EJE OXFORD
Clinical predictors of outcome in advanced adrenocortical carcinoma: a multicenter international ENSAT study
Alessandra Mangone (D 1,2, Barbara Altieri 3,4, Emanuele Ferrante 2, Irina Bancos 0 5, Michaela Luconi (D6, Barbara Ziółkowska7, Anja Barač Nekić8, Rossella Libe9, Filippo Ceccato (D 10,11, James F.H. Pittaway12, Marta Laganà13, Guido Di Dalmazi (D 14,15, Erika Peverelli1,2, Otilia Kimpel (D3, Bahar Bahrani Fard5, Letizia Canu ID6, Agnieszka Kotecka-Blicharz16, Darko Kastelan (D 8, Lucas Bouys (D9, Irene Tizianel10,11, Gillian Bennett12, Marc P. Schauer3, Yasir S. Elhassan @17, Mario Detomas ID3, Lorenzo Zanatta6, Maaz Sadiq18, Giovanna Mantovani @ 1,2 and Cristina L. Ronchi (D 17,19*
1Department of Clinical Sciences and Community Health, Dipartimento di Eccellenza 2023-2027, University of Milan, 20122, Milan, Italy
2Endocrinology Unit, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Via Francesco Sforza 35, 20122, Milan, Italy
3Division of Endocrinology and Diabetes, Department of Internal Medicine I, University Hospital, University of Würzburg, 97070 Würzburg, Germany 4Bavarian Cancer Research Center (BZKF), University Hospital, University of Würzburg, 97070, Würzburg, Germany
5Division of Endocrinology, Mayo Clinic, Rochester, Minnesota, 55902, United States
6Department Experimental and Clinical Biomedical Sciences, Endocrinology Section, University of Florence, 50139, Florence, Italy
7Second Clinic of Radiotherapy and Chemiotherapy, Maria Sklodowska-Curie National Research Institute of Oncology, 44-102, Gliwice Branch, Poland 8University Hospital Centre Zagreb, 10000, Zagreb, Croatia
9Department of Endocrinology and National Reference Center for Rare Adrenal Diseases, Hôpital Cochin, Assistance Publique Hôpitaux de Paris, 75014, Paris, France
10Department of Medicine DIMED, University of Padua, 35128, Padua, Italy
11Endocrine Unit, University Hospital of Padova, 35128, Padua, Italy
12Department of Endocrinology, St Bartholomew’s Hospital, EC1A 7BE, London, United Kingdom
13Medical Oncology Unit, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, ASST Spedali Civili, 25123, Brescia, Italy
14 Division of Endocrinology and Diabetes Prevention and Care, IRCCS Azienda Ospedaliero-Universitaria di Bologna, 40138, Bologna, Italy
15Department of Medical and Surgical Sciences (DIMEC), Alma Mater Studiorum University of Bologna, 40126, Bologna, Italy
16Department of Nuclear Medicine and Endocrine Oncology, Maria Sklodowska-Curie National Research Institute of Oncology, 44-102, Gliwice Branch, Poland
17 Department of Endocrinology, Queen Elizabeth Hospital Birmingham, University Hospitals Birmingham NHS Foundation Trust, B15 2GW, Birmingham, United Kingdom
18 Department of Oncology, Queen Elizabeth Hospital Birmingham, University Hospitals Birmingham NHS Foundation Trust, B15 2GW, Birmingham, United Kingdom
19 Department of Metabolism and Systems Science, University of Birmingham, Edgbaston B152TT, Birmingham, United Kingdom
*Corresponding author: Email: C.L.Ronchi@bham.ac.uk; Co-corresponding: Giovanna Mantovani, Email: giovanna.mantovani@unimi.it
Abstract
Objective Advanced adrenocortical carcinoma (ACC) is treated with mitotane alone or combined with cytotoxic chemotherapy, yet outcomes remain poor and prognostic models in this setting are lacking. This study aimed to evaluate the prognostic value of clinical parameters in a large cohort of patients with advanced ACC undergoing systemic therapy.
Methods Multicenter, international cohort study investigating 418 patients with advanced ACC (61.5% = women, median age = 52 years) from 11 centers. Patients received mitotane monotherapy (n = 161), etoposide + doxorubicin + cisplatin ± mitotane (n = 178), or second-line regimens (gemcitabine + capecitabine + mitotane or temozolomide + mitotane, n = 79). Variables included age, cortisol excess, performance status (ECOG-PS), tumor burden, and neutrophil-to-lymphocyte ratio (NLR) at start of therapy. Outcomes were overall survival (OS), time to progression (TTP), and best objective response. Results Tumor burden, cortisol excess, ECOG-PS, and NLR ≥5 independently predicted shorter OS (hazard ratio [HR] 1.55- 2.68). We developed an integrated ENSAT Risk Score for Advanced ACC combining these variables: tumor burden (0-2), cortisol excess (0/1), ECOG-PS (0-2), and NLR (0/1). A score >2 (poor-risk) was significantly associated with worse OS and TTP across all treatment groups (HRs for OS: 3.05-3.96; TTP: 2.53-3.08). It also predicted poorer response to mitotane (P <. 01) and second-line therapies (P= . 04).
Conclusions The ENSAT Risk Score for Advanced ACC is a practical, prognostic tool for patients with advanced ACC receiving systemic therapy. Based on accessible clinical and biochemical markers, it can support treatment decisions and facilitate informed discussions in routine care.
Keywords adrenal cancer, chemotherapy, mitotane, EDP, predictors
Significance statement
Pharmacological treatments for advanced adrenocortical carcinoma (ACC) remain largely ineffective, underscoring the urgent need for better prognostic tools.
Introducing the ENSAT Risk Score for Advanced ACC: We propose a novel, easy-to-use prognostic model developed from an international, multicenter cohort, combining key clinical and biochemical markers to guide treatment in advanced ACC. This score delivers powerful predictions accurately forecasting survival, disease progression, and response to both first- and second- line systemic therapies in a large patient population.
Ready for clinical integration: ENSAT Risk Score for Advanced ACC is designed for seamless use in routine practice, empowering clinicians and enhancing patient-centered decision-making in advanced ACC care.
Introduction
Adrenocortical carcinoma (ACC) is a rare and highly aggressive ma- lignancy with an incidence of 1-2 cases per 1 million inhabitants/ year and a median 5-year survival rate ranging from 10% to 60%.1-3 The most effective treatment for localized tumors is radical surgery; however, disease relapse is common. Moreover, a signifi- cant proportion of patients presents with metastatic or non- resectable disease, portending a dismal prognosis.4,5
For advanced ACC, the standard of care involves the adreno- lytic drug mitotane, either as monotherapy in cases with low tu- mor burden and more indolent disease, or in combination with etoposide, doxorubicin, and cisplatin (EDP) chemotherapy.2,6-8 However, the response rate to these treatments is limited, and efficacy of mitotane is subjected to reaching a therapeutic plas- ma level. Moreover, systemic therapy is often accompanied by considerable toxicity, which leads many patients to discontinue treatment or reduce dosages due to adverse events.2,6,7,9,10 In pa- tients with disease progression, several second-line regimens have been proposed, such as gemcitabine plus capecitabine (Gem-Cape),11 recommended by current guidelines, or more re- cently temozolomide (TMZ),12 though with even lower response rates. Given its rarity, ACC is classified as an orphan cancer, with limited options for new drug development. Alternative treat- ment strategies after failure of standard chemotherapy, includ- ing molecular-targeted therapies and immunotherapy,13-15 have shown limited clinical activity.
