
| Absent | 41 (46.1) | 23 (38.3) | |
| Present | 48 (53.9) | 37 (61.7) | .350 |
| Stage at diagnosis (ENS@T), n (%) | |||
| Stage I | 2 (2.2) | 1 (2) | |
| Stage II | 28 (31.5) | 15 (25) | |
| Stage III | 34 (38.2) | 17 (28) | |
| Stage IV | 25 (28.1) | 27 (45) | .199 |
ACC, Adrenocortical carcinoma; ENS@T, European Network for the Study of Adrenal Tumors.
surgical treatment. The operative procedures were performed via the transperitoneal approach, either extended subcostal incision or midline incision, or by retroperitoneal approach by poste- rior incision through the bed of the 11th rib, or a posterolateral flank incision, if necessary with thoraco-abdominal extension.
| R0 (n = 56) | R1 (n = 25) | |
|---|---|---|
| Recurrence, n (%) | ||
| Yes | 33 (59) | 20 (75) |
| No | 19 (34) | 5 (25) |
| Lost to follow-up | 4 (7) | 0 |
| Recurrence location, n (%) | ||
| Local | 21 (64) | 11 (55) |
| Lung | 13 (39) | 9 (45) |
| Liver | 9 (27) | 7 (35) |
| Lymph node | 8 (24) | 5 (25) |
| Bone | 1 (3) | 1 (5) |
| Other | 5 (15) | 4 (20) |
| Recurrence-free | 69 months | 22 months |
| survival | (36-102) | (0-50) |
R0 is defined as microscopically free margins; R1 is defined as microscop- ically incomplete resection and/or tumor spill. ACC, Adrenocortical carcinoma.
R0 resection was achieved in 56 patients (58%), 37 of whom were initially operated on in a DAN center. R1 resection was reported for 25 patients (17 of 64 patients in DAN center, 8 of 33 in local hospitals, P= . 805; 26%). Within this group, tumor spill occurred in 12 patients. R2 resection was noted for only 2 patients, both of whom were oper- ated on in a local hospital. For 14 patients (14%), resection status was unknown (Rx) because of a lack of information in the pathological and/or sur- gical report.
In three stage III patients, surgery was abandoned because of low performance status and because of the patients’ advanced age. Laparoscopic adrenalec- tomy was performed in five patients. Median tumor diameter was 7.0 cm (range, 4.0-9.0).
Of 75 stage IV patients, 52 (70%) underwent operation. The median age of the operated group was 50.1 years (range, 43.1-58.9) compared with 45.6 years (range, 41.3-49.9) in the nonoperated group (P =. 196). The mean number of organs containing adrenal metastases at diagnosis was 2.65 (SD 0.84) in the operated compared with 2.86 (SD 0.97) in the nonoperated group (P = . 330). Functional tu- mors were seen in 37 (71%) of the operated patients and in 12 (55%) of the nonoperated patients (P = .103). Chemotherapy, consisting of mitotane in com- bination with etoposide, doxorubicin and cisplatin, or streptozotocine, was given to surgical as well as nonsurgical patients (31% and 39% of the patients, respectively).
Twenty-nine of 62 patients (47%) received mito- tane after complete resection of the adrenocortical tumor. The duration of therapy ranged from 2 to 24
| Variable | Hazard ratio (95% CI) | Univariate (P) | Hazard ratio (95% | CI) Multivariate (P) |
|---|---|---|---|---|
| Gender | 1.08 (0.75-1.55) | .688 | x ☒ | x ☒ |
| Male | ||||
| Female | ||||
| Age | .020 | .040 | ||
| Side | 1.16 (0.83-1.63) | .376 | x ☒ | x ☒ |
| Right | ||||
| Left | ||||
| Stage at diagnosis (ENS@T) | .000 | .000 | ||
| Stage I vs stage II | 0.94 (0.13-7.13) | 0.52 (0.06-4.35) | ||
| Stage II vs stage III | 0.47 (0.27-0.84) | 1.76 (0.99-3.09) | ||
| Stage III vs stage IV | 0.33 (0.22-0.50) | 2.69 (1.68-4.29) | ||
| Hormonal secretion | 0.70 (0.48-1.02) | .060 | x ☒ | x ☒ |
| Present | ||||
| Absent | ||||
| Resection | ||||
| R0 | 2.55 (1.33-4.89) | .010 | 1.69 (0.82-3.53) | .160 |
| R1 | ||||
| Surgery at Dutch Adrenal Network Center | 2.34 (1.80-3.05) | .011 | 1.74 (1.34-2.26) | .014 |
| Yes No | ||||
ACC, Adrenocortical carcinoma; CI, confidence interval; ENS@T, European Network for the Study of Adrenal Tumors; R0, complete resection; R1, micro- scopically incomplete resection.
months. Therapeutic mitotane levels (≥14 mg/L) were reached in 8 of 29 treated patients. Three patients received adjuvant tumor bed radiation after microscopically incomplete resection. None of these patients developed local recurrence. In 2 of these patients, distant metastases were observed after 4 and 185 months. One patient is still alive 2 years after radiation therapy without evidence of disease.
