ENDOCRINE SOCIETY
OXFORD
Does Adjuvant Mitotane Impact Cure Rates in Adrenocortical Carcinoma? Insights From the ICARO-GETTHI/SEEN Registry
Alberto Carmona-Bayonas,10 Cristina Alvarez-Escola,20D Inmaculada Ballester Navarro,3 Jorge Hernando Cubero, 4D Miguel Angel Mangas Cruz,5D Rogelio Garcia-Centeno,6DD Clara Iglesias,70 Jesus Garcia-Donas, 80D Maria Jose Picon,90D Miguel Paja, 10
Lorena Gonzalez Batanero, 110D Lourdes Garcia, 12[D Teresa Alonso Gordoa, 130D Carlos Lopez, 14D Felicia Hanzu, 150 Javier Martinez-Trufero, 160D Beatriz Febrero, 17D Patricia Saiz-López, 18[D Concepción Blanco Carrera, 19(D Teresa Ramón y Cajal,20[D Brenda Veiguela,21
Oswaldo Gressani,22(D Nuria Valdes, 23ID and Paula Jimenez-Fonseca7 iD
1Medical Oncology Department, Hospital Universitario Morales Meseguer, Instituto Murciano de Investigación Biosanitaria (IMIB),
Universidad de Murcia (UMU), 30008 Murcia, Spain
2Endocrinology and Nutrition Department, Hospital Universitario La Paz, 28046 Madrid, Spain
3Medical Oncology Department, Hospital Universitario Morales Meseguer, 30008 Murcia, Spain
4Gastrointestinal and Endocrine Tumors Unit, Medical Oncology Department, Hospital Universitario Val de Hebrón, Vall d’Hebron Institute of Oncology (VHIO), 08035 Barcelona, Spain
5Endocrinology and Nutrition Department, Hospital Universitario Virgen del Rocío, 41013 Sevilla, Spain
6Endocrinology and Nutrition Department, Hospital General Universitario Gregorio Marañon, 28007 Madrid, Spain
7Medical Oncology Department, Hospital Universitario Central de Asturias, ISPA, 33011 Oviedo, Spain
8Medical Oncology Department, Centro Integral Oncológico HM Clara Campal, 28050 Madrid, Spain
9Endocrinology and Nutrition Department, Hospital Universitario Virgen de La Victoria, Biomedical Research Institute-IBIMA, Málaga, CIBER Pathophysiology of Obesity and Nutrition-CIBERON, 29010 Madrid, Spain
10Endocrinology Department, Hospital Universitario de Basurto, Bilbao, University of the Basque Country, University of Pais Vasco (UPV/EHU), 48013 Bilbao, Spain
11Medical Oncology Department, Hospital Universitario de Canarias, 38320 Tenerife, Spain
12Endocrinology and Nutrition Department, Hospital Universitario de Jerez, 11407 Jerez, Spain
13Medical Oncology Department, Hospital Universitario Ramón y Cajal, 28034 Madrid, Spain
14Medical Oncology Department, Hospital Universitario Marqués de Valdecilla, Instituto de Investigación Valdecilla (IDIVAL), Universidad de Cantabria (UNICAN), 39008 Santander, Spain
15Endocrinology and Nutrition Department, Hospital Clinic, 08036 Barcelona, Spain
16Medical Oncology Department, Hospital Universitario Miguel Servet, 50009 Zaragoza, Spain
17Endocrine Surgery Unit, General Surgery Department, Hospital Universitario Virgen de la Arrixaca, Instituto Murciano de Investigación Biosanitaria (IMIB) Pascual Parrilla, Universidad de Murcia, 30120 Murcia, Spain
18Medical Oncology Department, Hospital Universitario de Burgos, 09006 Burgos, Spain
19Endocrinology and Nutrition Department, Hospital Universitario Principe de Asturias, Alcalá de Henares, 28805 Madrid, Spain
20Medical Oncology Department, Hospital de la Santa Creu y Sant Pau, 08025 Barcelona, Spain
21Endocrinology and Nutrition Department, Hospital Universitario de Cabueñes, 33394 Gijón, Spain
22Interuniversity Institute for Biostatistics and statistical Bioinformatics (I-BioStat), Data Science Institute, Hasselt University, 3590 Hasselt, Belgium
23Endocrinology and Nutrition Department, Hospital Universitario de Cruces, Biobizkaia, University of Pais Vasco (UPV/EHU), CIBERDEM, CIBERER, Endo-ERN, 48903 Barakaldo, Spain
Correspondence: Paula Jimenez-Fonseca, MD, PHD, Medical Oncology Department, Hospital Universitario Central de Asturias, ISPA, Avenida Roma s/n, 33011 Oviedo, Asturias, Spain. Email: palucaji@hotmail.com.
Abstract
Context: Adrenocortical carcinoma (ACC) is a rare, aggressive malignancy with a high risk of postoperative recurrence. Although adjuvant mitotane is commonly used, its potential for achieving cure rather than simply postponing inevitable recurrence remains uncertain.
Objective: This study investigates whether mitotane impacts ACC recurrence patterns by preventing or delaying recurrence.
Methods: This retrospective analysis used data from the ICARO-GETTHI/SEEN registry, which includes 36 Spanish hospitals. Recurrence in nonmetastatic ACC patients after resection was analyzed using Cox models, flexible longitudinal models, and mixture cure models to evaluate the impact of mitotane.
Results: Among 244 patients, 133 (52%) received adjuvant mitotane, with therapeutic levels monitored in 84%. Findings suggest a possible “cure fraction” with a 32.5% estimated 30-year cure rate (95% CI, 23.4%-45.0%). Cox regression indicated a 39% reduced recurrence risk (HR 0.61; 95% CI, 0.39-0.95) for mitotane-treated patients, with effects diminishing over 24 months. Mixture cure models suggest mitotane primarily delays rather than prevents recurrence. Effect modification analysis showed significant benefit in male patients (HR 0.33; 95% CI, 0.16-0.69), younger patients, tumors with higher Ki-67% (modeled as a continuous variable), and those with venous invasion (HR 0.47; 95% CI, 0.27-0.82), with potential synergy when combined with radiotherapy.
Conclusion: This study underscores the intriguing possibility that adjuvant mitotane delays recurrence, yet questions remain as to its curative capacity. The early benefit suggests a cytostatic effect, but certain subgroups-especially males, younger individuals, and those with high-risk tumors-may experience a more durable outcome. Further research is needed to explore mitotane’s curative potential in ACC management.
Key Words: adrenocortical carcinoma, adjuvant, mixture cure model, mitotane, survival, recurrence
Adrenocortical carcinoma (ACC) is a rare malignancy, with an estimated incidence of 0.7 to 2 cases per million individuals an- nually worldwide (1). Although it has a low incidence, ACC is highly aggressive, even in localized stages after resection. Recurrence rates exceed 75% to 80% in certain high-risk pro- files, such as patients with high-grade tumors, nodal metasta- ses, or incompletely resected disease (R1-R2) (2). Most series have sought to improve prognosis through the use of adjuvant mitotane or radiotherapy (3-5). Mitotane, the only adrenolytic agent approved for advanced ACC treatment, is a compound originally derived from the insecticide dichloro-diphenyl- trichloroethane (DDT). Discovered serendipitously in the 1940s, mitotane was first isolated during research into DDT’s effects (3, 4). The efficacy of mitotane observed in ad- vanced ACC prompted its adoption as an adjuvant treatment in patients with localized, high-risk disease, as endorsed by clinical guidelines (5). However, the evidence supporting its benefit in this setting remains limited due to significant weak- nesses in the available studies, largely stemming from the rarity of the tumor (6, 7). Early investigations, primarily composed of small series, suggested a potential benefit, though they lacked definitive evidence (5, 8). A retrospective Italian-German series became one of the most influential stud- ies, showing longer median recurrence-free survival (RFS) in patients treated with mitotane compared to controls. Italian patients treated with mitotane had an RFS of 42 months, com- pared to 10 months in the Italian control group (hazard ratio [HR] 3.79; 95% CI, 2.27-6.32) and 25 months in the German control group (HR 2.93; 95% CI, 1.74-4.94) (9). Even low doses of mitotane (1-5 g/day) were thought to offer potential therapeutic benefits (10), with treatment strategies at the time focusing on careful monitoring of mitotane plasma levels to maintain a therapeutic range of 14 to 20 mg/L.