Due to the limited treatment options and poor outcomes in ad- vanced ACC, preserving quality of life by identifying potential res- ponders to standard therapies and avoiding unnecessary toxicity for non-responders is crucial.16 However, reliable predictors of treatment outcomes in advanced ACC are still lacking. 17-21
Previous studies have identified some negative prognostic or predictive factors in patients with metastatic ACC, including per- formance status, extent of metastatic lesions and involved tu- moral organs, mitotic index, and cortisol secretion.8,12,2-24 Moreover, preliminary series have suggested that inflammation- based scores (eg, neutrophil-to-lymphocyte ratio [NLR] and platelet-to-lymphocyte ratio [PLR]), may help identify patients with advanced disease more likely to benefit from first-line25 and second-line treatments.24
This study aimed to explore the prognostic value of easily as- sessable clinical parameters and in a large international cohort of patients with advanced ACC receiving systemic therapy.
Methods
Study design and population
This is a multicenter, international, retrospective cohort study conducted on behalf of the European Network for the Study of Adrenal Tumors (ENSAT, www.ensat.org). Only patients with ad- vanced ACC were included, defined as non-resectable or meta- static disease, treated with at least one of the following systemic therapies: mitotane monotherapy (at least 2 months) or mitotane + EDP or Gem-Cape or TMZ (at least one cycle). The list of the 11 participating ENSAT centers with number of cases included is reported in Table S1. Data were collected between October 2022 and October 2024.
Inclusion criteria were available clinical data at diagnosis and treatment initiation, blood cell counts (FBC) with white blood cell differential count at beginning of treatment (for inflammation-based-scores calculation), and follow-up data. We excluded patients with conditions that could potentially alter FBC including hematological diseases, immunomodulatory drugs, other active malignancies, severe cardiopathy, and active autoimmune disease. Patients who had already been included in previous stud- ies on clinical predictors in advanced disease were excluded.24,25
Treatment choices and radiological surveillance were decided by each participating center according to current European guidelines for management of ACC2 and through local multidis- ciplinary team discussion.
This study was conducted in accordance with the Declaration of Helsinki. The ENSAT ACC Scientific Board approved the study in October 2022 (Project 2022-001). All participating institutions ob- tained local ethics approval for recording of pseudonymized and standardized data in the ENSAT registry for use in adrenal tumor- related projects (www.ensat.org). The entire list of the participating centers and the details of the ethical approval for ENSAT multicen- ter studies are provided in the Table S1. All participants provided written informed consent prior participation to the study.
| Points | Definition |
|---|---|
| 1 | Small tumor burden (local recurrence <5 cm/small metastasis < 1 cm) |
| 2 | Local recurrence ≥ 5 cm or metastasis in 1-2 organs of small/intermediate size |
| 3 | Metastasis in >2 organs of intermediate/large size +/- local recurrence ≥ 5 cm |
| 4 | Advanced/diffuse multiple metastasis |
Clinical parameters
Demographic, clinical, biochemical, and radiological data were recorded at diagnosis, at systemic therapy initiation and during follow-up. Definitions of data collected at the time of diagnosis are detailed in Table S2, including the calculation of the S-GRAS score, as defined by Elhassan et al.4
The following clinical parameters at therapy initiation were in- vestigated: age, steroid secretion status (clinically and/or bio- chemically uncontrolled cortisol or androgen excess despite the ongoing medications, upon endocrine expert assessment), presence of symptoms (due to tumor mass, steroid hypersecre- tion, or cancer-related), Eastern Cooperative Oncology Group performance status (ECOG-PS),26 NLR (neutrophil count divided by lymphocyte count), PLR (platelet count divided by lympho- cyte count) and tumor burden (defined as detailed in Table 1). These parameters were collected short before (2-4 weeks) the ini- tiation of the treatment of interest. A total of 48/418 patients (11.5%) were already receiving mitotane in adjuvant setting when disease recurrence was recorded.
We also evaluated the number of patients who reached plasma mitotane target (≥14 mg/L) during treatment, maintaining these levels for at least 80% of time.27
Endpoints
Primary endpoints were overall survival (OS), defined as the interval from the initiation of treatment of interest to death or last follow-up visit, and time to progression (TTP), defined as the time elapsed from treatment initiation to the first radiologic- al evidence of disease progression or death.
Secondary endpoint was the best objective response to treat- ment, observed at any time during the treatment course, accord- ing to response evaluation criteria in solid tumors-RECIST1.128 (by cross sectional imaging; expert radiological review; and clin- ical judgment): progressive disease, stable disease, partial re- sponse, and complete response. Clinical benefit was defined as the sum of partial or complete response and stable disease.7
Statistical analysis
Descriptive statistics were used to report clinical variables. Qualitative variables are reported as absolute and relative (per- centage) frequencies. Continuous variables are presented as me- dian and IQR or range. Associations between categorical
variables were assessed by Chi-square or Fisher exact test, as ap- propriate. Survival analyses were performed using Cox propor- tional hazards models. Survival curves for OS and TTP were generated with the Kaplan-Meier method and outcomes were compared with log-rank test. Variables with potential prognostic value at univariable Cox regression (conservative P-value thresh- old ≤.2) were included in a multivariable Cox model. Results are given as hazard ratio (HR) with 95% CIs. The statistical signifi- cance was conventionally set at P < . 05 in the multivariable mod- el. Multivariable P-values were adjusted for covariates in the model.
For survival analysis, the cut-offs used to stratify according to inflammation-based-scores (calculated at the beginning of each treatment) were obtained from current literature on ACC: NLR = 5, PLR = 190. 24,25,29,30
The predictive accuracy of the test variable “ENSAT Risk Score for Advanced ACC’ (described below) was assessed with the receiver operator characteristic (ROC) curve in the entire co- hort, using “death” as the reference variable. The optimal ROC-derived cut-off (2.5) was identified using the Liu Method.31
To evaluate the stability and robustness of the univariate Cox findings, a non-parametric bootstrap procedure with 1000 repli- cations was implemented. In each bootstrap replication, a sam- ple of the same size as the original dataset was drawn with replacement from the original patient data. Bootstrap-derived standard errors, biases, and 95%CIs were subsequently calcu- lated to confirm the precision and reliability of HRs obtained from the original Cox model.
