Prognostic Parameters of Metastatic Adrenocortical Carcinoma
Guillaume Assié, Guillemette Antoni, Frédérique Tissier, Bernard Caillou, Gwenaelle Abiven, Christine Gicquel, Sophie Leboulleux, Jean-Paul Travagli, Clarisse Dromain, Xavier Bertagna, Jérôme Bertherat, Martin Schlumberger, and Eric Baudin
Service de Médecine Nucléaire et de Cancérologie Endocrinienne (G.As., S.L., J .- P.T., C.D., M.S., E.B.), Département de Santé Publique (G.An.) and Département d’Anatomo-Pathologie (B.C.), Institut Gustave-Roussy, Université Paris XI, 94800 Villejuif, France; Département d’Anatomo-Pathologie (F.T.) and Service d’Endocrinologie (G.Ab., X.B., J.B.), Hôpital Cochin, Université Paris V René Descartes, Faculté de Médecine, Site Cochin-Port-Royal, 75014 Paris, France; Département Endocrinologie-Métabolisme-Cancer (F.T., X.B., J.B.), Institut National de la Santé et de la Recherche Médicale U567 and Centre National de la Recherche Scientifique Unité Mixte de Recherche 8104, Institut Cochin, 75014 Paris, France; and Laboratoire d’Explorations Fonctionnelles Endocriniennes (C.G.), Institut National de la Santé et de la Recherche Médicale U-515, Hôpital Armand Trousseau, Université Paris VI, 75012 Paris, France
Context: Prognostic parameters of metastatic adrenocortical carci- noma (ACC) are poorly characterized.
Objective: The objective of the study was to describe the clinical presentation of metastatic ACC and determine prognostic factors for survival.
Design: This was a retrospective cohort study (1988-2004).
Setting: The study was conducted in an institutional practice.
Patients: Participants included 124 consecutive patients with metastatic ACC, 70 from Gustave-Roussy Institute (main cohort) and 54 patients from the Cochin Hospital (validation cohort). Clin- ical data concerning all patients, histopathologic slides of primary tumors (44 in the main cohort and 40 in the validation cohort), and molecular biology data on 15 primary tumors (main cohort) were analyzed.
Intervention: There was no intervention.
Main Outcome: The main outcome was the specific survival after discovery of the first metastasis (Kaplan-Meier method). This in- cluded univariate analysis on the main cohort, confirmed on the validation cohort and then analyzed in a multivariate analysis.
Results: In the main cohort, overall median survival was 20 months. In univariate analysis, the presence of hepatic and bone metastases, the number of metastatic lesions and the number of tumoral organs at the time of the first metastasis, a high mitotic rate (>20 per 50 high-power field), and atypical mitoses in the primary tumor pre- dicted survival (P = 0.05, 0.003, 0.046, 0.001, 0.01, and < 0.001, respectively). The number of tumoral organs and a high mitotic rate were confirmed on the validation cohort (P = 0.009 and 0.03, respec- tively). These two parameters were confirmed in multivariate anal- ysis (P = 0.0058 and 0.049).
Conclusion: Metastatic ACC is a heterogeneous disease with poor outcome. The combination of the number of tumoral organs at the time of the first metastasis and the mitotic rate can predict different outcomes. (J Clin Endocrinol Metab 92: 148-154, 2007)
A DRENOCORTICAL CARCINOMA (ACC) is a rare dis- ease characterized by an aggressive behavior with 5-yr survival ranging from 16 to 60% (1-12). Surgery remains the only curative modality in ACC patients, and the impact of medical management including 1,1-dichlorodiphenildi- chloroethane (o,p’DDD) and chemotherapy on survival re- mains questionable (13). The disease stage and completeness of initial surgery emerged as major ACC prognostic factors for survival in several studies (1, 2, 4, 5, 7, 8, 10, 11, 14). Survival drops once the tumor spreads outside the adrenal gland as reported in one recent study, in which 5-yr survival is 58-66 and 0-24% in patients with intraadrenal and ex- traadrenal ACC, respectively (8). After curative resection, a
Abbreviations: ACC, Adrenocortical carcinoma; CT, computerized tomography; HPF, high-power field; LOH, loss of heterozygosity; o,p- ‘DDD, 1,1-dichlorodiphenildichloroethane.
JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the en- docrine community.
5-yr survival rate ranging from 11 to 62% has been reported (1-3, 6-8, 10, 14, 15), which compares favorably with sur- vival of less than a year in patients in whom resection was incomplete (1, 2, 8, 14, 16). In addition, other studies sug- gested that the mitotic count or the Ki67 index of the primary tumor should be taken into account to define the 5-yr sur- vival (12, 17-19). Finally, age as well as functioning activity (8, 10, 20) was demonstrated in some (4) but not all studies (5, 7, 14) to be a prognostic factor for survival.
After complete resection of adrenocortical tumors, Gicquel et al. (21) demonstrated that age, Weiss pathological criteria, and 17p13 loss of heterozygosity (LOH) were independent factors for relapse-free survival. Harrison et al. (22) found that combination of parameters including the size of ACC but also the mitotic count and hemorrhage identify subgroups of patients whose 5-yr survival range from 0 to 83%. However, in both studies, stage was either not specified or heterogeneous.