Nonetheless, the retrospective, nonrandomized design and small sample size of this and similar studies (11), along with the omission of key prognostic factors-such as tumor burden, Ki-67%, and detailed subgroup data-have consistently raised concerns (12). To address these uncertainties, the ADIUVO-1 study became the first randomized phase 3 trial to evaluate the role of adjuvant mitotane in a controlled setting, specifically targeting patients at low risk of recurrence. The trial included patients with stage I-III ACC who had undergone R0 resection and had a Ki-67 index of 10% or less. A total of 91 patients were randomized, with 45 receiving adjuvant mitotane and 46 assigned to surveillance. The study was terminated early due to slow recruitment, with an interim Bayesian analysis sug- gesting a low probability of reaching the primary endpoint. At 5 years, the RFS was 79% (95% CI, 67-94) in the mitotane
group compared to 75% (95% CI, 63-90) in the surveillance group (HR 0.74; 95% CI, 0.30-1.85) (13).
While the potential benefits of mitotane remain uncertain, its tolerability over extended use has been shown to be moder- ate, often challenged by off-target effects such as neurological symptoms, which frequently necessitate dose adjustments. Despite its wide application, the precise therapeutic effect and determinants of dose modification remain poorly defined (14).
Current expert opinion suggests that low-risk patients, as defined by ADIUVO trial criteria, may be managed with ob- servation, while high-risk patients might benefit from adju- vant mitotane therapy for 2 to 5 years (5). For low-risk patients with specific adverse prognostic features, some ex- perts advocate shorter courses of mitotane (eg, 2 years), with a low threshold for discontinuation. However, a key fac- tor in the decision to discontinue treatment is determining whether patients are truly cured or still at risk of relapse des- pite the therapy. Mitotane’s dual cytotoxic and cytostatic properties are central to this discussion, as cytostatic therapies often require prolonged use to suppress micrometastatic growth, while cytotoxic agents are typically administered in short, intensive courses aimed at eradicating residual disease (8, 15). For instance, if mitotane’s effects in typical clinical set- tings are more compatible with a cytostatic model-delaying relapse rather than preventing it-this would substantially al- ter the perception of this therapy, suggesting that extended treatment might be warranted depending on the patient’s re- currence risk.
Given these uncertainties, we aimed to evaluate the efficacy of adjuvant mitotane using data from the multicenter ICARO-GETTHI/SEEN registry, which includes contribu- tions from 36 Spanish hospitals. Additionally, we sought to determine whether the benefits of mitotane diminish over time, assessing if it provides a lasting reduction in recurrence risk or merely delays disease recurrence.
Methods
Study Design and Patients
The ICARO-GETTHI/SEEN registry is a multicenter, obser- vational study of patients with ACC. Promoted by the Spanish Group of Transversal Oncology and Rare and Orphan Tumors (GETTHI) and the Spanish Society of Endocrinology and Nutrition (SEEN), this registry collects data from 36 hospitals across Spain to ensure the representa- tivity of this rare disease. To minimize selection bias, patient recruitment is consecutive.
Eligibility criteria for the ICARO-GETTHI/SEEN registry are as follows: adult patients aged 18 years or older with his- tologically confirmed ACC are included, while patients with a follow-up of less than 3 months, except in cases of early death, are excluded. Participating investigators and centers commit to reporting all ACC cases treated at their institutions, regard- less of diagnosis year, tumor stage, care setting, or type of management received.
This study adheres to the ethical principles outlined in the Declaration of Helsinki and follows Good Clinical Practice guidelines and their updates. The ICARO-GETTHI/SEEN registry protocol was approved by the Medical Research Ethics Committee (MREC) of the Hospital Universitario Central de Asturias, acting as the reference MREC, and by the MRECs of all participating hospitals. Informed consent was obtained from all living patients prior to their inclusion in the registry.
Variables
The online database (http://www.icaro.com) collects a wide range of demographic, clinical, biochemical, imaging, histo- logical, and treatment-related information. The web applica- tion is designed to prioritize data reliability by minimizing free-text fields and complex data entries. Instead, it uses drop- down menus, checkboxes, and predefined ranges. Integrated filters, validations, and help prompts further enhance data quality throughout the platform.
To assess the effect of adjuvant therapies in localized tu- mors, the following confounding factors were considered: year of treatment, age, sex, tumor size, Ki-67 index, age, T stage, lymph node involvement, venous invasion, and surgical margin (R0-1-2). These factors were selected a priori based on literature review and expert consultation. Disease was staged according to the European Network for the Study of Adrenal Tumors (ENSAT) classification. High-risk disease was defined as the presence of positive lymph nodes, positive resection margins, or Ki-67 ≥ 10%.
Perioperative chemotherapy was defined as systemic chemotherapy initiated prior to surgery (neoadjuvant) and continued after surgery (adjuvant), typically in patients with aggressive, high-grade tumors.
Treatment efficacy, particularly of mitotane, was assessed using event-free survival (EFS) and overall survival (OS). EFS was defined as the time from surgery to the first tumor re- currence or death from any cause, whichever occurred first. EFS was chosen as the primary endpoint to account for com- peting risks, such as death without recurrence, ensuring that all clinically relevant events were included in the analysis. Additionally, as part of a sensitivity analysis, RFS was also evaluated, focusing solely on tumor recurrence as the event of interest. OS was defined as the time from surgery to death from any cause. Patients without events at the end of follow- up were considered censored. Toxicity was classified accord- ing to the CTCAE v4.0 criteria in effect at the time of the treat- ment administration.
Statistical Analysis
Descriptive statistics summarized the study population: cat- egorical variables were reported as frequencies and percen- tages, while continuous variables were presented as means, medians, and ranges. EFS was analyzed using Cox proportion- al hazards regression, adjusting for tumor size, Ki-67 index,
age, sex, T stage, lymph node involvement, venous invasion, year, and radiotherapy to isolate the adjuvant effect of mito- tane. Multiple imputation using predictive mean matching was applied (20 imputed datasets). To avoid immortal time bias, the temporal structure of exposures, including mitotane treatment duration and longitudinal mitotane levels, was pre- served by analyzing them as time-dependent covariates (16). The proportional hazards assumption was assessed using Schoenfeld residuals to explore a potential time-varying effect of mitotane. Subsequently, a multivariable Royston-Parmar spline model was applied to characterize the evolving hazard ratios, capturing the diminishing effect of mitotane over time.
Changes in hazard associated with mitotane discontinu- ation were also examined. To further investigate, a mixture cure model with Laplacian-P-splines was used to determine whether the diminishing effect influenced cure rates, distin- guishing cured from noncured patients and adjusting for rele- vant covariates. Longitudinal mitotane levels (mitotanemia) were evaluated using mixed-effects models with splines for age, sex, treatment year, and time. A joint model integrated these longitudinal mitotanemia levels with recurrence risk through an association parameter (a), where a negative a indi- cated that higher mitotanemia levels observed at any point lowered recurrence risk. We set a 5% significance level for main effects and used a less conservative 10% threshold for in- teractions in subgroup analyses, facilitating the detection of signals warranting further investigation in this orphan disease context (17). A nontechnical overview of the models, designed for clinicians, along with a comparison of goodness-of-fit in dynamic scenarios between the Cox model and the Royston-Parmar model, is presented in Supplementary Appendix 1 (18). All analyses were performed in R version 4.1.0 (19), utilizing the survival, rms, mixcurelps, JM, and rstpm2 packages.