Model discrimination of prognostic scores [ENSAT Risk Score for Advanced ACC and S-GRAS,4 see below] was assessed using Harrell’s C concordance statistic (C-index, Stata command estat concordance). Differences between C-indices were tested using Somers’ D rank correlation, calculated from the linear predictors of each model (Stata command somersd). Finally, the incremen- tal prognostic value of the ENSAT Risk Score for Advanced ACC beyond S-GRAS was evaluated by comparing Cox regression models using a likelihood ratio test. Analyses were restricted to patients with complete data for both scores.
All analyses were performed using STATA v.17 and GraphPad Prism v.9.
Results
Overview of patients’ characteristics and response to treatment
A detailed flowchart for patient inclusion is shown in Figure S1. A total of 453 patients with advanced ACC were initially recruited from 11 ENSAT centers. Thirty-five patients were excluded due to incomplete data as per inclusion criteria. A final cohort of 418 eligible patients was enrolled (61.5% = women, median age =52 years), subdivided into 3 cohorts depending on the treatment received: mitotane monotherapy (mitotane cohort, n = 161), etoposide + cisplatin ± doxorubicin + mitotane (EDP co- hort, n = 178) or 2 different second-line chemotherapy schemes (n =79; Gem-Cape ± mitotane n = 56 or TMZ + mitotane, n = 23). Treatment change due to disease progression was undertaken in 85 patients (20%), who were therefore included in more than
one cohort. Within the EDP cohort, 31 patients (17.4%) under- went reduced etoposide + cisplatin + mitotane scheme omitting doxorubicin as per clinical decision; due to the limited number of patients they were not analyzed separately.
The baseline demographic and clinical characteristics of the patients included in the 3 cohorts, as well as survival data and response to treatment, are described in Table 2.
Around half of patients presented with metastatic disease at diagnosis-ENSAT stage IV (52.2% in the mitotane cohort, 56.7% in the EDP cohort, and 38% in the second-line cohort). Primary surgery was not performed in 16%, 17.4%, and 6% of patients in the 3 cohorts, respectively.
The median OS from the start of mitotane, EDP and second- line therapies was respectively of 22 (range 2-276 months), 16 (1-152) and 12.7 months (0.5-169). Median TTP was 5 (1-156), 5 (0.5-129), and 3 (1-30) months in the 3 respective cohorts. Overall, a clinical benefit from therapy was observed in 44.1% of patients treated with mitotane, 56.2% with EDP, and 34.2% with second-line therapies.
Predictors of overall-survival and time-to-progression
Univariable and multivariable survival analyses were performed in the entire cohort (n = 418, Table 3 for OS and Table S3 for TTP) and separately in the 3 cohorts of treatment (Table 4 for OS and Table S4 for TTP).
Overall, all variables except for age resulted predictive of OS, but at multivariable analysis only tumor burden, cortisol excess, ECOG-PS, and NLR maintained their significance (HR between 1.5 and 2.6). Tumor burden, cortisol excess and ECOG-PS also con- firmed their independent predictive role for TTP. In particular, tu- mor burden and ECOG-PS resulted the strongest predictor of both OS (HR 2.68 and 2.15, respectively) and TTP (HR 2.2 and 1.87, respectively).
Resection of the primary tumor also resulted associated with longer OS (P <. 01), but not with TTP (P =. 81). This variable was not included in the multivariate analysis due to the different time point at evaluation compared to other variables (ie, at time of diagnosis).
Complete data on mitotane levels during treatment were available for a subset of patients (289/418, 69%), and 41.9% of them reached targeting plasma levels for ≥ 80% of time. This percentage was similar among the 3 groups (42.2% in mi- totane, 38.9% in EDP and 51.4% in second-line cohorts, re- spectively). Patients in the mitotane group with levels in range had longer OS and TTP compared to those that did not (48.9 vs 31.2 months, P =. 014; and 17.7 vs 7.5 months, P= .006). This variable was not included in the multivariate ana- lysis due to the reduced number of available data and the dif- ferent time point at evaluation (ie, after the start of index treatment). There was no significant difference in the EDP or second-line line groups.
Bootstrap analysis
The bootstrap analysis confirmed the stability of the estimates obtained by univariate Cox regression analysis using the original
sample, with similar standard errors, low biases and overlapping confidence intervals for both OS and TTP (Tables S5 and 6).
Combined ENSAT risk score for advanced ACC
We selected the clinical variables that maintained a significant prognostic value at multivariable analyses on OS and used them to create a newly proposed combined ENSAT Risk Score for Advanced ACC (range 0-6), calculated as a sum of the follow- ing points: tumor burden (1=0, 2=1, 3/4=2), cortisol excess (absent =0, present=1), ECOG-PS (0=0, 1=1, 2/3=2), and NLR (<5=0, ≥5=1). The combined score showed a significantly better model fit than individual parameters, assessed by likelihood-ratio testing (P <. 001).
Figure 1 shows the Kaplan-Maier curves for OS and TTP for pa- tients in the 3 cohorts, stratified according to pre-treatment ENSAT Risk Score for Advanced ACC. We used a dichotomous variable based on the optimal score ROC-derived cut-off (1= combined score 0-2 = “favourable-risk group”; 2 = combined score > 2= “poor-risk group”) for the survival analysis. Importantly, high score >2 (poor-risk) resulted a strong predictor of shorter OS and TTP in the entire cohort (OS HR = 3.28, 95% CI =2.52-4.27, P <. 01; TTP HR =2.76, 95% CI = 2.13-3.59, P < .01), and separately in all the 3 cohorts of therapy (mitotane co- hort: OS HR = 3.05, 95% CI = 2.06-4.51, P <. 01; TTP HR = 2.68, 95% CI = 1.81-3.98, P <. 01; EDP cohort: OS HR = 3.38, 95% CI = 2.22-5.14, P <. 01; TTP HR=2.53, 95% CI = 1.68-3.84, P <. 01; second-line cohort: OS HR =3.96, 95% CI = 1.68-9.29, P <. 01; TTP HR = 3.08, 95% CI = 1.31-7.24, P <. 05).
Best response to therapy
We assessed the relationship between ENSAT Risk Score for Advanced ACC and response to therapy in all 3 categories of pa- tients (Figure 2). Patients in the mitotane cohort with score > 2 (poor-risk group) showed a poorer response to therapy com- pared to those in the favorable risk group (clinical benefit in 34.2% vs 55.9%; P =. 006). In the EDP cohort, a similar trend was (51.4% vs 64.8%, respectively, P = . 07). Finally, poor-risk cat- egory (score > 2) also predicted worse response to second-line therapies (clinical benefit in 29% vs 56.2%; P = . 04).
Comparative analysis with S-GRAS score
For completeness, we compared the prognostic performance of the S-GRAS score4,32 in the present cohort. Complete data re- quired to calculate S-GRAS were available in 266/418 patients (63.6%), specifically in 108, 106, and 52 patients in the mitotane, EDP, and second-line cohorts, respectively. Comparative ana- lyses were restricted to the subgroup of patients with both scores available.