Because stage is a major prognostic parameter, prognostic studies should focus on ACC patients selected according to disease stage. In clinical practice, results of prognostic stud-
ies dealing with patients presenting with various stage at diagnosis are poorly applicable when facing a patient at a given stage. Namely, ACC patients with distant metastases continue to be a major therapeutic challenge. Between 17 and 53% of ACC patients present with distant metastases at the time of the diagnosis (1-12, 23, 24), and their 5-yr survival ranges from 0 to 17% (6-8, 20, 25, 26), with a survival of less than 13 months (1, 3-6). Despite the prognosis of ACC pa- tients presenting with distant metastasis is commonly seen to be heterogeneous, prognostic factors for survival remain unknown.
In an attempt to clarify this question, we conducted a study on clinical, pathological, and biological prognostic factors in ACC patients with distant metastases, followed up in a single center (Gustave-Roussy Institute, first cohort). The main goal was to define prognostic factors for survival in this subgroup of patients. Factors influencing survival were then validated in a second cohort of ACC patients recruited from another French center (Cochin Hospital, second cohort).
Patients and Methods
Patients of the Gustave-Roussy Institute cohort
From January 1988 to April 2003, 76 patients with metastatic ACC followed up at the Gustave-Roussy were enrolled. Six patients were excluded due to uncertainty regarding the diagnosis (two patients) or because they were lost to follow-up (four patients). All histopathological diagnoses were confirmed by a single pathologist (B.C.). Complete clin- ical, biological, and morphological evaluations, including an abdominal ultrasound, a chest and abdomen computerized tomography (CT) scan (conventional and spiral CT since 1998), a chest x-ray, and bone scin- tigraphy, were performed within 4 months of the diagnosis. In three patients, only chest x-rays were repeated, but they remained negative. In three other patients, no bone scan was performed and the initial bone evaluation was deemed negative because of lack of bone pain. Clinically functional tumors were confirmed by appropriate hormonal tests. If tumors were not clinically functional, biological tests include measure- ment of potassium level, plasma and/or urinary excretion of cortisol, and androgen elevation by measuring plasma testosterone and/or an- drostenedione and/or dehydroepiandrosterone sulfate. During the study period, therapeutic management was homogeneous and included primary surgery, first-line systemic o,p’DDD chemotherapy and in case of disease progression, a cisplatinum-based regimen as second-line ther- apy, with or without liver chemoembolization, as previously described (4, 27-29). Various treatment options have been given as a third-line therapy, and reoperation was discussed case by case. Information on clinical follow-up was available for all patients
Patients of the Cochin Hospital cohort
A second cohort of patients originating from Cochin Hospital was analyzed to validate prognostic parameters found significant at univar- iate analysis in the Gustave-Roussy cohort. From April 1988 to August 2004, 60 patients with metastatic ACC followed up at the Cochin Hos- pital were enrolled in this study. Six patients were excluded due to missing data. Among the 54 remaining patients, histopathological di- agnostic was confirmed for 52 by a single pathologist (F.T.), and for the remaining two patients, the diagnosis was confirmed by the presence of Cushing’s syndrome associated with evidence of metastatic spread. As recently described, complete clinical, biological, and morphological evaluations were performed, including systematic chest and abdomen CT scan and bone scintigraphy depending on clinical presentation (30). During the study period, therapeutic management was homogeneous and included primary surgery and systemic o,p’DDD chemotherapy as a first-line chemotherapy. Various therapeutic options were given as a second line. Clinical follow-up was available for all patients.
Pathological examination
For the Gustave-Roussy cohort, pathological slides from 44 patients were reviewed by a single pathologist (B.C.). Hematoxylin and eosin- stained slides of the primary tumor were available and systematic evalu- ation of 11 pathological criteria was performed, including the nine Weiss pathological criteria as previously defined (17) and two additional criteria: intratumor hemorrhage and nuclear cytoplasmic pseudoinclusions. The median number of slides per patient was three (range from 1 to 10).
For the Cochin cohort, pathological slides of 40 patients were re- viewed by one pathologist (F.T.). The nine Weiss pathological criteria were recorded.
The size of the primary tumor was recorded taking into account the size of the tumor at gross pathological examination.
Molecular biology analysis
In the Gustave-Roussy cohort, primary adrenal tumor total RNA and DNA were extracted and purified, as previously described (31). The allelic status of the 17p13 and 11p15 loci were determined in 13 and 15 patients, respectively, by Southern blotting and PCR. IGF-II overexpres- sion was determined, in 15 patients, by dot-blot quantification, as pre- viously described (21). The presence of IGF-II overexpression was de- fined as a ratio 10-fold above that of the normal adrenal gland.
Informed consent was obtained from each patient.
Statistical analysis
In the Gustave-Roussy cohort, the following clinical criteria were recorded in all patients (see Table 1 for classes): gender, age, size of the primary, the MacFarlane stage at initial diagnosis (see Table 1 footnotes), tumor side, renal or inferior vena cava tumor thrombus, hormonal or cortisol secretion, time between the diagnosis of the primary tumor and the first metastasis, location of tumoral organs, total number of tumoral organs (including the primary or a locoregional recurrence, liver, lung, bone, other), and total number of metastases. As detailed above, 11 pathological criteria were recorded in 44 patients and the biological parameters (analyzed as present or absent) in 13-15 patients.
The primary end point was specific survival between the diagnosis of the first metastasis and the end of the study or the patient’s death. Correlation was sought between the above-mentioned criteria and spe- cific survival.