The original dataset and the R code used to reproduce the analyses are available in the supplementary materials and have been deposited in an online repository (20, 21).
Results
Patients and Surgery
The registry comprises data from 244 patients with nonmeta- static, resectable ACC (TNM stages I-III) who underwent pri- mary tumor resection. The cases recruited per year and center are detailed in Supplementary Appendix Figures 1 and 2 (18). The cohort had a mean age of approximately 50 years with a slight female predominance (42.5% vs 57.5%). Baseline char- acteristics and summary of basic therapies, stratified by mito- tane use, are presented in Table 1. The median interval from diagnosis to surgery was 40 days (interquartile range, 78 days). Surgical approaches included laparoscopic procedures in 77 patients (31.6%), open surgery in 157 patients (64.3%), and combined techniques in 10 patients (4.1%). Resection margins were classified as R0 in 175 patients (71.7%), R1 in 36 patients (14.8%), R2 in 4 patients (1.6%), and were unknown in 29 patients (11.9%). The types of surger- ies performed were adrenalectomy in 165 patients (68%), adrenalectomy with nephrectomy in 34 patients (14%), adre- nalectomy with lymphadenectomy in 24 patients (10%), adre- nalectomy with resection of other structures in 10 patients (4%), adrenalectomy with cytoreduction of other disease in 4 patients (2%), and other unspecified surgeries in 7 patients (3%).
| Variables | Overall 244 (100%) | No adjuvant mitotane 114 (100%) | Adjuvant mitotane 130 (100%) |
|---|---|---|---|
| Baseline characteristics | |||
| Age, years, mean (SD) | 50.3 (16.1) | 49.9 (17.2) | 51.4 (15.3) |
| Sex, female | 152 (62.3) | 63 (55.3) | 89 (68.5) |
| ECOG PS | |||
| 0 | 112 (45.9) | 47 (41.2) | 65 (50.0) |
| 1 | 93 (38.1) | 43 (37.7) | 50 (38.5) |
| 2 | 12 (4.9) | 6 (5.3) | 6 (4.6) |
| 3 | 21 (8.6) | 15 (13.2) | 6 (4.6) |
| NA | 6 (2.5) | 3 (2.6) | 3 (2.3) |
| Multidisciplinary committee | 138 (56.6) | 46 (40.4) | 92 (70.8) |
| Diagnosis year | |||
| 1983-1991 | 6 (2.5) | 5 (4.4) | 1 (0.8) |
| 1992-1999 | 16 (6.6) | 14 (12.3) | 2 (1.5) |
| 2000-2008 | 28 (11.5) | 24 (21.1) | 4 (3.1) |
| 2009-2016 | 81 (33.2) | 43 (37.7) | 38 (29.2) |
| 2017-2024 | 113 (46.3) | 28 (24.6) | 85 (65.4) |
| Incidental diagnosis | 83 (34.0) | 36 (31.6) | 47 (36.2) |
| Clinical presentation by tumor mass | 103 (42.2) | 49 (43.0) | 54 (51.5) |
| Clinical presentation by | 94 (38.5) | 37 (32.5) | 57 (43.8) |
| endocrine symptoms | |||
| Cushing syndrome | 56 (23.0) | 21 (18.4) | 35 (26.9) |
| Mineralocorticoid excess | 19 (7.8) | 7 (6.1) | 12 (9.2) |
| syndrome | |||
| Hirsutism, virilization, | 62 (25.4) | 24 (21.1) | 38 (29.2) |
| and/or amenorrhea | |||
| Familial hereditary syndrome | 7 (2.9) | 5 (4.4) | 2 (1.5) |
| Ki-67%, mean (SD) | 24.1 (20) | 26.4 (22.8) | 23.1 (18.5) |
| Tumor size (cm), mean (SD) | 11.3 (8.7) | 10.7 (6.5) | 11.8 (8.0) |
| Weiss index, median (IQR) | 6 (4) | 4.4 (2) | 6.6 (2) |
| ENSAT stage | |||
| I | 22 (9.0) | 15 (13.2) | 7 (5.4) |
| II | 124 (50.8) | 64 (61.1) | 60 (46.2) |
| III | 98 (40.2) | 35 (30.7) | 63 (48.5) |
| cT | |||
| cT1 | 23 (9.4) | 15 (13.2) | 8 (6.2) |
| cT2 | 136 (55.7) | 69 (60.5) | 67 (51.5) |
| cT3-4 | 85 (34.8) | 30 (26.3) | 55 (42.3) |
| Positive cN | 31 (12.7) | 12 (10.5) | 19 (14.6) |
| Treatments other than mitotane | |||
| Surgery type | |||
| Laparoscopic | 77 (31.6) | 35 (30.7) | 42 (32.3) |
| Open | 157 (64.3) | 76 (66.7) | 81 (62.3) |
| Both | 10 (4.1) | 3 (2.6) | 7 (5.4) |
| Resection margin | |||
| R0 | 175 (71.7) | 85 (74.6) | 90 (69.2) |
| R1 | 36 (14.8) | 11 (9.6) | 25 (19.2) |
| R2 | 4 (1.6) | 2 (1.8) | 2 (1.5) |
| Unknown | 29 (11.9) | 16 (14.0) | 13 (10.0) |
| Adjuvant radiotherapy | 36 (14.7) | 11 (9.6) | 25 (19.2) |
(continued)
| Variables | Overall 244 (100%) | No adjuvant mitotane 114 (100%) | Adjuvant mitotane 130 (100%) |
|---|---|---|---|
| Perioperative chemotherapy | 25 (10.2) | 8 (7) | 17 (13.1) |
Values are presented as n (%) unless otherwise specified. Percentages are calculated by columns.
Abbreviations: cN, clinical Node status; cT, clinical Tumor stage; cTNM, clinical Tumor, Node, Metastasis staging system; ECOG PS, Eastern Cooperative Oncology Group Performance Status; IQR, interquartile range; NA, not available; RO, No residual tumor (complete resection with negative margins); R1, Microscopic residual tumor; R2, Macroscopic residual tumor.
Severe postoperative complications were reported in 21 pa- tients (9%).
Adjuvant Treatment
Regarding adjuvant therapy, 36 patients (14.7%) received postoperative radiotherapy, 133 patients (52%) were treated with adjuvant mitotane, and 25 patients (10.2%) received perioperative chemotherapy. Of those treated with mitotane, 43 patients (32%) remained on treatment at the time of ana- lysis. The median initial mitotane dose was 2500 mg/day. Primary reasons for not prescribing mitotane included insuffi- cient clinician experience (37 patients, 30.6%), lack of sup- porting evidence (30 patients, 24.8%), unavailability of the drug at the center (29 patients, 24%), unfavorable risk-benefit assessment (22 patients, 18.2%), and inability to monitor serum drug levels (3 patients, 2.5%). The median duration of mitotane therapy was 23 months (95% CI, 18.8-24.3 months).
Therapeutic drug monitoring was conducted in 111 of the 133 patients (84%). Trends in mitotane serum levels over time are illustrated in Supplementary Appendix Figures 3 and 4 (18), with no evidence of an impact of sex or age on mi- totane levels. However, in a longitudinal model, the year of therapy was significantly associated with mitotane levels (esti- mate: 0.031; SE: 0.016; t = 2.01; P = . 047), reflecting changes in clinical practice over time (see Supplementary Appendix Table 1) (18).