The S-GRAS score was significantly associated with OS and TTP in the overall population (OS HR =1.4, 95% CI = 1.2-1.7, P <. 01; TTP HR = 1.2, 95% CI = 1.0-1.4, P =. 02). In subgroup ana- lyses, S-GRAS retained prognostic significance for OS in the mito- tane (OS HR = 1.4, 95% CI = 1.1-1.9, P =. 05) and EDP cohorts (OS
| Demographic data | Mitotane cohort | EDP cohort | Second-line cohort |
|---|---|---|---|
| Cases, n | 161 | 178 | 79 |
| Sex, F/M (% F) | 101/60 (63) | 109/69 (61) | 47/32 (59) |
| Clinical data at diagnosis | |||
| ENSAT tumor stage | |||
| I (%) | 6 (3.7) | 5 (2.8) | 3 (3.8) |
| II (%) | 34 (21.1) | 42 (23.6) | 25 (31.6) |
| III (%) | 31 (19.3) | 27 (15.2) | 20 (25.3) |
| IV (%) | 84 (52.2) | 101 (56.7) | 30 (38) |
| Unknown | 6 (3.7) | 3 (1.7) | 1 (1.3) |
| Surgery not performed (%) | 26 (16.1) | 31 (17.4) | 5 (6.3) |
| Median Ki67% (IQR) | 20 (10-30) | 20 (12-32) | 20 (10-34) |
| <10 (%) | 18 (11.1) | 20 (11.2) | 12 (15.2) |
| 10-19 (%) | 39 (24.2) | 34 (19.1) | 13 (16.4) |
| ≥20 (%) | 73 (45.3) | 76 (42.7) | 39 (49.3) |
| Unknown/not applicable | 31 (19.2) | 48 (26.9) | 15 (19) |
| Resection status | |||
| R0 (%) | 68 (42.2) | 71 (39.8) | 20 (25.3) |
| RX (%) | 20 (12.4) | 21 (11.7) | 6 (7.6) |
| R1 (%) | 21 (13) | 20 (11.2) | 5 (6.3) |
| R2 (%) | 21 (13) | 15 (8.4) | 3 (3.7) |
| Unknown/not applicable | 31 (19.2) | 51 (28.6) | 45 (57) |
| Clinical data at the beginning of systemic therapy for advanced disease | |||
| Median age (IQR), years | 54 (41-64) | 50 (40-61) | 52 (42-62) |
| Cortisol excess (%) | 60 (37.27) | 64 (36) | 17 (21.5) |
| ECOG performance status | |||
| 0 (%) | 73 (47.4) | 64 (38.3) | 12 (16) |
| 1 (%) | 54 (35.1) | 68 (40.7) | 41 (54.7) |
| 2 (%) | 25 (16.2) | 30 (18) | 13 (17.3) |
| 3 (%) | 2 (1.3) | 5 (3) | 9 (12) |
| Tumor burden | |||
| 1 (%) | 40 (24.8) | 18 (10.2) | 2 (2.5) |
| 2 (%) | 62 (38.5) | 56 (31.6) | 21 (26.6) |
| 3 (%) | 44 (27.3) | 67 (37.8) | 24 (30.4) |
| 4 (%) | 15 (9.3) | 36 (20.3) | 32 (40.5) |
| Survival data and response to therapy | |||
| Dead at last follow-up (%) | 109 (67.7) | 115 (64.6) | 60 (76) |
| Median survival from start of therapy (IQR), months | 22 (10-48) | 16 (8-30) | 12.7 (4.6-39) |
| Median time to progression from start of therapy (IQR), months | 5 (3-13) | 5 (2-14) | 3 (2-5) |
| Best objective response: | |||
| Progressive disease (%) | 85 (52.8) | 76 (42.7) | 51 (64.5) |
| Stable disease (%) | 45 (27.9) | 52 (29.2) | 23 (29.1) |
| Partial response (%) | 18 (11.2) | 40 (22.5) | 4 (5.1) |
| Complete response (%) | 8 (5.0) | 6 (3.4) | 0 (0) |
| Unknown | 5 (3.1) | 4 (2.25) | 1 (1.3) |
Abbreviations: F, female; M, male, ENSAT, European Network for the Study of Adrenal Tumors; ECOG, Eastern Cooperative Oncology Group; IQR, interquartile range; EDP, etoposide, doxorubicin and cisplatin.
HR = 1.9, 95% CI = 1.4-2.6, P <. 01), but not in the second-line cohort.
When directly compared, the ENSAT Risk Score for Advanced showed a higher HR for OS than S-GRAS (HR = 3.4 vs 1.4), main- taining its significance also in the second-line cohort. Discrimination assessed by the C-index was similar between
the 2 scores (OS C-index S-GRAS = 0.59 vs ENSAT Risk score C = 0.65, P =. 1).
Finally, S-GRAS was not significantly associated with TTP in in- dividual treatment cohorts and did not predict best objective re- sponse to therapy in any group. Therefore, no comparative analyses were performed for these endpoints.
Downloaded from https://academic.oup.com/ejendo/article/194/3/381/8509481 by WT Cox Information Services user on 03 April 2026
| Total cohort (n = 418) | Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Clinical variable | ||||||
| Age < 50 | 1 | Ref | ||||
| Age ≥ 50 | 1.2 | .95-1.52 | .13 | NS | ||
| ECOG Performance Status =0 | 1 | Ref | ||||
| ECOG Performance Status =1 | 1.82 | 1.36-2.43 | <. 001 | 1.53 | 1.14-2.06 | .005 |
| ECOG Performance Status ≥1 | 3.05 | 2.24-4.17 | <. 001 | 2.15 | 1.39-2.96 | <. 001 |
| ECOG Performance Status (continuous) | 1.75 | 1.5-2.0 | <. 001 | 1.46 | 1.25-1.72 | <. 001 |
| Cortisol excess (no) | 1 | Ref | ||||
| Cortisol excess (yes) | 2.38 | 1.87-3.03 | <. 001 | 1.79 | 1.39-2.31 | <. 001 |
| Tumor burden =1 | 1 | Ref | ||||
| Tumor burden =2 | 2.19 | 1.39-3.46 | .001 | 1.8 | 1.13-2.86 | .01 |
| Tumor burden ≥3 | 3.74 | 2.42-5.8 | <. 001 | 2.68 | 1.71-4.19 | <. 001 |
| Tumor burden (continuous) | 1.85 | 1.54-2.21 | <. 001 | 1.57 | 1.31-1.9 | <. 001 |
| Inflammation-based scores | ||||||
| NLR <5 | 1 | Ref | ||||
| NLR ≥ 5 | 2.04 | 1.59-2.63 | <. 001 | 1.55 | 1.19-2.02 | .001 |
| PLR < 190 | 1 | Ref | NS | |||
| PLR ≥ 190 | 1.26 | .99-1.59 | .05 | |||
Variables with a potential prognostic value at univariate Cox regression were included in the multivariate Cox model (P-value threshold ≤.2). Abbreviations: HR, hazard ratio; 95% CI, confidence interval, ECOG, Eastern Cooperative Oncology Group; NLR, neutrophil-to-lymphocyte-ratio; PLR, platelet-to-lymphocyte ratio; NS, not significant; EDP, etoposide, doxorubicin and cisplatin; NS, not significant. Bold values indicate statistically significant p- values.