The predictive factors for survival identified within the Gustave- Roussy cohort were validated on the Cochin cohort in univariate anal- ysis. Factors validated in univariate analysis were further tested in multivariate analysis in the Gustave-Roussy cohort. Survival rates and their 95% confidence intervals were estimated by the Kaplan-Meier method (32) and the Rothman method (33). The log-rank test (34) was used to compare survival curves (univariate analysis). The relative risks (95% confidence interval) were estimated using Cox’s proportional haz- ards regression model (35) (univariate and multivariate analysis). All tests were two sided.
Results
Clinical features of the Gustave-Roussy cohort (Table 1)
Seventy patients with metastatic ACC (21 males and 49 females; median age 43.6 yr; range 18-69 yr) constitute the study population. The main patient characteristics are de- tailed in Table 1. Briefly, primary tumors were smaller than 10 cm in 21 patients (30%) including four tumors measuring 5 cm or less; at diagnosis the MacFarlane stage was IV in 36 patients (51%). Secretory activity was demonstrated in 50 patients (71%), consisting of Cushing’s syndrome in 38 (54%). The median time between the diagnosis of the primary tumor and the discovery of the first metastasis was 12.3 months (range 0-100 months), being less than 4 months in the ma- jority of patients (33 of 70: 47%). Of note, 80 and 90% of metastases were discovered within 17 and 32 months, re- spectively, after the diagnosis of the primary tumor. At the
| Variables | n | 2-yr survival (%) | 5-yr survival (%) | Unadjusted relative risk (95% CI) | P value (univariate) |
|---|---|---|---|---|---|
| Gender | |||||
| Male | 21 | 41 | 7 | 1 | |
| Female | 49 | 41 | 17 | 0.86 | 0.65 |
| Age at first metastasis | |||||
| 45 yr or younger | 39 | 51 | 10 | 1 | |
| Older than 45 yr | 31 | 29 | 13 | 1.33 (0.69-2.56) | 0.55 |
| Size of the primary tumor | |||||
| ≤10 cm | 21 | 30 | 10 | 1 | |
| 11-15 | 28 | 54 | 17 | 0.61 (0.32-1.17) | 0.14 |
| >15 cm | 21 | 37 | 8 | 0.82 (0.42-1.62) | 0.54 |
| MacFarlane staged | |||||
| <4 | 34 | 40 | 11 | 1 | |
| 4 | 36 | 43 | 12 | 1.00 (0.57-1.73) | 0.99 |
| Side | |||||
| Right | 35 | 42 | 6 | 1 | |
| Left | 35 | 40 | 17 | 0.96 (0.55-1.66) | 0.88 |
| Thrombus in renal or inferior vena cava | |||||
| No | 53 | 41 | 13 | 1 | |
| Yes | 17 | 42 | 9 | 1.10 (0.60-2.04) | 0.76 |
| Cortisol secretion | |||||
| No | 32 | 57 | 21 | 1 | |
| Yes | 38 | 33 | 9 | 1.32 (0.75-2.34) | 0.34 |
| Time between initial diagnosis and first metastasis | |||||
| ≤4 months | 33 | 44 | 13 | 1.19 (0.61-2.31) | 0.60 |
| 4-12 months | 18 | 26 | 1.77 (0.80-3.92) | 0.16 | |
| >12 months | 19 | 49 | 11 | 1 | |
| Liver metastases | |||||
| No | 29 | 54 | 24 | 1 | |
| Yes | 41 | 32 | 5 | 1.77 (1.00-3.15) | 0.05ª |
| Lung metastases | |||||
| No | 23 | 48 | 9 | 1 | |
| Yes | 47 | 38 | 14 | 0.95 (0.54-1.69) | 0.86 |
| Bone metastases | |||||
| No | 61 | 47 | 14 | 1 | |
| Yes | 09 | 0 | 0 | 3.20 (1.43-7.20) | 0.003b |
| Presence of an adrenal tumor | |||||
| No | 17 | 45 | 19 | 1 | |
| Yes | 53 | 40 | 11 | 1.54 (0.77-3.07) | 0.22 |
| No. of tumoral organs (including adrenals) | |||||
| 1 | 10 | 63 | 25 | 1 | |
| 2 | 36 | 53 | 16 | 1.85 (0.71-4.83) | 0.21 |
| >2 | 24 | 11 | 5.44 (1.97-15.05) | 0.001€ | |
| No. of metastatic lesions (excluding adrenal lesions) | |||||
| 1 | 12 | 73 | 32 | 1 | |
| 2-5 | 20 | 48 | 15 | 1.288 (0.49-3.33) | 0.61 |
| >5 | 34 | 26 | 6 | 2.45 (1.02-5.94) | 0.046ª |
| Mitotic count | |||||
| <20 | 28 | 51 | 19 | 1 | |
| >20 | 16 | 29 | 0 | 2.53 (1.22-5.27) | 0.01b |
| Atypical mitoses | |||||
| No | 31 | 55 | 19 | 1 | |
| Yes | 13 | 9 | 0 | 4.44 (2.03-9.70) | <0.001c |
| Capsular invasion | |||||
| No | 6 | 50 | 17 | 1 | |
| Yes | 38 | 41 | 13 | 1.56 (0.59-4.11) | 0.37 |
| Clear cells | |||||
| ≤25% | 26 | 43 | 7 | 1 | |
| <25% | 18 | 43 | 19 | 1.15 (0.57-2.27) | 0.70 |
| Necrosis | |||||
| No | 5 | 30 | 0 | 1 | |
| Yes | 39 | 44 | 15 | 0.85 (0.30-2.44) | 0.77 |
| Venous embolism | |||||
| No | 23 | 35 | 17 | 1 | |
| Yes | 21 | 50 | 9 | 0.95 (0.48-1.89) | 0.88 |
| Diffuse architecturee | |||||
| No | 3 | ||||
| Yes | 41 |
| Variables | n | 2-yr survival (%) | 5-yr survival (%) | Unadjusted relative risk (95% CI) | P value (univariate) |
|---|---|---|---|---|---|
| Cytonuclear grade | |||||
| 1 or 2 | 17 | 46 | 8 | 1 | |
| 3 or 4 | 27 | 40 | 18 | 0.84 (0.42-1.66) | 0.61 |
| Intratumor hemorrhage | |||||
| No | 31 | 44 | 14 | 1 | |
| Yes | 3 | 37 | 12 | 1.34 (0.64-2.82) | 0.43 |
| Nuclear pseudoinclusions | |||||
| No | 30 | 43 | 17 | 1 | |