Cancer recurrence and survival. Of 56 patients with stage I-III disease and complete resection, recurrent disease occurred in 34 (61%; Table III). Twenty-two patients suffered a local recurrence. Median time to recurrence was 28 months (range, 11-45).
Among patients with stage I-III disease and a microscopically incomplete resection, recurrence occurred in 19 patients (76%). In 10 patients (53%), this was a local recurrence. Median time to recurrence was 15 months (range, 9-21).
Table IV shows multiple factors influencing OS of ACC patients. Patients initially operated on in a DAN center had significantly longer survival times than patients operated in local hospitals in both univariate (P= . 011) and multivariate analysis (P= . 014; Fig 1). Patients with a complete resection had significantly longer OS times than patients with an incomplete or unknown resection (P = .010; Fig 2). RFS was different between these 2 groups in univariate analysis (P= . 052).
Among the 8 patients who received adjuvant mitotane treatment reaching therapeutic plasma
levels, median OS was 164 months (range, 84- 244). Among the patients who were treated adjuvantly without reaching therapeutic concen- trations, median OS was 83 months (range, 9-157; P = . 278). In the group who was not treated with mitotane, median OS was 250 months (range, 190-310).
For the adjuvantly treated group, RFS was 69 months (range, 33-105) vs 48 months (range, 0- 109) for patients who did not receive mitotane treatment (P= . 601). Stage IV patients who under- went surgery exhibited significantly longer survival times compared with nonoperated patients (P = .002; Fig 3). Multivariable analysis showed no signif- icant influence of chemotherapy on overall survival (hazard ratio 1.01, 95% confidence interval 0.60- 1.67), whereas surgery does have a significant favor- able effect (hazard ratio 0.46, 95% confidence inter- val 0.27-0.77).
DISCUSSION
Treatment in specialized centers offering multi- disciplinary approach is beneficial to patients with ACC, because improved survival was observed in patients initially operated within the DAN com- pared with patients initially treated in a non-DAN hospital.
Aggressive surgery in this approach is essential because surgery is the only curative option.7,8 Se- ries performed in specialized centers report high percentages (up to 80-90%) of complete resection
Survival Functions
1,0
0,8
Cum Survival
0,6
0,4
0,2
0,0
0
30
60
90
120
150
Time (months)
Dutch Adrenal Network Center n=89
Local hospital n=60
Survival Functions
1.0
0.8
Cum Survival
0.6
0.4-
0,2
RO (n=56) .. “>R1 (n=25) — Rx (n=14)
0.0
0
50
100
150
200
Time (months)
Survival Functions
1.0
0.8
Cum Survival
0,6
0,4
0.2
Surgery (n=52)
0,0
.. ‘“No surgery (n=23)
0
30
60
90
120
150
Time (months)
for patients with stage I-III disease.3,9,10 The rela- tively low percentage of complete resections in the present study (55%) is probably caused by the diffuse distribution of patients over several hos- pitals in the Netherlands.
Recurrent disease has been shown to occur frequently (60-80%) even after complete resec- tion, as observed in our series.11,12 Furthermore, more local recurrences were seen in the R0 group, suggesting wrongfully classified complete resec- tions. Pathological diagnosis of ACC should there- fore be performed by or in consultation with an expert pathologist.
The role of mitotane as an adjuvant therapy after complete surgery remains uncertain, although the impact of adjuvant mitotane on the survival of patients with ACC was recently demonstrated in one retrospective study.13 Improved survival could not be demonstrated in our study, probably because of the small population under study. Furthermore, duration, dose and serum level of mitotane varied substantially between patients making it hard to draw firm conclusions based on these data.
In cases of high risk of recurrence, a ki-67 index >10%, and/or tumor size >8 cm, R1 resection or, in case of tumor spillage, adjuvant mitotane is ad- vised.14 In case of low risk, a ki-67 index <10%, and/ or tumor size <8 cm, mitotane can be discussed, and inclusion in ADIUVO (trial on efficacy of adjuvant
mitotane treatment; www.epiclin.it/adiuvo) must be considered. Although our study was not designed for this purpose, a survival benefit was observed for patients reaching therapeutic plasma concentra- tions in adjuvant mitotane treatment.