Analysis of maximum mitotane concentrations revealed that 37% of patients had peak levels ≤14 µg/mL, indicating that a third did not reach the desired therapeutic range. Conversely, 21% achieved maximum levels within the thera- peutic range (14-20 µg/mL), while 42% exceeded this range at some point, reaching levels above 20 µg/mL (Supplementary Appendix Figure 5) (18). On average, patients spent 72.5% of the time below the therapeutic range (<14 µg/mL), 8.8% within the optimal range (14-20 µg/mL), and 18.7% above it (>20 µg/mL) (Supplementary Appendix Figure 6) (18). Adherence rates-defined as the percentage of the planned dose taken-were as follows: adherence of over 90% was ob- served in 91 patients (68.4%), 75% to 90% in 13 patients (9.8%), 50% to 74% in 4 patients (3.0%), 25% to 50% in 2 patients (1.5%), less than 25% in 2 patients (1.5%), with adherence unknown in 18 patients (15.8%). Reasons for dis- continuation of mitotane therapy included toxicity (29 pa- tients, 32.2%), completion of the planned course (24 patients, 26.7%), disease recurrence (18 patients, 20%), pa- tient preference (2 patients, 2.2%), and other reasons (17 pa- tients, 19.8%).
The prescription of adjuvant mitotane has been increasingly influenced by clinical, demographic, and tumor-specific fac- tors over time. Over time, mitotane prescriptions rose dramat- ically from 6.7% in 1983-1991 to 72.5% in 2004-2016 (P <. 001). There is also a notable rise in both multidisciplin- ary committee involvement and mitotane use in recent years (Supplementary Appendix Figures 7 & 8) (18). Patients re- viewed by multidisciplinary teams were more likely to receive mitotane (65.1% vs 35.2%, P <. 001). Patients prescribed mitotane had significantly higher Ki-67% levels (mean 27.3% vs 17.4%, P =. 0011) and Weiss scores (median 5 vs 4, P =. 0001), reflecting more aggressive tumor biology. Female patients were more frequently prescribed mitotane than male patients (57.5% vs 43.6%, P =. 0446). Mitotane utilization increased with more advanced TNM stages- 30.4% in stage I, 48.1% in stage II, and 62.7% in stage III (P =. 0075)-mainly attributable to higher T stages, while no- dal involvement showed no significant association. The clinic- al presentation before surgery did not significantly affect mitotane prescribing patterns. Notably, 69% of the 37 pa- tients who underwent adjuvant radiotherapy also received mi- totane. Other baseline characteristics of patients treated with adjuvant radiotherapy or perioperative chemotherapy are shown in Supplementary Appendix Table 2 (18).
Effect of Mitotane and Radiotherapy on EFS and OS
A total of 133 events were observed, including 121 disease re- currences and 12 deaths without prior recurrence, with a me- dian EFS of 30.2 months (95% CI, 23.7-68.6 months). Among patients who experienced recurrence, the median time to relapse was 36 months (95% CI, 26-106 months). For patients without events, the mean follow-up was 9.4 years, based on restricted mean survival up to the maximum of 35 years. At this time point, the proportion of patients achiev- ing cure is estimated at 32.5% (95% CI, 23.4%-45%). Overall, 111 deaths were recorded, with a median OS of 114 months (95% CI, 61.3-150 months). Among surviving patients at the time of analysis, the median follow-up was 83.4 months (95% CI, 67.8-96.2 months) (Supplementary Appendix Figure 9) (18).
The site of recurrence was identified in 120 of the 121 cases. For patients treated with mitotane compared to those with- out, the distant recurrence was 57.1% vs 52.6%, and the local recurrence rate was 42.9% vs 47.4%, respectively (P =. 754) (Supplementary Appendix Table 3A) (18). Among patients re- ceiving radiotherapy, the distant recurrence rates were 38.9% and 25% in those without radiotherapy, while the local recur- rence rate was 13.9% compared to 23.6%, respectively (P =. 125). Patterns of distant and local recurrence, consider- ing mitotane or radiotherapy alone or in combination, are in- cluded in Supplementary Appendix Table 3B (18).
Kaplan-Meier plots for EFS, stratified by venous invasion, Ki-67%, resection margin, tumor size, and surgical approach, are presented in Supplementary Appendix Figure 10 (18). The curves stratified by mitotane use initially diverged but con- verged after 24 months, with no significant differences observed in OS (log-rank test, P = . 91) or EFS (log-rank test, P =. 81) (Fig. 1).
A multivariable Cox proportional hazards model, adjusted for relevant covariates (Fig. 2A), demonstrated that mitotane use was associated with a 39% reduction in events (HR for EFS 0.61; 95% CI, 0.39-0.95). In a sensitivity analysis
excluding patients who had received perioperative chemother- apy, mitotane’s benefit was comparable (HR 0.60; 95% CI, 0.37-0.96 for EFS). In a second sensitivity analysis considering RFS as the endpoint (focusing exclusively on recurrence events), the HR was 0.59 (95% CI, 0.36-0.97).
Notably, chemotherapy itself showed no clear effect (Supplementary Appendix Figure 11) (18). In an additional sensitivity analysis, the reason for omitting mitotane was also considered. The benefit was consistent when the reason for omission was clinician perception of an unfavorable risk- benefit balance (HR 0.32; 95% CI, 0.17-0.61), difficulties in achieving therapeutic levels (HR 0.64; 95% CI, 0.14-2.78), lack of physician familiarity with the drug (HR 0.55; 95% CI, 0.30-1.01), and unavailability of the drug at the treatment center (HR 0.47; 95% CI, 0.26-0.85). Supplementary Appendix Figure 12 (18) presents a Cox model for the OS end- point, showing consistent results. Specifically, the HR for OS associated with adjuvant mitotane was 0.57 (95% CI, 0.35-0.93). To rule out survivor bias, data were analyzed by treatment year quartiles, and no evidence of heterogeneity in HRs was found (Supplementary Appendix Figure 13) (18).
The longitudinal joint model showed a non-significant asso- ciation between mitotane levels and reduced recurrence risk (HR 0.88; 95% CI, 0.35-1.35). The joint model is reported in Supplementary Appendix Table 1 (18). Kaplan-Meier curves stratified by mitotane peak levels, along with a modeled patient example, are presented in Supplementary Appendix Figure 14 (18).
Predictive Factors and Subgroup Analysis
Subgroup analyses within the Cox model revealed differential effects based on Ki-67 levels, with tumors exhibiting a lower Ki-67 index (Ki-67 <10%) showing minimal to no benefit from mitotane (interaction P =. 05) (Fig. 3). For example, the HR for patients with a Ki-67 of 8% was 0.91 (95% CI, 0.53-1.57) compared with 0.43 (95% CI, 0.23-0.79) for those with a Ki-67 of 40%. The model also identified a sex-based subgroup effect, indicating a more pronounced benefit in men (HR 0.33; 95% CI, 0.16-0.69) compared with women (HR 0.89; 95% CI, 0.53-1.49) (interaction P =. 012). Additionally, older patients appeared to derive less benefit from mitotane, with an HR of 0.45 (95% CI, 0.24-0.83) for 40-year-olds vs 0.90 (95% CI, 0.50-1.60) for 70-year-olds (interaction P =. 088). Additional subgroup effects of mito- tane efficacy are presented in Fig. 3. Patients with venous inva- sion demonstrated a greater benefit, with an HR of 0.47 (95% CI, 0.27-0.82), compared with an HR of 0.92 (95% CI, 0.49-1.73) in those without venous invasion (interaction P = . 090). No significant interactions were observed for lymph node status (P = . 53) or tumor size (P = . 63). For patients who received adjuvant radiotherapy, the interaction term ap- proached significance (P =. 10). Those receiving radiotherapy showed a benefit from mitotane use, with an HR of 0.32 (95% CI, 0.12-0.87), compared to an HR of 0.77 (95% CI, 0.49-1.22) for mitotane in patients who did not receive radio- therapy. Figure 4 presents Kaplan-Meier plots stratified by ad- juvant mitotane use and by ENSAT risk status or Ki-67 levels alone. We did not find evidence of a subgroup effect based on the presence of an adrenal hormone hypersecretion syndrome or hypercortisolism on the efficacy of mitotane (see Supplementary Appendix Figure 15) (18).