| Yes | 14 | 39 | 8 | 1.15 (0.56-2.35) | 0.70 |
| IGF-II overexpression | |||||
| No | 6 | 33 | 0 | 1 | |
| Yes | 9 | 56 | 17 | 0.42 | 0.15 |
| 17p13 LOH | |||||
| No | 5 | 40 | 0 | 1 | |
| Yes | 8 | 31 | 16 | 1.05 | 0.94 |
| 11p15 LOH | |||||
| No | 4 | 50 | 0 | 1 | |
| Yes | 11 | 42 | 16 | 0.72 | 0.60 |
Results of the univariate analysis of specific survival are shown. CI, Confidence interval.
” P ≤ 0.05.
b P ≤ 0.01.
c P ≤ 0.001.
d MacFarlane stage I (T1 NO, MO), II (T2 NO, MO), III (T3 MO), IV (T4 MO or MI).
e Survival could not be determined in patients without diffuse architecture (too few).
time of study entry, tumoral organs included the liver (41 patients; 59%), lung (47 patients; 67%), bone (nine patients; 13%), peritoneum (eight patients; 11%), and miscellaneous organs (nine patients; 13%) including 10 lesions (two ovaries, one mediastinal lymph node, one neck lymph node, one skin, one brain, two thyroid, one pancreas, one spleen). Metastases confined to a single organ involved the liver, lungs, and bone in 19 (27%), 22 (31%), and no patients, respectively. An ad- renal tumor was present when metastatic disease was dis- covered in 53 patients (76%), including 17 locoregional re- currences (24%). The majority of patients experienced two tumoral organs at presentation and more than five metastatic locations. Of note, in four patients the exact number of me- tastases could not be assessed.
The median follow-up since the diagnosis of the first me- tastasis, calculated according to the inverse Kaplan-Meier method, was 51 months.
The primary tumor was resected in 67 patients and was macroscopically complete in 62 (89%). Three patients did not undergo resection of the primary because of advanced met- astatic disease. All patients received o,p’DDD as first-line medical therapy as either adjuvant (35 patients) or palliative (35 patients) treatment. Of the 70 patients, 62 (89%) received cisplatin-based second-line chemotherapy, and 45 of them were treated with an etoposide-cisplatin regimen. Liver che- moembolization was performed in 12 patients. Twenty-five patients underwent reoperation for the primary tumor (11 patients), metastases (13 patients), or both (1 patient).
Pathological features of the Gustave-Roussy cohort (Table 1)
The pathological features of primary ACC from the 44 patients of the Gustave-Roussy cohort are detailed in Table 1. Briefly, the mitotic count exceeded 20 per high-power field (HPF) in 16 (36%), and atypical mitoses were seen in 13 patients (30%). Capsular invasion, a low level of clear cells,
necrosis, venous embolism, and a diffuse architecture were identified in 38 (86%), 18 (41%), 39 (89%), 21 (48%), and 41 (93%) tumors, respectively. Finally, intratumor hemorrhage and intranuclear cytoplasmic pseudoinclusions were found in 13 (30%) and 14 (32%) tumors, respectively.
17p13 and 11p15 allelic status and IGF-II mRNA tumor content in the Gustave-Roussy cohort (Table 1)
Biological results are given in Table 1. Overexpression of the IGF-II gene was found in nine of 15 tumors analyzed (60%). 17p13 LOH was found in eight of 13 tumors ana- lyzed (62%), and 11p15 LOH was found in 11 of 15 tumors analyzed (74%). Of note, as previously reported (31), eight of the nine patients with documented IGF-II expression had an uniparental disomy in 11p15.
Outcome data in the Gustave-Roussy cohort
The 1-, 2-, and 5-yr survival rates were 72% (61-82%), 41% (30%-54%), and 12% (5-25%). The overall median survival for the entire cohort was 20 months. At the time of the last follow-up, 52 (74%) patients had died of their disease.
Prognostic factors for survival in the Gustave-Roussy cohort (univariate analysis; Table 1)
Liver metastases (P = 0.05), bone metastases (P = 0.003), more than five metastatic lesions (P = 0.04), more than two tumoral organs (P = 0.001), a high mitotic count (>20/50 HPF; P = 0.01), and presence of atypical mitoses (P < 0.001) were identified as predictive factors for survival at univariate analysis. Of note, we found a trend toward a better outcome in patients whose metastases were diagnosed more than 1 yr after resection of the primary ACC, compared with those who developed a recurrence before this date (2-yr survival of 49 and 26%, respectively, P = 0.16). The number of met-
astatic lesions and number of tumoral organs (trend test: P = 0.008) as well as mitotic count and atypical mitoses (x test: P = 0.0058) were highly intricated. These parameters were further explored for the validation study. Others parameters were found not significantly related with survival.