Discussion
We present one of the largest studies to date on prognostic strati- fication in patients with advanced ACC, a rare malignancy char- acterized by heterogeneous clinical behavior and molecular profiles. The substantial cohort of 418 patients was made pos- sible through the collaborative efforts of the ENSAT network. This study holds significant clinical relevance, particularly in light of the limited efficacy of systemic therapies and the ongoing ab- sence of reliable predictive markers in ACC.17-21 In such a challen- ging therapeutic landscape, it is essential to prioritize quality of life by identifying patients likely to benefit from pharmacological treatments, minimizing unnecessary toxicity in non-responders, and facilitating informed discussions about treatment expecta- tions and potential benefits.16
We assessed clinical and biochemical parameters that are readily available in routine practice, including in patients who are inoperable and lack histopathological data. This approach is especially pertinent given that more than one-third of patients present with metastatic disease and are not candidates for pri- mary surgery.4,5 Furthermore, a considerable interval may elapse between the initial diagnosis and the progression to advanced disease, during which the tumor biology may evolve and the dis- ease become more aggressive. Consequently, prognostic indica- tors typically evaluated at time of diagnosis (primary surgery), such as the ENSAT tumor stage or the Ki67 index, may no longer accurately reflect the disease status at the time when systemic therapy is initiated for advanced disease. For these reasons, we focused on exploring newer, easily accessible and treatment start-specific parameters that may offer complementary value
in the advanced-disease setting. By focusing on the point of sys- temic treatment initiation, our study offers insights that are dir- ectly applicable to the clinical management of patients in advanced ACC.
We examined individual parameters previously reported to have prognostic value in advanced ACC and we identified varia- bles with independent prognostic significance-namely tumor burden, cortisol excess, ECOG-PS, and NLR-to develop a com- bined score, the ENSAT Risk Score for Advanced ACC. This score demonstrated strong prognostic and predictive performance in patients with advanced ACC undergoing standard first- and second-line systemic therapies. Its predictive value was consist- ent across key clinical outcomes, including overall survival, time to progression, and best response to therapy. Based on these findings, we propose that the ENSAT Risk Score for Advanced ACC may serve as a valuable tool for supporting clinical decision- making, enabling more effective patient counseling and selec- tion prior to initiating systemic therapy, thereby advancing a pre- cision medicine approach in ACC.
We evaluated the utility of the ENSAT Risk Score for Advanced ACC across 3 patient cohorts treated with different therapeutic regimens. Mitotane monotherapy is indicated for patients with lower tumor burden or more indolent disease, whereas systemic chemotherapy is recommended for those with more aggressive or disseminated disease. EDP-M is the preferred first-line chemo- therapy, as supported by the FIRM-ACT trial.7 As expected, pa- tient selection varied across treatment groups, with the EDP cohort exhibiting a higher tumor burden compared to the mito- tane cohort. While both treatments generally offer modest
| Mitotane cohort (n = 171) | Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Clinical variable | ||||||
| Age < 50 | 1 | Ref | ||||
| Age ≥ 50 | 1.58 | 1.07-2.33 | .02 | NS | ||
| ECOG Performance Status = 0 | 1 | Ref | ||||
| ECOG Performance Status = 1 | 1.87 | 1.2-2.89 | .005 | 1.41 | .89-2.23 | .14 |
| ECOG Performance Status ≥1 | 2.75 | 1.68-4.5 | <. 001 | 1.69 | .98-2.91 | .06 |
| ECOG Performance Status (continuous) | 1.67 | 1.31-2.12 | <. 001 | 1.3 | .99-1.70 | .05 |
| Cortisol excess (no) | 1 | Ref | ||||
| Cortisol excess (yes) | 2.88 | 1.95-4.24 | <. 001 | 1.9 | 1.22-2.97 | .005 |
| Tumor burden = 1 | 1 | Ref | ||||
| Tumor burden = 2 | 1.85 | 1.07-3.22 | .03 | 1.60 | .91-2.81 | .1 |
| Tumor burden ≥3 | 3.24 | 1.88-5.6 | <. 001 | 2.18 | 1.22-3.89 | .008 |
| Tumor burden (continuous) | 1.79 | 1.38-2.32 | <. 001 | 1.46 | 1.11-1.92 | .006 |
| Inflammation-based scores | ||||||
| NLR <5 | 1 | Ref | ||||
| NLR ≥ 5 | 2.36 | 1.52-3.66 | <. 001 | 1.75 | 1.09-2.81 | .02 |
| PLR < 190 | 1 | Ref | ||||
| PLR ≥ 190 | 1.27 | 0.87-1.86 | .021 | NS | ||
| EDP cohort (n = 178) | Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Clinical variable | ||||||
| Age < 50 | 1 | Ref | ||||
| Age ≥50 | .95 | .66-1.39 | .803 | NS | ||
| ECOG Performance Status = 0 | 1 | Ref | ||||
| ECOG Performance Status = 1 | 1.57 | .998-2.46 | .05 | 1.45 | .91-2.31 | .11 |
| ECOG Performance Status ≥1 | 3.25 | 1.99-5.28 | <. 001 | 2.46 | 1.50-4.05 | <. 001 |
| ECOG Performance Status (continuous) | 1.8 | 1.4-2.32 | <. 001 | 1.57 | 1.22-2.02 | <. 001 |
| Cortisol excess (no) | 1 | Ref | ||||
| Cortisol excess (yes) | 2.33 | 1.6-3.38 | <. 001 | 1.84 | 1.25-2.71 | .002 |
| Tumor burden = 1 | 1 | Ref | ||||
| Tumor burden = 2 | 2.64 | 1.02-6.78 | .04 | 1.96 | .76-5.07 | .17 |
| Tumor burden ≥3 | 4.73 | 1.91-11.72 | .001 | 3.09 | 1.23-7.81 | .02 |
| Tumor burden (continuous) | 1.97 | 1.43-2.7 | <. 001 | 1.67 | 1.20-2.34 | .002 |
| Inflammation-based scores | ||||||
| NLR <5 | 1 | Ref | ||||
| NLR ≥ 5 | 2.13 | 1.45-3.14 | <. 001 | 1.67 | 1.11-2.51 | .014 |
| PLR < 190 | 1 | Ref | ||||
| PLR ≥ 190 | 1.54 | 1.06-2.23 | .022 | NS | ||
| Second-line cohort (n = 79) | Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Clinical variable | ||||||
| Age < 50 | 1 | Ref | ||||
| Age ≥ 50 | 1.08 | .65-1.81 | .76 | NS | ||
| ECOG Performance Status = 0 | 1 | Ref | ||||
| ECOG Performance Status = 1 | 2.35 | .91-6.07 | .08 | 2.21 | .84-5.85 | .11 |
| ECOG Performance Status ≥1 | 3.51 | 1.32-9.32 | .01 | 3.2 | 1.13-9.0 | .03 |
| ECOG Performance Status (continuous) | 1.71 | 1.15-2.55 | .008 | 1.72 | 1.13-2.63 | .01 |
| Cortisol excess (no) | 1 | Ref | ||||
(continued)
Downloaded from https://academic.oup.com/ejendo/article/194/3/381/8509481 by WT Cox Information Services user on 03 April 2026
| Second-line cohort (n = 79) | Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Cortisol excess (yes) | 2.49 | 1.31-4.71 | .005 | 2.12 | 1.08-4.16 | 03 |
| Tumor burden = 1 | NA* | NA* | NA* | NA* | NA* | NA* |
| Tumor burden = 2 | NA* | NA* | NA* | NA* | NA* | NA* |
| Tumor burden ≥3 | NA* | NA* | NA* | NA* | NA* | NA* |
| Tumor burden (continuous) | 1.84 | 1.08-3.15 | .02 | 1.78 | 1.02-3.12 | .04 |
| Inflammation-based scores | ||||||
| NLR <5 | 1 | Ref | ||||
| NLR ≥ 5 | 1.42 | .82-2.45 | .21 | 1.1 | .60-2.01 | .76 |
| PLR < 190 | 1 | Ref | ||||
| PLR ≥ 190 | .83 | .49-1.43 | .50 | NS | ||
Abbreviations: HR, hazard ratio; 95% CI, confidence interval, ECOG, Eastern Cooperative Oncology Group; NLR, neutrophil-to-lymphocyte-ratio; PLR, platelet-to-lymphocyte ratio; NS, not significant; EDP, etoposide, doxorubicin and cisplatin. NS, not significant. NA* not applicable for limited number of events. Bold values indicate statistically significant p-values.