1.00
+ No mitotane
Event-free survival, %
0.75
Adjuvant mitotane administration
0.50
Log-rank test
0.25
p = 0.81
0.00
0
6
12
18
24
30
36
42
48
54
60
Time (months)
Number at risk
114
85
71
66
56
51
49
45
44
41
38
130
103
81
71
54
46
38
35
33
29
27
1.00
No mitotane
Adjuvant mitotane administration
0.75
Overall survival, %
0.50
Log-rank test p = 0.91
0.25
0.00
0
12
24
36
48
60
72
84
96
108
120
Time (months)
Number at risk
114
93
81
68
58
51
44
40
35
32
30
130
107
88
69
57
41
29
23
17
17
12
A
Hazard Ratio for EFS (66,95, 99% CI)
0.40
0.75
1.00
1.40
1.80
2.20
2.60
HR (95% CI)
Year - 2018 vs 2009
1.10 (0.84-1.43)
Age - 61 vs 40
1.64 (1.26-2.12)
Ki67% - 35 vs 10
1.67 (1.26-2.21)
Tumor size (cm) - 14 : 7
1.14 (0.97-1.33)
Stage - I vs II
0.64 (0.28-1.45)
Stage - III vs II
1.52 (1.01-2.28)
Adjuvant radiotherapy
0.79 (0.47-1.33)
Margin - R1/2 vs RO
1.09 (0.63-1.90)
Adjuvant mitotane
0.60 (0.39-0.94)
Venous invasion
1.50 (1.00-2.26)
B Time-dependent hazard ratio (EFS) - Adjuvant mitotane
2.0
1.5
Hazard ratio (EFS)
1.0
0.5
0.0
0
10
20
30
40
50
60
Time (months)
Dynamic Effect and Cure Modeling With Mitotane
The Schoenfeld residuals test indicated a violation of the pro- portional hazards assumption, suggesting a time-varying ef- fect for mitotane (x2=7.22, df=1, P =. 0072). After excluding patients who stopped treatment due to recurrence, mitotane discontinuation was linked to a higher risk (HR for EFS 2.37; 95% CI, 1.24-4.56), independent of the reason for stopping: discontinuation due to toxicity had an HR of 1.35 (95% CI, 0.46-3.97), while discontinuation based on physician or patient choice was associated with an HR of 2.53 (95% CI, 0.91-7.06) compared to continued therapy.
Time-varying HR analysis (Fig. 2B) showed an initial pro- tective effect of mitotane at 6 months (HR, 0.67; 95% CI, 0.42-1.05), which progressively diminished, reaching an HR of 1.12 (95% CI, 0.77-1.62) by 2 years.
To assess the impact of time-dependent effects on cure probability, a mixture cure model was applied (Table 2). This model found that age was the only variable associated with the probability of noncure, with an odds ratio (OR) of 1.08 (95% CI, 1.01-1.16). There was no evidence that adju- vant radiotherapy impacted cure probability (OR of 6.54, 95% CI, 0.53-83.33). While mitotane delayed recurrence among noncured patients (HR, 0.54; 95% CI, 0.30-0.95), representing a 1.85-fold decrease in risk, it did not significant- ly influence the probability of achieving definitive cure. For the cure endpoint, interactions were explored between mitotane and factors such as sex, ENSAT risk, and age. None were sig- nificant; for instance, in high ENSAT risk patients, the odds of cure with mitotane were OR 1.22 (95% CI, 0.46-3.25), and in male patients, OR 1.14 (95% CI, 0.21-6.25).
Ki-67: 5%
1
1.01 (0.54-1.89)
Ki-67: 15%
.
0.81 (0.51-1.28)
.
Ki-67: 25%
0.65 (0.42-1.01)
Ki-67: 35%
.
0.52 (0.29-0.94)
Ki-67: 45%
.
0.42 (0.19-0.94)
.
Ki-67: 55%
0.33 (0.12-0.97)
Ki-67: 65%
-
0.27 (0.07-1.02)
Ki-67: 75%
.
0.22 (0.04-1.08)
.
Lymph nodes (absent, pNO)
0.71 (0.45-1.14)
Subgroup or point estimate
Lymph nodes (present, pN+)
0.56 (0.23-1.40)
No venous invasion
.
0.94 (0.52-1.69)
.
Venous invasion present
0.52 (0.30-0.92)
Tumor size (3 cm)
0.60 (0.34-1.07)
Tumor size (13 cm)
.
0.69 (0.45-1.08)
.
Tumor size (23 cm)
0.80 (0.45-1.43)
Tumor size (33 cm)
L
0.92 (0.38-2.23)
Tumor size (43 cm)
.
1.06 (0.31-3.62)
.
Age (18 years)
”
0.34 (0.12-0.94)
Age (30 years)
.
0.43 (0.20-0.91)
Age (42 years)
.
0.54 (0.31-0.93)
.
Age (54 years)
0.68 (0.44-1.05)
Age (66 years)
.
0.86 (0.51-1.45)
Age (78 years)
-
1.08 (0.52-2.25)
.
Margin RO
0.85 (0.52-1.38)
Margin R1/2
L
0.30 (0.13-0.72)
No radiotherapy
.
0.77 (0.49-1.22)
.
Adjuvant radiotherapy
0.32 (0.12-0.87)
Male
-
0.38 (0.19-0.74)
Female
.
0.97 (0.57-1.65)
1
ENSAT low-risk
1.14 (0.30-4.30)
ENSAT high risk
+
0.62 (0.37-1.05)
.
0.0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
2.4
2.7
3.0
3.3
3.6
3.9
4.2
4.5
4.8
Hazard Ratio (EFS)
Discussion
In this analysis using data from the ICARO-GETTHI/SEEN registry, we evaluated the efficacy of adjuvant mitotane in ACC, with a focus on its time-varying effects and potential im- pact on cure rates. Extended follow-up of these patients, in- cluding some observed for over 35 years, indicated that approximately one-third were cured of their disease. This raised an intriguing question: did adjuvant mitotane affect the cure fraction?
Overall, the findings suggest that mitotane offers a tempor- ary benefit by delaying recurrence rather than promoting cure, aligning more closely with the effects of a cytostatic agent. Evidence supporting this included early separation and later convergence of Kaplan-Meier curves, elevated recurrence risk after treatment, a significant dynamic effect test showing a diminishing HR over time, and results from cure models.
Regarding comparison with previous literature, the effect of mitotane on recurrence in our cohort (HR, 0.61; 95% CI, 0.39-0.95) was consistent with a meta-analysis of 12 studies (4606 patients), showing a similar pooled RFS benefit (HR 0.63; 95% CI, 0.44-0.92) (22). Our findings also align well with the ADIUVO-1 trial, which showed no benefit from mi- totane in low-risk ACC (HR for RFS, 0.74; 95% CI, 0.30-1.85) (13). Indeed, in our cohort, the benefit was primar- ily observed in high-risk patients, as defined by ENSAT, who showed a significant benefit. These results, therefore, support current guidelines recommending adjuvant mitotane for
patients with Ki-67> 10%, positive margins, or stage III ACC (5).
Interaction analysis showed significant subgroup effects, with a marked benefit in male patients, whereas the benefit in females appeared limited. Allegra et al provide a possible explanation, suggesting that sex may influence plasma mitotane and o, p’-DDE levels, affecting the likelihood of achieving therapeutic mitotane concentrations (23). However, our analysis found no evidence of an impact of sex on mitotane levels over time, as shown in the longitudinal model and illustrated in Supplementary Appendix Figures 4 (18). This suggests that oth- er factors, potentially hormonal, genetic, or multifactorial, may be more relevant contributors in this context. Furthermore, age appears to play a significant role in treatment outcomes. In our series, older patients, particularly those aged 65 and above, dem- onstrated reduced benefit from mitotane. Although data on this are limited, a multicenter registry of Asian patients reported that individuals aged ≥60 had worse OS (HR, 1.47; 95% CI, 1.24-1.75) (24). Potential explanations for the reduced efficacy of mitotane in older patients include age-related physiological changes in the adrenal glands, increased comorbidities, distinct molecular tumor profiles, or clinical challenges such as the in- creased difficulty of adrenal surgery in elderly patients (25, 26).