Validation of the prognostic factors for survival in the Cochin cohort (univariate analysis; Table 2)
The median follow-up since the diagnosis of the first me- tastasis was 36.7 months. The primary tumor was resected in all patients except for four patients because of advanced metastatic disease. The overall median survival for the entire cohort was 14 months.
In the Cochin cohort, more than two tumoral organs at the time of the first metastasis and more than 20 mitoses per 50 HPF in the primary tumor were confirmed as deleterious predictors for survival in univariate analysis (P = 0.009 and 0.03, respectively; Table 2). The number of metastatic lesions could not be tested.
Multivariate analysis of the validated prognostic factors in the Gustave-Roussy cohort (Table 3)
The number of tumoral organs (P = 0.0058) and the num- ber of mitoses (P = 0.049) showed a significant effect on survival on multivariate analysis.
Based on the number of tumoral organs (one or two and more than two) and the number of mitoses (less than 20 and more than 20), a model is proposed to predict survival (Fig. 1): a 2.8-, 1.2-, and 0.8-yr median survival was observed in ACC patients with none, one, and two deleterious prognostic factors, the 5-yr survival being 25, 0, and 0%, respectively (P = 0.0008).
Objective responses to o,p’DDD- or cisplatinum-based chemotherapy in ACC patients with none, one, and two deleterious prognostic factors were observed in three of 21 (14%), three of 14 (21%), and none of nine patients, respec- tively (trend test: non significant).
Discussion
To our knowledge, this study is the first prognostic study focusing on the survival of ACC patients with distant metas-
| Variables | n | Unadjusted relative risk (95% CI) | P value (univariate) |
|---|---|---|---|
| No. of tumoral organs (including adrenals) | |||
| 1 | 7 | 1 | |
| 2 | 26 | 3.49 (0.82-14.9) | 0.09 |
| 3 | 20 | 7.19 (1.64-31.57) | 0.009ª |
| Mitotic count | |||
| <20 | 31 | 1 | |
| >20 | 9 | 2.46 (1.10-5.53) | 0.03b |
| Atypical mitoses | |||
| No | 30 | 1 | |
| Yes | 10 | 1.03 (0.45-2.33) | 0.95 |
CI, Confidence interval.
ª P ≤ 0.01.
b P ≤ 0.05.
| Gustave-Roussy cohort | |||
|---|---|---|---|
| n | Adjusted relative risk (95% CI) | P value | |
| No. of tumoral organs (including adrenals) | |||
| 1-2 | 8 | 1 | |
| >2 | 23 | 3.83 (1.45-10.11) | 0.0058ª |
| Mitotic count | |||
| <20 | 28 | 1 | |
| >20 | 16 | 2.26 (1.00-5.19) | 0.049b |
CI, Confidence interval.
@ P ≤ 0.01.
b P ≤ 0.05.
tases. The rationale for aggressive chemotherapy and deciding to include a patient in a trial as well as reoperation are the usual questions haunting physicians facing patients with metastatic ACC. The latter may benefit from the present study.
Our study has several limitations: it is a retrospective study, performed on a limited number of patients, and only a median of three slides were reviewed to assess the patho- logical status, and pathological and biological data were analyzed in only 44 and 15 ACC patients. Finally, we cannot extensively rule out the impact of therapeutic interventions on our results. Some concessions can be made for these limited data given the rarity of the disease, and the homo- geneous imaging and therapeutic management of ACC pa- tients in a single referral cancer center during the study period. Moreover, the reliability of the study is greatly in- creased by the use of an independent cohort of patients for the validation of the results. Concerning the number of slides, only representative slides, as judged by the first pa- thologist, were reviewed for Weiss score determination. However, we cannot totally exclude an underestimation of the pathological Weiss score.
The clinical characteristics of patients with metastatic ACC deserve several remarks. The largest diameter of the primary tumor was less than 5 cm in 6% of patients and less than five mitoses per 50 HPF were found in 32% of patients, suggesting that these two parameters should not be regarded as reliable
Survival rate
1.00
0.90
Gustave-Roussy
0.80
logrank : chi2 ( 2 df) = 14.25 , p=0.0008
0.70
0.60
No risk factor
0.50
1 risk factor
0.40
2 risk factors
0.30
0.20
0.10
0.00
Years
0
1
2
3
4
5
6
7
8
9
single criteria of benign disease. The majority of ACC pa- tients had two tumoral organs and more than five metasta- ses, and the liver and lung were the most frequently affected organs, which is in accordance with a recent review (36). The 5-yr survival of 12% and a median survival of 20 months compare favorably with previous studies of patients with unresectable ACC including metastatic ACC (1, 3-8, 25, 26). Whether this slightly improved outcome is related to better therapeutic management, earlier diagnosis of metastatic dis- ease, population heterogeneity, or better supportive care re- mains to be established.