The role of surgery in metastatic disease is often disputed.15 Our data indicate longer survival times among patients with stage IV disease who have un- dergone surgery compared with their nonoperated counterparts. Although patients who undergo op- eration may be a selected group of patients, clini- cal characteristics such as age, gender, tumor secretion, and number of metastatic organs did not differ from nonoperated patients. It is there- fore advised to consider in all patients with meta- static disease. However, the final decision to perform surgery should be tailored to the individ- ual patient and should be discussed in a multidisci- plinary team, including an experienced surgeon.
Although this study has resulted in improved insight into our surgical performance and with it improvement in Dutch ACC surgery in the near future, the limitations of this study are acknowl- edged. Selection bias is one of the limitations because the database only contains patients who have been treated in a DAN center at any time during the course of their disease. Patients oper- ated on in local hospitals without consultation with or referral to a DAN center are missing. This group
could comprise patients with a complete resection and an uneventful follow-up. For the same reason, there probably is an underrepresentation of pa- tients with stage I disease. However, whether such a subpopulation exists in the Netherlands is un- known, let alone the number of patients.
Despite these limitations, initiatives like the DAN in rare diseases such as ACC should be stimulated to improve patient care. The collection of prospective data is, of course, essential. For the future, the aim is to improve patient registration and to further centralize and evaluate ACC treatment in the Netherlands. This should be possible by increasing awareness of this disease among doctors in local hospitals and standardizing treatment in DAN centers. Surgery is a key component in this project.
In conclusion, these data suggest the relevance of national cooperation and centralized surgery in ACC. The multidisciplinary approach offered by the DAN centers will improve the care and survival of patients with ACC. Surgery of stage IV disease can be beneficial in extending survival for selected patients. The retrospective analysis presented here, helped setting new standards, and thereby will improve the quality of ACC surgery.
REFERENCES
1. Allolio B, Fassnacht M. Clinical review: adrenocortical carci- noma: clinical update. J Clin Endocrinol Metab 2006;91: 2027-37.
2. Crucitti F, Bellantone R, Ferrante A, Boscherini M, Crucitti P. The Italian Registry for Adrenal Cortical Carcinoma: analysis of a multiinstitutional series of 129 patients. The ACC Italian Registry Study Group. Surgery 1996;119:161-70.
3. Grubbs EG, Callender GG, Xing Y, et al. Recurrence of ad- renal cortical carcinoma following resection: surgery alone
can achieve results equal to surgery plus mitotane. Ann Surg Oncol 2010;17:263-70.
4. Schulick RD, Brennan MF. Long-term survival after com- plete resection and repeat resection in patients with adrenocortical carcinoma. Ann Surg Oncol 1999;6: 719-26.
5. Fassnacht M, Johanssen S, Quinkler M, et al. Limited prog- nostic value of the 2004 International Union Against Can- cer staging classification for adrenocortical carcinoma: proposal for a Revised TNM Classification. Cancer 2009; 115:243-50.
6. Haak HR, Hermans J, van de Velde CJ, et al. Optimal treat- ment of adrenocortical carcinoma with mitotane: results in a consecutive series of 96 patients. Br J Cancer 1994;69: 947-51.
7. Dackiw AP, Lee JE, Gagel RF, Evans DB. Adrenal cortical car- cinoma. World J Surg 2001;25:914-26.
8. Miller BS, Ammori JB, Gauger PG, Broome JT, Hammer GD, Doherty GM. Laparoscopic resection is inappropriate in patients with known or suspected adrenocortical carci- noma. World J Surg 2010;34:1380-5.
9. Gonzalez RJ, Shapiro S, Sarlis N, et al. Laparoscopic resec- tion of adrenal cortical carcinoma: a cautionary note. Sur- gery 2005;138:1078-85.
10. Wangberg B, Khorram-Manesh A, Jansson S, et al. The long- term survival in adrenocortical carcinoma with active surgi- cal management and use of monitored mitotane. Endocr Relat Cancer 2010;17:265-72.
11. Pommier RF, Brennan MF. An eleven-year experience with adrenocortical carcinoma. Surgery 1992;112:963-70.
12. Stojadinovic A, Ghossein RA, Hoos A, et al. Adrenocortical carcinoma: clinical, morphologic, and molecular character- ization. J Clin Oncol 2002;20:941-50.
13. Terzolo M, Angeli A, Fassnacht M, et al. Adjuvant mitotane treatment for adrenocortical carcinoma. N Engl J Med 2007;356:2372-80.
14. Berruti A, Fassnacht M, Baudin E, et al. Adjuvant therapy in patients with adrenocortical carcinoma: a position of an in- ternational panel. J Clin Oncol 2010;28:e401-2.
15. Fassnacht M, Allolio B. Clinical management of adrenocor- tical carcinoma. Best Pract Res Clin Endocrinol Metab 2009; 23:273-89.