Additional exploratory interactions, noted at a 10% signifi- cance level, warrant further investigation in upcoming trials, particularly in relation to radiotherapy, where some hetero- geneity in effect was observed. Mitotane was associated
A
+ No mitotane - ENSAT high risk
1.00
+ No mitotane - ENSAT low-risk
+ Mitotane - ENSAT high risk
+ Mitotane - ENSAT low risk
0.75
Event-free survival, %
0.50
Log-rank
0.25
p = 0.0084
0.00
0
6
12
18
24
30
36
42
48
54
60
Time (months)
Number at risk
58
39
31
26
21
18
17
16
16
15
15
14
14
13
13
11
9
9
7
7
6
5
106
85
64
55
41
34
29
27
26
22
20
9
8
7
6
5
5
5
4
3
3
3
B
No mitotane - Ki67% <10
1.00
+ No mitotane - Ki67% ≥10
+ Mitotane - Ki67% <10
Mitotane - Ki67% ≥10
0.75
Event-free survival, %
0.50
Log-rank
0.25
p = 0.0017
0.00
0
6
12
18
24
30
36
42
48
54
60
Time (months)
Number at risk
25
23
22
21
19
16
16
13
13
11
10
19
13
10
9
7
5
4
4
4
4
4
24
22
17
16
13
12
12
10
8
8
8
77
62
46
38
27
21
17
16
16
13
11
| Variable | Estimate | Standard error | Z | P value | CI 95% lower | CI 95% upper |
|---|---|---|---|---|---|---|
| Incidence model | ||||||
| Intercept | -69.01288 | 98.631 | -0.7 | .484 | -262.329 | 124.303 |
| Mitotane | 0.66748 | 0.872 | 0.766 | .444 | -1.041 | 2.376 |
| Ki-67% | 0.0835 | 0.054 | 1.533 | .125 | -0.023 | 0.19 |
| Age | 0.08375 | 0.036 | 2.33 | .02 | 0.013 | 0.154 |
| Lymph nodes involvement | 0.50449 | 1.069 | 0.472 | .637 | -1.591 | 2.6 |
| No surgical margin involvement | -1.2531 | 1.286 | -0.974 | .33 | -3.774 | 1.268 |
| Hormonal hypersecretion | -0.13425 | 0.867 | -0.155 | .877 | -1.833 | 1.564 |
| Sex (female) | 1.13749 | 0.958 | 1.187 | .235 | -0.74 | 3.015 |
| Radiotherapy | -1.87824 | 1.282 | -1.465 | .143 | -4.391 | 0.635 |
| Tumor size (cm) | 0.19228 | 0.108 | 1.786 | .074 | -0.019 | 0.403 |
| Year of diagnosis | 0.03149 | 0.049 | 0.645 | .519 | -0.064 | 0.127 |
| Latency process | ||||||
| Mitotane | -0.61146 | 0.29 | -2.105 | .035 | -1.181 | -0.042 |
| Ki-67% | -0.0002 | 0.015 | -0.014 | .989 | -0.029 | 0.028 |
| Age | -0.03079 | 0.009 | -3.49 | -0.048 | -0.013 | |
| Lymph nodes involvement | 0.51764 | 0.321 | 1.611 | .107 | -0.112 | 1.147 |
| No surgical margin involvement | -0.80937 | 0.264 | -3.071 | .002 | -1.326 | -0.293 |
| Tumor size (cm) | -0.0489 | 0.026 | -1.851 | .064 | -0.101 | 0.003 |
| Hormonal hypersecretion | 0.95493 | 0.372 | 2.569 | .01 | 0.226 | 1.684 |
| Sex (female) | 0.05475 | 0.261 | 0.21 | .834 | -0.456 | 0.565 |
The table summarizes the estimates from a mixture cure model using Laplacian-P-splines. The model includes an incidence part and a latency part. The incidence component assesses the probability of being in the noncured group, while the latency component evaluates the time until recurrence or event for noncured individuals. Covariates such as mitotane use, Ki-67, age, lymph node involvement (N), surgical margin status, hormonal hypersecretion, sex, radiotherapy, tumor size, and year of diagnosis are included in the model. The estimates, standard errors, and 95% confidence levels are reported. A negative estimate in the latency component for mitotane suggests a protective effect against recurrence among noncured patients, with the exponentiated value interpreted as a hazard ratio (a hazard ratio <1 indicating a protective effect on recurrence in noncured patients). In the incident process, a positive estimate increases the likelihood of noncure; the exponentiated estimates in this part are interpreted as odds of noncure, where an odds ratio <1 implies a benefit in terms of definitive cure. Abbreviation: CI, credible interval.
with improved outcomes in patients also receiving radiother- apy, although we could not gather sufficient evidence to sug- gest that radiotherapy alone contributes to cure. Beyond the limited number of patients treated with radiotherapy in this registry, a potential synergy with mitotane could explain the observed benefit. Although a synergistic effect between mito- tane and radiotherapy has been hypothesized in advanced dis- ease, definitive evidence remains limited (22, 27). Supporting this potential advantage, a meta-analysis consistent with our data showed that adjuvant radiotherapy was associated with improved OS (HR, 0.69; 95% CI, 0.58-0.83; P < 0.001) (22).
Regarding Ki-67 and venous invasion, the 2022 WHO clas- sification emphasizes venous invasion as a key component in multiparametric prognostic systems (28). In our series, higher Ki-67 and presence of venous invasion were both associated with greater benefit from mitotane, underscoring their poten- tial role in treatment selection and risk stratification.
A key finding from the ICARO-GETTHI/SEEN registry is the potential time-varying effect of mitotane, initially sug- gested by the distinctive “banana-shaped” morphology in Kaplan-Meier curves, which showed temporary separation followed by convergence around 24 months. This pattern cor- responds with the median discontinuation time of mitotane at 23 months (95% CI, 18.8-24.3 months) and the subsequent decline in mitotane levels. Detailed statistical analyses using flexible models confirmed a time-dependent effect on recur- rence hazards. Similar effects for adjuvant therapies have
been observed in breast cancer, highlighting the presence of noncured patients (29). This pattern had not been previously described in ACC literature, raising the question of whether long-term cure is achievable. Using a mixture cure model, we differentiated between potentially cured postsurgery pa- tients and those at risk of recurrence, represented by a latent process. Cure models are well-established in statistical litera- ture but are less widely recognized clinically. They are particu- larly valuable in adjuvant therapy studies where the assumption that all subjects will eventually experience the event does not hold, as violating this assumption can lead to biased estimates and interpretative challenges (30). This ana- lysis indicated that mitotane delayed recurrence (HR, 0.46; 95% CI, 0.31-0.70); however, there was no evidence of in- creased cure rates. Based on this result, our hypothesis is that, although mitotane is theoretically expected to have a cytotoxic effect that eradicates residual disease and increases cure rates-a common assumption in the literature-it may not be its primary mechanism. Supporting this hypothesis, in a Cox model with mitotane discontinuation as a time- dependent variable, we observed that the hazard of recurrence doubled in patients who were no longer exposed. This finding could, of course, be influenced by usage patterns such as tim- ing and prolonged subtherapeutic dosing in this registry. Additional data on this topic are scarce; however, insights from other tumors with presumably similar cellular dynamics may shed some light. For instance, in adjuvant treatment for
GIST, recurrence rates typically increase within 6 to 12 months after discontinuing imatinib, which suggests a dormancy-inducing rather than a curative effect on micrometa- stases (31). Preclinical data support this, indicating imatinib primarily induces quiescence without promoting cell death, thus delaying rather than preventing recurrence (32). This has clinical implications, as studies have shown similar recurrence hazards after discontinuation, regardless of treatment duration (33, 34), supporting the need for long-term therapy (35). A similar pattern is observed with endocrine therapy for breast cancer, where long-term data reveal a persistent risk of relapse that supports extended therapy for certain patients (36, 37). Multiple studies have shown that the hazard rate for recurrence remains consistently above zero, indicating ongoing risk (29). Even in randomized trials where extending therapy beyond 7 to 8 years shows no added advantage, the absence of a plateau suggests that tumors are not fully eradicated (38, 39).