Our study highlights the number of tumoral organs at the time of the first metastasis as a major predictive factor for survival of patients with metastatic ACC. This result sug- gests that the tumor burden has a higher prognostic value than the location of any involved organ (lung, bone, liver) including the primary. Indeed, the lower survival of patients with hepatic or bone metastasis we observed in univariate analysis was actually related to a higher tumor burden. In particular, all patients with bone metastases had metastatic disease in other organs. Furthermore, no difference in sur- vival between isolated lung or liver metastases patients were found in this study (data not shown). Locoregional exten- sion, presence of a thrombus, or functional status were not identified as major prognostic parameters in metastatic ACC patients, supporting that an active medical or surgical control of these ACC characteristics is not an overriding priority in comparison with tumor burden control and therefore the use of active cytotoxic agents (4, 8, 10, 12, 20, 30).
The pathological analysis of adrenal tumors was first in- troduced as a diagnostic tool for ACC (18, 37, 38). The Weiss score is the most widely used. From a prognostic point of view, the mitotic index or the Ki-67 staining index has emerged as single parameters correlated with survival in various studies (11, 12, 17, 19, 39, 40). However, in only two studies was the influence of the stage concomitantly ana- lyzed (12, 40), the proliferative index remaining significantly associated with survival in one (12). Finally, none of these studies determined whether the prognostic influence of the mitotic count was applicable in the subgroup of ACC pre- senting with distant metastasis. Our study highlights that both a high number of mitoses and the presence of atypical mitoses could predict survival of metastatic ACC patients and that these two parameters were highly correlated in the Gustave-Roussy cohort. However, in the validation cohort, only the number of mitoses, and not the presence of atypical mitoses, could predict survival in metastatic ACC patients. This discrepancy may be explained by either a random effect in our cohort or a lack of reproducibility of atypical mitoses diagnosis between the two pathologists. Atypical mitoses have also been considered as a significant diagnostic param- eter in a one-score pathological system (37), associated with a low disease-free survival in a univariate analysis in another study (12), and identified as an important predictor of ma- lignancy in childhood adrenocortical tumors (41). The po- tential biological relevance of atypical mitoses in these pa- tients remains to be understood.
Together with clinical and pathological criteria, we ad- dressed the question of whether the IGF-II expression and 17p13 and 11p15 LOH play a role in the prediction of out-
come in a limited number of patients with metastatic ACC. IGF-II overexpression and 11p15 locus rearrangement are frequently observed in sporadic ACC (31, 42). Using DNA microarray analysis, two recent studies underlined that IGF-II overexpression is a major diagnostic target in sporadic ACC (43, 44). Also, from a diagnostic standpoint, 17p13 LOH was found to be an independent prognostic marker of re- currence (21). Furthermore, IGF-II overexpression and 17p13 and 11p15 LOH were found to be correlated with Weiss criteria, tumor stage and vascular endothelial growth factor expression (21, 42, 45). However, in the present study, none of these biological parameters were found to be correlated with survival. Despite the fact that we cannot rule out that the lack of prognostic value may be related to the low number of samples, these data may suggest an early initiating rather than a progressive role of these markers in the carcinogenic process in ACC. As recently emphasized, it is one major goal of the recently created European Network for the study of Adrenal Tumors network to gather an adequate number of ACC tissue samples which, in the field of prognostic studies, may turn into the discovery of putative therapeutic targets (46).
Finally, at multivariate analysis, the number of tumoral organs at the time of first metastasis and the mitotic index in the primary were found to be the two major predictors of survival in patients with metastatic ACC. By combining these parameters, different outcomes can be predicted. In partic- ular, patients with less than two tumoral organs and less than 20 mitoses per 50 HPF in the primary tumor constitute a favorable subgroup of metastatic patients associated with a 25%, 5-yr-specific survival. In this subgroup, a repeated sur- gical resection may be appropriate. In addition, previous studies suggest that the ability to perform a complete resec- tion is a critical parameter that is associated to a better out- come (3,7, 15). Finally, the interval between primary surgery and the diagnosis of the first metastasis may also help the physician to decide between a general or a locoregional ther- apeutic approach. Conversely, in patients with at least one pejorative prognostic factor, we suggest that a systemic med- ical control of patients with metastatic ACC should be an overriding priority before locoregional therapeutic manage- ment such as a repeated surgical resection. The questionable value of medical management, including o,p’DDD- and cis- platinum-based chemotherapy indicates that new drugs are urgently needed in ACC patients.
Our study highlights for the first time prognostic factors for survival of patients with metastatic ACC. The number of tumoral organs together with a high mitotic count in the primary tumor may serve to divide a heterogeneous popu- lation of patients into smaller subgroups with more predict- able outcomes. Results of therapeutic trials should take into account the heterogeneity of patients with metastatic ACC.
Acknowledgments
We thank Dr. Laplanche for statistical advice, Professor Plouin for the coordination of the “COMETE” (COrtico et MEdullosurrénale Tumeurs Endocrines) network, the nurses of our departments, Dr. Zbuk for re- viewing English language, and the following pathologists: Drs. Amat, Vetter, Hoang, Delcourt, Capron, Molinié, Anger, Prevot, Hem, Duch- atelle, Trouette, Spiekermann, Morcellet, Guettier, and Plenat.
Received March 31, 2006. Accepted October 13, 2006.
Address all correspondence and requests for reprints to: E. Baudin, Institut Gustave-Roussy, 39 rue Camille Desmoulins, 94805 Villejuif Cedex, France. E-mail: baudin@igr.fr.
The “COMETE” network was supported by PHRC Grant AOM02068 and grants from INSERM and Ministère Délégué à la Recherche et des Nouvelles Technologies.
Disclosure Statement: The authors have nothing to disclose.