A
Overall survival - Mitotane cohort
B
1.00
Overall survival - EDP cohort
C
Overall survival - Second-line cohort
8.
1.00
-
5
0.75
0.75
0
median OS 33 months
8
0.50
median OS 27 months
0.50
0
median OS 34 months
3
0.25
3
0
median OS 13 months
0
median OS 12 months
median OS 10 months
DO
0.00
0
0.00
0
100
Analysis time
200
300
0
50
Analysis time
100
150
0
50
100
Analysis time
150
200
Favourable-risk
Poor-risk
p<0.01
Favourable-risk
Poor-risk
p<0.01
Favourable-risk
Poor-risk
p<0.01
D
Time To Progression - Mitotane cohort
E
Time To Progression - EDP cohort
F
1.00
Time To Progression - Second-line cohort
DI
~
1.00
7’s
0.75
0.75
0
median TTP 3.2 months
$ 0
0.50
median TTP 8 months
median TTP 7 months
0.50
2 0
0.25
2
median TTP 4 months
median TTP 4 months
median TTP 3 months
0.00
0.00
0
0
50
Analysis time
100
150
0
50
Analysis time
100
150
0
10
Analysis time
20
30
Favourable-risk-
Poor-risk
p<0.01
Favourable-risk
Poor-risk
p<0.01
Favourable-risk
Poor-risk
p<0.05
efficacy, individual responses vary considerably. In our cohort, approximately 50% of patients treated with EDP experienced clinical benefit, a relatively high proportion compared to existing literature.7,18,33 This discrepancy may partly reflect the retro- spective nature of our data collection, where non-responders may have been underrepresented due to incomplete records. Nonetheless, when the ENSAT Risk Score for Advanced ACC was assessed at the initiation of therapy, it effectively identified patients more likely to benefit from EDP: over 60% of those with a “favourable-risk” score achieved clinical benefit. This finding highlights the potential of the ENSAT Risk Score for Advanced ACC to support clinical practice by facilitating more informed
and individualized discussions with patients regarding treatment expectations.
Second-line chemotherapy is generally reserved for patients who are clinically fit and willing to pursue further treatment fol- lowing failure of EDP-M, which may introduce selection bias when evaluating response rates. According to current European guidelines, and depending on local availability and in- stitutional protocols, second-line options may include Gem-Cape, streptozotocin, temozolomide, immune checkpoint inhibitors, or cabozantinib.33 In this study, we focused on 2 regi- mens more commonly used across ENSAT centers: Gem-Cape and temozolomide. Consistent with previous findings, a low
MITOTANE cohort
EDP cohort
Second-line cohort
progression (n=85)
progression (n=76)
progression (n=51)
clinical benefit (n=71)
clinical benefit (n=98)
clinical benefit (n=27)
p=0.006
p=0.07
p=0.04
100
100-
100
75
75
75
% of patients
% of patients
% of patients
50
50-
50
25
25
25
0
0
T
0
Favourable-risk (n=84)
Poor-risk (n=72)
Favourable-risk (n=70)
Poor-risk (n=104)
Favourable-risk (n=16)
Poor-risk (n=62)
ENSAT Risk Score for Advanced ACC was associated with a great- er likelihood of clinical benefit from second-line therapy, with re- sponse rates of 56% in the “favourable-risk” group compared to 29% in patients with “poor-risk”.
During systemic chemotherapy, continuation of mitotane is recommended.2 The importance of maintaining mitotane levels within the therapeutic range to maximize efficacy has been well established.23,34 In line with guideline recommendations, mitotane levels should be monitored at multiple time points. Therefore, we used the metric “mitotane in range ≥80% of time” as a proxy for adequate treatment exposure. Complete data on mitotane control during therapy were available for ap- proximately 70% of patients, which slightly limited the statistical power of this sub-analysis. Among those with available data, only 41.9% achieved therapeutic levels for more than 80% of the treatment duration, consistent with previous reports.9,35 Unsurprisingly, patients who maintained therapeutic mitotane levels experienced longer OS and TTP. However, this association was observed only in the mitotane monotherapy group, with no significant differences in the other treatment cohorts. We chose not to include this variable in the multivariate analysis for several reasons. First, the number of patients with complete mitotane data was lower than for other variables. Second, due to the retro- spective nature of the study and the complexity of the data, the reliability of mitotane level measurements was limited. Finally, this variable reflects longitudinal treatment exposure, which dif- fers methodologically from the ENSAT Risk Score for Advanced ACC and its component parameters.
As expected, resection of the primary tumor, indicative of less disseminated disease at the time of diagnosis, was associated with longer survival. This parameter was not included in the multivariate analysis, as it represents a fixed baseline character- istic, in contrast to the ENSAT Risk Score for Advanced ACC var- iables, which were collected at the initiation of systemic therapy. This approach also helped prevent overrepresentation of the sur- gery variable among patients included in multiple treatment reg- imens, while dynamic variables more accurately reflect tumor
progression and allow for the inclusion of the same patient across different cohorts.