In vitro evidence is compatible with our cure model. Preclinical data suggest that mitotane, the primary treatment for advanced ACC, operates via dual mechanisms. It induces G2-phase arrest through disruption of the G2 checkpoint (Cdk1-cyclin B complex) and may also exert cytolytic effects, eradicating cancer cells through mitochondria-mediated apoptosis and leading to cell death via aponecrosis. The inter- play of mechanisms seems concentration-dependent, consist- ent with the low doses applied in this cohort (27, 40, 41). In our registry, findings may have been influenced by inadequate mitotane exposure-with 72% of patients spending most of their treatment time outside the therapeutic range-and by treatment duration. However, given mitotane’s toxicity, opti- mizing these parameters may prove challenging. An alterna- tive strategy could involve adding cytotoxic chemotherapy in high-risk patients, a rationale reflected in the ADIUVO-2 trial, which compares mitotane with and without cytotoxic chemotherapy in this population.
Our study has several limitations. First, as a retrospective analysis, it shares the inherent limitations of this design; how- ever, we included additional clinical and demographic factors to enhance robustness, and key variables such as recurrence date and survival are well-documented. Second, while the sample size is relatively small, it is substantial for ACC studies, allowing complex modeling with covariates that capture a sig- nificant portion of patient heterogeneity, which previous stud- ies have not addressed (11).
The extended recruitment period, while essential for captur- ing long-term outcomes, presents inherent limitations due to changes in clinical practice over the years. For example, cases from earlier years are more likely to include long-term survi- vors, as patients with rapid progression may have been over- looked in earlier registries. To address this, we incorporated year of diagnosis as a covariate in all multivariable models and provided detailed visualizations of practice evolution, in- cluding the use of mitotane, monitoring of mitotanemia, and the role of multidisciplinary tumor boards. Despite these ef- forts, selection bias remains a limitation that readers should consider when interpreting the findings.
We were unable to determine whether the delay in recur- rence, without correlation to OS and with no apparent impact on cure, carries clinical implications (ie, extending symptom- free periods or improving quality of life).
Mitotane’s impact on cure rates may be influenced by sub- optimal treatment patterns and prolonged exposure to sub- therapeutic levels, challenges inherent to its long-term
administration. These patterns, while pragmatic in real-world settings, are difficult to optimize further due to the drug’s nar- row therapeutic window. Taken together, the data suggest that approximately two-thirds of participants achieved peak concentrations above the lower threshold of the therapeutic range, despite spending 72% of the treatment period below this margin. This aligns with the transient effect observed in our analysis. Attempts to extend therapy or intensify dosing to sustain therapeutic levels, however, risk exacerbating tox- icity, particularly neurological symptoms, asthenia, and gastrointestinal side effects.
Our study did not analyze radiotherapy dose or field, and the number of patients treated is small, introducing a degree of uncertainty. Finally, evolving clinical practices have led to missing values for variables like Ki-67 and Weiss index. We addressed this with multiple imputation which, while redu- cing bias, adds some uncertainty to parameter estimates.
In terms of clinical applicability, our study offers valuable guidance for decision-making. Long-term mitotane therapy appears most beneficial in patients with high-risk features, while it may be avoided in low-risk patients. Special caution is advised for older women, who may experience reduced bene- fit. Venous invasion should also be evaluated alongside other high-risk criteria, such as Ki-67 index and tumor size, to inform treatment strategies more effectively. Additionally, although our analysis did not demonstrate a clear relationship between therapeutic levels and outcomes, this lack of evidence should not be interpreted as proof of futility, and monitoring plasma mitotane levels remains clinically relevant. Exploring hazard rates after therapy discontinuation in other cohorts is war- ranted to validate these findings and optimize dosing strategies.
In conclusion, this study highlights the cytostatic role of mi- totane in delaying recurrence rather than achieving cure in ACC. These findings suggest clinically relevant interactions that support further investigation into optimizing treatment duration, dosing, and sequencing to maximize benefits for high-risk patients.
Acknowledgments
The authors are grateful to the ICARO registry researchers for their contributions to this study and to GETTHI and SEEN for their support in promoting it. Special thanks to Natalia Cateriano, Miguel Vaquero, and the IRICOM SL team for de- veloping, maintaining, and providing ongoing IT support for the registry’s web platform. To Miguel Ángel and Sofía Carmona, for reminding me that a little chaos can be the best “cure” for prolonged writing sessions.
Funding
This work is sponsored by the Grupo Español de Medicina Traslacional y Tumores Huérfanos e Infrecuentes (GETTHI) and the Sociedad Española de Endocrinología y Nutrición (SEEN). The database is funded by Esteve-HRA Pharm Rare Disease. The funders were not involved in the study design, data collection, analysis, interpretation, writing of this article, or the decision to submit it for publication.
Author Contributions
A.C .- B., P.J .- F., C.A .- E., N.V., and I.B.N. developed the pro- ject. A.C .- B. and P.J .- F. analyzed the data and drafted the
manuscript. The other authors recruited patients and pro- vided clinical information, feedback, and revisions to the manuscript. All authors contributed to the interpretation and discussion of the data and have read and approved the fi- nal manuscript.
Disclosures
A.C .- B. reports receiving lecture grants from Esteve, Lilly, and Astellas, travel support from Amgen, and research funding from HRA Pharma-Esteve. J.H.C. reports receiving funding from Novartis, Eisai, Ipsen, Bayer, and HRA Pharma- Esteve. P.J .- F. reports receiving honoraria for speakers’ bureau participation, and serving on advisory boards from Astellas, AstraZeneca, Bristol-Myers Squibb (BMS), HRA Pharma- Esteve, Merck Sharp & Dohme (MSD), Novartis, Nutricia, Pfizer, Rovi, Takeda, and Viatri. All other authors declare no competing interests related to the scope of this work.
Availability of Data and Materials
Original data generated and analyzed during this study are in- cluded in this published article or in the data repositories listed in References. Any citation, reference, or use of these materials should appropriately reference the original study.
Ethics Approval and Consent to Participate
The study is being conducted in accordance with the principles of the Declaration of Helsinki, the International Council for Harmonisation Guidelines for Good Clinical Practice, and lo- cal regulations. The study protocol was approved by the refer- ence CEIM of Hospital Universitario Central de Asturias on December 29th, 2017 (Ref: 218/17), and by the CEIMs of all participating centers. Written informed consent was ob- tained from all living patients.
References
1. Libé R, Huillard O. Adrenocortical carcinoma: diagnosis, prognos- tic classification and treatment of localized and advanced disease. Cancer Treat Res Commun. 2023;37:100759.
2. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer. 2008;113(11):3130-3136.
3. Nelson AA, Woodard G. Severe adrenal cortical atrophy (cyto- toxic) and hepatic damage produced in dogs by feeding 2,2-bis(parachlorophenyl)-1,1-dichloroethane (DDD or TDE). Arch Pathol. 1949;48(5):387-394.
4. Nichols J, Green HD. Effect of DDD treatment on metabolic re- sponse of dogs to ACTH injection. Am. J. Physiol .- Leg. Content. 1954;176(3):374-376.
5. Fassnacht M, Dekkers OM, Else T, et al. European Society of Endocrinology clinical practice guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European network for the study of adrenal tumors. Eur J Endocrinol. 2018;179(4):G1-G46.
6. De León DD, Lange BJ, Walterhouse D, Moshang T. Long-term (15 years) outcome in an infant with metastatic adrenocortical carcin- oma. J Clin Endocrinol Metab. 2002;87(10):4452-4456.