References
1. Henley DJ, van Heerden JA, Grant CS, Carney JA, Carpenter PC 1983 Ad- renal cortical carcinoma-a continuing challenge. Surgery 94:926-931
2. Soreide JA, Brabrand K, Thoresen SO 1992 Adrenal cortical carcinoma in Norway, 1970-1984. World J Surg 16:663-667; discussion 668
3. Pommier RF, Brennan MF 1992 An eleven-year experience with adrenocor- tical carcinoma. Surgery 112:963-970; discussion 970-971
4. Luton JP, Cerdas S, Billaud L, Thomas G, Guilhaume B, Bertagna X, Laudat MH, Louvel A, Chapuis Y, Blondeau P, Bonnin A, Bricaire H 1990 Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mitotane therapy. N Engl J Med 322:1195-1201
5. Crucitti F, Bellantone R, Ferrante A, Boscherini M, Crucitti P 1996 The Italian Registry for Adrenal Cortical Carcinoma: analysis of a multiinstitutional series of 129 patients. The ACC Italian Registry Study Group. Surgery 119:161-170
6. Barzon L, Fallo F, Sonino N, Daniele O, Boscaro M 1997 Adrenocortical carcinoma: experience in 45 patients. Oncology 54:490-496
7. Schulick RD, Brennan MF 1999 Long-term survival after complete resection and repeat resection in patients with adrenocortical carcinoma. Ann Surg Oncol 6:719-726
8. Icard P, Goudet P, Charpenay C, Andreassian B, Carnaille B, Chapuis Y, Cougard P, Henry JF, Proye C 2001 Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French Association of Endocrine Surgeons study group. World J Surg 25:891-897
9. Favia G, Lumachi F, D’Amico DF 2001 Adrenocortical carcinoma: is prognosis different in nonfunctioning tumors? results of surgical treatment in 31 patients. World J Surg 25:735-738
10. Kendrick ML, Lloyd R, Erickson L, Farley DR, Grant CS, Thompson GB, Rowland C, Young Jr WF, van Heerden JA 2001 Adrenocortical carcinoma: surgical progress or status quo? Arch Surg 136:543-549
11. Vassilopoulou-Sellin R, Schultz PN 2001 Adrenocortical carcinoma. Clinical outcome at the end of the 20th century. Cancer 92:1113-1121
12. Stojadinovic A, Ghossein RA, Hoos A, Nissan A, Marshall D, Dudas M, Cordon-Cardo C, Jaques DP, Brennan MF 2002 Adrenocortical carcinoma: clinical, morphologic, and molecular characterization. J Clin Oncol 20:941-950
13. Latronico AC, Chrousos GP 1997 Extensive personal experience: adrenocor- tical tumors. J Clin Endocrinol Metab 82:1317-1324
14. Lee JE, Berger DH, el-Naggar AK, Hickey RC, Vassilopoulou-Sellin R, Gagel RF, Burgess MA, Evans DB 1995 Surgical management, DNA content, and patient survival in adrenal cortical carcinoma. Surgery 118:1090-1098
15. Bellantone R, Ferrante A, Boscherini M, Lombardi CP, Crucitti P, Crucitti F, Favia G, Borrelli D, Boffi L, Capussotti L, Carbone G, Casaccia M, Cavallaro A, Del Gaudio A, Dettori G, Di Giovanni V, Mazziotti A, Marrano D, Masenti E, Miccoli P, Mosca F, Mussa A, Petronio R, Piat G, Ruberti U, Serio G, Marazano L, Alexander R, Akerstrom G, Skogseid MB 1997 Role of reoperation in recurrence of adrenal cortical carcinoma: results from 188 cases collected in the Italian National Registry for Adrenal Cortical Carcinoma. Surgery 122:1212-1218
16. Zografos GC, Driscoll DL, Karakousis CP, Huben RP 1994 Adrenal adeno- carcinoma: a review of 53 cases. J Surg Oncol 55:160-164
17. Weiss LM, Medeiros LJ, Vickery Jr AL 1989 Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 13:202-206
18. van Slooten H, Schaberg A, Smeenk D, Moolenaar AJ 1985 Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 55: 766-773
19. Terzolo M, Boccuzzi A, Bovio S, Cappia S, De Giuli P, Ali A, Paccotti P, Porpiglia F, Fontana D, Angeli A 2001 Immunohistochemical assessment of Ki-67 in the differential diagnosis of adrenocortical tumors. Urology 57:176-182
20. Berruti A, Terzolo M, Sperone P, Pia A, Casa SD, Gross DJ, Carnaghi C, Casali P, Porpiglia F, Mantero F, Reimondo G, Angeli A, Dogliotti L 2005 Etoposide, doxorubicin and cisplatin plus mitotane in the treatment of ad- vanced adrenocortical carcinoma: a large prospective phase II trial. Endocr Relat Cancer 12:657-666
21. Gicquel C, Bertagna X, Gaston V, Coste J, Louvel A, Baudin E, Bertherat J, Chapuis Y, Duclos JM, Schlumberger M, Plouin PF, Luton JP, Le Bouc Y 2001 Molecular markers and long-term recurrences in a large cohort of patients with sporadic adrenocortical tumors. Cancer Res 61:6762-6767
22. Harrison LE, Gaudin PB, Brennan MF 1999 Pathologic features of prognostic
significance for adrenocortical carcinoma after curative resection. Arch Surg 134:181-185
23. Wooten MD, King DK 1993 Adrenal cortical carcinoma. Epidemiology and treatment with mitotane and a review of the literature. Cancer 72:3145-3155