The S-GRAS score has been extensively validated as powerful prognostic tool for ACC but requires available histopathological data from primary tumors and was not designed to predict re- sponse to systemic treatment.4,32 Our ENSAT Risk Score for Advanced ACC was not conceived as a substitute of the S-GRAS score, but rather as a complementary tool addressing a different clinical setting. Nevertheless, for completeness, we compared the performance of our score with the S-GRAS score.4 The S-GRAS score retained prognostic value for overall survival in the present cohort, confirming the robustness of this established model. However, the S-GRAS score showed slightly weaker per- formance metrics in this context showing a non-significant role on response to treatment. Importantly, the ENSAT Risk Score for Advanced ACC could be applied to a much larger proportion of patients, since complete histopathological data at time of diagnosis were frequently unavailable.
This study has some limitations. Its retrospective design and the lack of centralized radiological and histopathological review may introduce variability. As with all prognostic studies, the po- tential confounding effects of additional treatments, such as pri- or or concurrent local therapies or adjuvant mitotane, on survival outcomes cannot be entirely excluded. Another limitation is the heterogeneity of the population, as patients were evaluated at different stages of their disease course and some individuals con- tributed to more than one treatment group over time. Furthermore, detailed information on the number of patients re- quiring dose reductions or treatment interruptions was not con- sistently available, which may have influenced dose intensity and, consequently, treatment response. Lastly, the relatively small size of the second-line treatment group (n = 79) limits the generalizability of the ENSAT Risk Score for Advanced ACC in this subset, underscoring the need for external validation in lar- ger cohorts. While this score was not designed to guide treatment selection, it may support a more informed discussion with pa- tients by providing individualized estimates of prognosis and
likelihood of benefit at the start of systemic therapy. Prospective studies are clearly required to validate its role and to explore whether it may inform therapeutic strategies in advanced ACC.
Despite these limitations, the study has several notable strengths. It presents a unique, large, and well-characterized co- hort of patients with advanced ACC, all treated at internationally recognized adrenal expert centers. Assembling such a substan- tial cohort in the context of a rare cancer reflects a significant col- laborative effort by the ENSAT network and provides a valuable resource for prognostic stratification in patients receiving sys- temic therapy. Additionally, the use of standardized data extrac- tion from the ENSAT registry, followed by systematic clarification of ad hoc queries, ensured data consistency and quality. Importantly, the ENSAT Risk Score for Advanced ACC is based on clinical parameters that are readily available in expert cen- ters, facilitating its integration into routine clinical practice.
In conclusion, the ENSAT Risk Score for Advanced ACC repre- sents a non-invasive, cost-effective, and easy-to-use tool for pre- dicting outcomes in patients with advanced ACC undergoing systemic therapy. It holds promise for supporting treatment de- cisions and supporting a more personalized approach to care in this rare malignancy.
Acknowledgments
We thank the COST Action CA20122 Harmonization for support- ive networking. We acknowledge the European Reference Network for Rare Endocrine Conditions (Endo-ERN).
Authors’ contributions
Alessandra Mangone (Data curation, Investigation, Writing- review & editing [equal], Writing-original draft [lead]), Barbara Altieri (Data curation, Validation, Visualization, Writing-review & editing [equal]), Emanuele Ferrante (Formal analysis, Methodology, Resources, Software, Writing-original draft [equal]), Irina Bancos (Data curation, Validation, Visualization, Writing-review & editing [equal]), Michaela Luconi (Data cur- ation, Validation, Visualization, Writing-review & editing [equal]), Barbara Ziólkowska (Data curation, Validation, Visualization, Writing-review & editing [equal]), Anja Barac Nekic (Data curation, Validation, Visualization, Writing-review & editing [equal]), Rosella Libe (Data curation, Validation, Visualization, Writing-review & editing [equal]), Filippo Ceccato (Data curation, Validation, Visualization, Writing-review & editing [equal]), James Pittaway (Data curation, Validation, Visualization, Writing-review & editing [equal]), Marta Laganà (Data curation, Validation, Visualization, Writing-review & edit- ing [equal]), Guido Di Dalmazi (Data curation, Validation, Visualization, Writing-review & editing [equal]), Erika Peverelli (Funding acquisition, Resources, Validation, Visualization [equal]), Otilia Kimpel (Data curation, Validation, Visualization, Writing-review & editing [equal]), Bahar Bahrani Fard (Data cur- ation, Validation, Visualization, Writing-review & editing [equal]), Letizia Canu (Data curation, Validation, Visualization, Writing-review & editing [equal]), Agnieszka Kotecka-Blicharz (Data curation, Validation, Visualization, Writing-review & edit- ing [equal]), Darko Kastelan (Data curation, Validation, Visualization, Writing-review & editing [equal]), Lucas Bouys
(Data curation, Validation, Visualization, Writing-review & edit- ing [equal]), Irene Tizianel (Data curation, Validation, Visualization, Writing-review & editing [equal]), Gillian Bennett (Data curation, Validation, Visualization, Writing-review & edit- ing [equal]), Marc P Schauer (Data curation, Validation, Visualization, Writing-review & editing [equal]), Yasir S Elhassan (Data curation, Validation, Visualization, Writing- review & editing [equal]), Mario Detomas (Data curation, Validation, Visualization, Writing-review & editing [equal]), Lorenzo Zanatta (Data curation, Validation, Visualization, Writing-review & editing [equal]), Maaz Sadiq (Validation, Visualization, Writing-review & editing [equal]), Giovanna Mantovani (Funding acquisition, Project administration, Supervision, Validation, Visualization, Writing-review & editing [equal]), and Cristina L Ronchi (Conceptualization, Supervision, Writing-review & editing [lead], Project administration, Writing-original draft [equal])
Supplementary material
Supplementary material is available at European Journal of Endocrinology online.
Conflict of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research re- ported. CR is recipient of a research grant from HRA Pharma Rare Disease (part of Esteve).
Funding
This research was funded by an AIRC (Associazione Italiana Ricerca Cancro) grant to EP (IG 2021-25920) and Ministero Italiano dell’Università e della Ricerca (Progetti di Ricerca di Interesse Nazionale -PRIN 20222KAYY5, PNRR M4.C2.1.1 - Next Generation EU funded by the European Community) to ML. This work has been supported by the Bavarian Cancer Research Center (BZKF) (BA).
Data avaiability
The data supporting the findings of this study are available from the corresponding author upon reasonable request. The raw data are not publicly available because of ethical restriction.
References
1. Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endocr Rev. 2014;35(2):282-326. https://doi.org/10.1210/er. 2013-1029
2. Fassnacht M, Dekkers OM, Else T, et al. European Society of Endocrinology clinical practice guidelines on the manage- ment 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. https://doi.org/10. 1530/EJE-18-0608
3. Kerkhofs TMA, Verhoeven RHA, der Zwan JMV, et al. Adrenocortical carcinoma: a population-based study on inci- dence and survival in The Netherlands since 1993. Eur J Cancer (Oxford, England: 1990). 2013;49(11):2579-2586. https://doi.org/10.1016/j.ejca.2013.02.034
4. Elhassan Y, Altieri B, Berhane S, et al. S-GRAS score for prog- nostic classification of adrenocortical carcinoma: an inter- national, multicenter ENSAT study. Eur J Endocrinol. 2021;186(1):25-36. https://doi.org/10.1530/EJE-21-0510
5. sPuglisi S, Calabrese A, Ferraù F, et al. New findings on pres- entation and outcome of patients with adrenocortical cancer: results from a national cohort study. J Clin Endocrinol Metab. 2023;108:2517-2525. https://doi.org/10.1210/clinem/dgad199
6. Fassnacht M, Assie G, Baudin E, et al. Adrenocortical carcin- omas and malignant phaeochromocytomas: ESMO-EURACAN clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2020;11:1476-1490. https://doi.org/10.1016/j.annonc.2022.12.006.