7. Luton JP, Cerdas S, Billaud L, et al. Clinical features of adrenocort- ical carcinoma, prognostic factors, and the effect of mitotane ther- apy. N Engl J Med. 1990;322(17):1195-1201.
8. Del Rivero J, Else T, Hallanger-Johnson J, et al. A review of mito- tane in the management of adrenocortical cancer. Oncologist. 2024;29(9):747-760.
9. Terzolo M, Angeli A, Fassnacht M, et al. Adjuvant mitotane treat- ment for adrenocortical carcinoma. N Engl J Med. 2007;356(23): 2372-2380.
10. Dickstein G, Shechner C, Arad E, Best L-A, Nativ O. Is there a role for low doses of mitotane (o,p’-DDD) as adjuvant therapy in adrenocort- ical carcinoma? J Clin Endocrinol Metab. 1998;83(9):3100-3103.
11. Tang Y, Liu Z, Zou Z, Liang J, Lu Y, Zhu Y. Benefits of adjuvant mitotane after resection of adrenocortical carcinoma: a systematic review and meta-analysis. Biomed Res Int. 2018;2018:9362108.
12. Huang H, Fojo T. Adjuvant mitotane for adrenocortical cancer-a recurring controversy. J Clin Endocrinol Metab. 2008;93(10): 3730-3732.
13. Terzolo M, Fassnacht M, Perotti P, et al. Adjuvant mitotane versus surveillance in low-grade, localised adrenocortical carcinoma (ADIUVO): an international, multicentre, open-label, randomised, phase 3 trial and observational study. Lancet Diabetes Endocrinol. 2023;11(10):720-730.
14. Hermsen IG, Fassnacht M, Terzolo M, et al. Plasma concentrations of o,p’DDD, o,p’DDA, and o,p’DDE as predictors of tumor re- sponse to mitotane in adrenocortical carcinoma: results of a retro- spective ENS@T multicenter study. J Clin Endocrinol Metab. 2011;96(6):1844-1851.
15. Corso CR, Acco A, Bach C, Bonatto SJR, De Figueiredo BC, De Souza LM. Pharmacological profile and effects of mitotane in adre- nocortical carcinoma. Br J Clin Pharmacol. 2021;87(7):2698-2710.
16. Therneau T, Crowson C, Atkinson E. Using time dependent cova- riates and time dependent coe_cients in the cox model. Survival Vignettes. 2017;2(3):1-25.
17. Edgar K, Jackson D, Rhodes K, Duffy T, Burman C-F, Sharples L. Frequentist rules for regulatory approval of subgroups in phase III trials: a fresh look at an old problem. Stat Methods Med Res. 2021;30(7):1725-1743.
18. Carmona-Bayonas A. Supplementary data (Tables & Figures) to article “Does adjuvant mitotane impact cure rates in adrenocortical carcinoma? Insights from the ICARO-GETTHI/SEEN registry”. Figshare. 2025. https://doi.org/10.6084/m9.figshare.28324220.v1
19. R: The R Project for Statistical Computing. Accessed October 21, 2024. https://www.r-project.org/
20. Carmona-Bayonas A. ICARO Adrenocortical Carcinoma Dataset. ICARO-GETTHI/SEEN Registry. Figshare. 2025. https://doi.org/ 10.6084/m9.figshare.28314812.v1
21. Carmona-Bayonas A. ICARO R Code for Data Analysis. ICARO- GETTHI/SEEN Registry. Figshare. 2025. https://doi.org/10.6084/ m9.figshare.28314815.v1
22. Tsuboi I, Kardoust Parizi M, Matsukawa A, et al. The efficacy of adjuvant mitotane therapy and radiotherapy following adrenalec- tomy in patients with adrenocortical carcinoma: a systematic re- view and meta-analysis. Urol Oncol Semin Orig Investig. 2024. Doi: 10.1016/j.urolonc.2024.09.014
23. Allegra S, Puglisi S, Brescia I, et al. Sex differences on mitotane con- centration and treatment outcome in patients with adrenocortical carcinoma. Life Basel Switz. 2021;11(3):266.
24. He S, Huang X, Zhao P, Zhang P. The prognosis difference between elderly and younger patients with adrenocortical carcinoma. Front Genet. 2023;13:1029155.
25. Lo CY, Van Heerden JA, Grant CS, Soreide JA, Warner MA, Ilstrup DM. Adrenal surgery in the elderly: too risky? World J Surg. 1996;20(3):368-374.
26. Vermeulen A, Deslypere JP, Schelfhout W, Verdonck L, Rubens R. Adrenocortical function in old age: response to acute adrenocorti- cotropin stimulation *. J Clin Endocrinol Metab. 1982;54(1): 187-191.
27. Cerquetti L, Bucci B, Marchese R, et al. Mitotane increases the radiotherapy inhibitory effect and induces G2-arrest in combined treatment on both H295R and SW13 adrenocortical cell lines. Endocr Relat Cancer. 2008;15(2):623-634.
28. Mete O, Erickson LA, Juhlin CC, et al. Overview of the 2022 WHO classification of adrenal cortical tumors. Endocr Pathol. 2022;33(1):155-196.
29. Jatoi I, Bandos H, Jeong J-H, et al. Time-Varying effects of breast cancer adjuvant systemic therapy. J Natl Cancer Inst. 2016;108(1): djv304.
30. Amico M, Van Keilegom I. Cure models in survival analysis. Annu Rev Stat Its Appl. 2018;5(1):311-342.
31. Joensuu H, Eriksson M, Sundby Hall K, et al. One vs three years of adjuvant imatinib for operable gastrointestinal stromal tumor: a randomized trial. JAMA. 2012;307(12):1265-1272.
32. Boichuk S, Parry JA, Makielski KR, et al. The DREAM complex mediates GIST cell quiescence and is a novel therapeutic target to enhance imatinib-induced apoptosis. Cancer Res. 2013;73(16): 5120-5129.
33. Eisenberg BL. The SSG XVIII/AIO trial: results change the current adjuvant treatment recommendations for gastrointestinal stromal tumors. Am J Clin Oncol. 2013;36(1):89-90.
34. Raut CP, Espat NJ, Maki RG, et al. Efficacy and tolerability of 5-year adjuvant imatinib treatment for patients with resected intermediate- or high-risk primary gastrointestinal stromal tumor: the PERSIST-5 clinical trial. JAMA Oncol. 2018;4(12): e184060.
35. Lin J-X, Chen Q-F, Zheng C-H, et al. Is 3-years duration of adju- vant imatinib mesylate treatment sufficient for patients with high-
risk gastrointestinal stromal tumor? A study based on long-term follow-up. J Cancer Res Clin Oncol. 2017;143(4):727-734.
36. Pedersen RN, Esen BÖ, Mellemkjær L, et al. The incidence of breast cancer recurrence 10-32 years after primary diagnosis. J Natl Cancer Inst. 2022;114(3):391-399.
37. Pan H, Gray R, Braybrooke J, et al. 20-Year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med. 2017;377(19):1836-1846.
38. Gnant M, Fitzal F, Rinnerthaler G, et al. Duration of adjuvant aromatase-inhibitor therapy in postmenopausal breast cancer. N Engl J Med. 2021;385(5):395-405.
39. Blok EJ, Kroep JR, Meershoek-Klein Kranenbarg E, et al. Optimal duration of extended adjuvant endocrine therapy for early breast cancer; results of the IDEAL trial (BOOG 2006-05). J Natl Cancer Inst. 2018;110(1):40-48.
40. Poli G, Guasti D, Rapizzi E, et al. Morphofunctional effects of mi- totane on mitochondria in human adrenocortical cancer cells. Endocr Relat Cancer. 2013;20(4):537-550.
41. Schteingart DE, Sinsheimer JE, Counsell RE, et al Comparison of the adrenalytic activity of mitotane and a methylated homolog on normal adrenal cortex and adrenal cortical carcinoma. Cancer Chemother Pharmacol. 1993;31(6):459-466.