24. Wajchenberg BL, Albergaria Pereira MA, Medonca BB, Latronico AC, Cam- pos Carneiro P, Alves VA, Zerbini MC, Liberman B, Carlos Gomes G, Kirschner MA 2000 Adrenocortical carcinoma: clinical and laboratory obser- vations. Cancer 88:711-736
25. Sullivan M, Boileau M, Hodges CV 1978 Adrenal cortical carcinoma. J Urol 120:660-665
26. Bodie B, Novick AC, Pontes JE, Straffon RA, Montie JE, Babiak T, Sheeler L, Schumacher P 1989 The Cleveland Clinic experience with adrenal cortical carcinoma. J Urol 141:257-260
27. Schlumberger M, Brugieres L, Gicquel C, Travagli JP, Droz JP, Parmentier C 1991 5-Fluorouracil, doxorubicin, and cisplatin as treatment for adrenal cortical carcinoma. Cancer 67:2997-3000
28. Bonacci R, Gigliotti A, Baudin E, Wion-Barbot N, Emy P, Bonnay M, Cailleux AF, Nakib I, Schlumberger M 1998 Cytotoxic therapy with etoposide and cisplatin in advanced adrenocortical carcinoma. Reseau Comete INSERM. Br J Cancer 78:546-549
29. Baudin E, Pellegriti G, Bonnay M, Penfornis A, Laplanche A, Vassal G, Schlumberger M 2001 Impact of monitoring plasma 1,1-dichlorodiphenildi- chloroethane (o,p’DDD) levels on the treatment of patients with adrenocortical carcinoma. Cancer 92:1385-1392
30. Abiven G, Coste J, Groussin L, Anract P, Tissier F, Legmann P, Dousset B, Bertagna X, Bertherat J 2006 Clinical and biological features in the prognosis of adrenocortical cancer: poor outcome of cortisol-secreting tumors in a series of 202 consecutive patients. J Clin Endocrinol Metab 91:2650-2655
31. Gicquel C, Bertagna X, Schneid H, Francillard-Leblond M, Luton JP, Girard F, Le Bouc Y 1994 Rearrangements at the 11p15 locus and overexpression of insulin-like growth factor-II gene in sporadic adrenocortical tumors. J Clin Endocrinol Metab 78:1444-1453
32. Kaplan EL, Meier P 1958 Nonparametric estimation from incomplete obser- vations. J Am Stat Assoc 53:457-481
33. Rothman KJ 1978 Estimation of confidence limits for the cumulative proba- bility of survival in life table analysis. J Chronic Dis 31:557-560
34. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG 1977 Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. analysis and examples. Br J Cancer 35:1-39
35. Cox DR 1972 Regression models and life-tables. J R Statist Soc 34:187-220
36. Allolio B, Hahner S, Weismann D, Fassnacht M 2004 Management of adre- nocortical carcinoma. Clin Endocrinol (Oxf) 60:273-287
37. Weiss LM 1984 Comparative histologic study of 43 metastasizing and non- metastasizing adrenocortical tumors. Am J Surg Pathol 8:163-169
38. Hogan TF, Gilchrist KW, Westring DW, Citrin DL 1980 A clinical and patho- logical study of adrenocortical carcinoma: therapeutic implications. Cancer 45:2880-2883
39. Goldblum JR, Shannon R, Kaldjian EP, Thiny M, Davenport R, Thompson N, Lloyd RV 1993 Immunohistochemical assessment of proliferative activity in adrenocortical neoplasms. Mod Pathol 6:663-668
40. Evans HL, Vassilopoulou-Sellin R 1996 Adrenal cortical neoplasms. A study of 56 cases. Am J Clin Pathol 105:76-86
41. Wieneke JA, Thompson LD, Heffess CS 2003 Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol 27:867-881
42. Gicquel C, Raffin-Sanson ML, Gaston V, Bertagna X, Plouin PF, Schlum- berger M, Louvel A, Luton JP, Le Bouc Y 1997 Structural and functional abnormalities at 11p15 are associated with the malignant phenotype in spo- radic adrenocortical tumors: study on a series of 82 tumors. J Clin Endocrinol Metab 82:2559-2565
43. Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL, Sanders D, Aljundi RT, Gauger PG, Thompson NW, Taylor JM, Hanash SM 2003 Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am J Pathol 162:521-531
44. de Fraipont F, El Atifi M, Cherradi N, Le Moigne G, Defaye G, Houlgatte R, Bertherat J, Bertagna X, Plouin PF, Baudin E, Berger F, Gicquel C, Chabre O, Feige JJ 2005 Gene expression profiling of human adrenocortical tumors using complementary deoxyribonucleic acid microarrays identifies several candidate genes as markers of malignancy. J Clin Endocrinol Metab 90:1819-1829
45. de Fraipont F, El Atifi M, Gicquel C, Bertagna X, Chambaz EM, Feige JJ 2000 Expression of the angiogenesis markers vascular endothelial growth factor-A, thrombospondin-1, and platelet-derived endothelial cell growth factor in hu- man sporadic adrenocortical tumors: correlation with genotypic alterations. J Clin Endocrinol Metab 85:4734-4741
46. Allolio B, Fassnacht M 2006 Clinical review: adrenocortical carcinoma: clinical update. J Clin Endocrinol Metab 91:2027-2037
JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.