7. Fassnacht M, Terzolo M, Allolio B, et al. Combination chemo- therapy in advanced adrenocortical carcinoma. N Engl J Med. 2012;366:2189-2197. https://doi.org/10.1056/NEJMoa1200966
8. Terzolo M, Daffara F, Ardito A, et al. Management of adrenal cancer: a 2013 update. J Endocrinol Invest. 2014;37:207-217. https://doi.org/10.1007/s40618-013-0049-2
9. Megerle F, Herrmann W, Schloetelburg W, et al. Mitotane monotherapy in patients with advanced adrenocortical car- cinoma. J Clin Endocrinol Metab. 2018;103:1686-1695. https:// doi.org/10.1210/jc.2017-02591
10. Turla A, Laganà M, Grisanti S, et al. Supportive therapies in patients with advanced adrenocortical carcinoma submitted to standard EDP-M regimen. Endocrine. 2022;77:438-443. https://doi.org/10.1007/s12020-022-03075-y
11. Henning JEK, Deutschbein T, Altieri B, et al. Gemcitabine-based chemotherapy in adrenocortical carcin- oma: a multicenter study of efficacy and predictive factors. J Clin Endocrinol Metab. 2017;102:4323-4332. https://doi. org/10.1210/jc.2017-01624
12. Cosentini D, Badalamenti G, Grisanti S, et al. Activity and safety of temozolomide in advanced adrenocortical carcin- oma patients. Eur J Endocrinol. 2019;181:681-689. https:// doi.org/10.1530/EJE-19-0570
13. Cosentini D, Grisanti S, Volta AD, et al. Immunotherapy failure in adrenocortical cancer: where next? Endocr Connect. 2018;7:E5-E8. https://doi.org/10.1530/EC-18-0398
14. Tourneau CL, Hoimes C, Zarwan C, et al. Avelumab in pa- tients with previously treated metastatic adrenocortical car- cinoma: phase 1b results from the JAVELIN solid tumor trial. J Immunother Cancer. 2018;6:111. https://doi.org/10.1186/ S40425-018-0424-9
15. Altieri B, Ronchi CL, Kroiss M, Fassnacht M. Next-generation therapies for adrenocortical carcinoma. Best Pract Res Clin Endocrinol Metab. 2020;34:101434. https://doi.org/10.1016/j. beem.2020.101434
16. Steenaard RV, Michon LA, Haak HR. Health-related quality of life in adrenocortical carcinoma. Cancers (Basel). 2019;11: 1500. https://doi.org/10.3390/cancers11101500
17. Roca E, Berruti A, Sbiera S, et al. Topoisomerase 2a and thy- midylate synthase expression in adrenocortical cancer. Endocr Relat Cancer. 2017;24:319-327. https://doi.org/10. 1530/ERC-17-0095
18. Laufs V, Altieri B, Sbiera S, et al. ERCC1 as predictive biomark- er to platinum-based chemotherapy in adrenocortical carcin- omas. Eur J Endocrinol. 2018;178:181-188. https://doi.org/10. 1530/EJE-17-0788
19. Puglisi S, Calabrese A, Basile V, et al. Mitotane concentra- tions influence outcome in patients with advanced adreno- cortical carcinoma. Cancers (Basel). 2020;12(3):740. https:// doi.org/10.3390/cancers12030740.
20. Altieri B, Sbiera S, Herterich S, et al. Effects of germline CYP2W1*6 and CYP2B6*6 single nucleotide polymorphisms on mitotane treatment in adrenocortical carcinoma: a multi- center ENSAT study. Cancers (Basel). 2020;12:359. https://doi. org/10.3390/cancers12020359
21. Altieri B, Lalli E, Faggiano A. Mitotane treatment in adreno- cortical carcinoma: mechanisms of action and predictive markers of response to therapy. Minerva Endocrinol (Torino). 2022;47:203-214. https://doi.org/10.23736/S2724- 6507.21.03601-0
22. Libé R, Borget I, Ronchi CL, et al. Prognostic factors in stage III-IV adrenocortical carcinomas (ACC): an European network for the study of adrenal tumor (ENSAT) study. Ann Oncol. 2015;26:2119-2125. https://doi.org/10.1093/annonc/mdv329
23. Malandrino P, Ghuzlan AA, Castaing M, et al. Prognostic markers of survival after combined mitotane- and platinum- based chemotherapy in metastatic adrenocortical carcin- oma. Endocr Relat Cancer. 2010;17:797-807. https://doi.org/ 10.1677/ERC-09-0341
24. Grisanti S, Cosentini D, Laganà M, et al. Clinical prognostic factors in patients with metastatic adrenocortical carcinoma treated with second line gemcitabine plus capecitabine chemotherapy. Front Endocrinol (Lausanne). 2021;12:1-9. https://doi.org/10.3389/fendo.2021.624102
25. Mangone A, Altieri B, Detomas M, et al. Inflammation-based scores as predictors of treatment response in advanced adre- nocortical carcinoma. Endocr Relat Cancer. 2023;30:e220372. https://doi.org/10.1530/ERC-22-0372
26. Oken MM, Creech RH, Tormey DC, et al. Toxicity and re- sponse criteria of the eastern cooperative oncology group. Am J Clin Oncol. 1982;5:649-655. https://api.semanticscholar. org/CorpusID:2261448.
27. Terzolo M, Baudin AE, Ardito A, et al. Mitotane levels predict the outcome of patients with adrenocortical carcinoma treated adjuvantly following radical resection. Eur J Endocrinol. 2013;169:263-270. https://doi.org/10.1530/EJE- 13-0242
28. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer (Oxford, England : 1990). 2009;45: 228-247. https://doi.org/10.1016/j.ejca.2008.10.026
29. Bagante F, Tran TB, Postlewait LM, et al. Neutrophil-lymphocyte and platelet-lymphocyte ratio as pre- dictors of disease specific survival after resection of adreno- cortical carcinoma. J Surg Oncol. 2015;112:164-172. https:// doi.org/10.1002/jso.23982
30. Mochizuki T, Kawahara T, Takamoto D, et al. The neutrophil-to-lymphocyte ratio (NLR) predicts adrenocorti- cal carcinoma and is correlated with the prognosis. BMC Urol. 2017;17:1-5. https://doi.org/10.1186/s12894-017-0240-4
31. Liu X. Classification accuracy and cut point selection. Stat Med. 2012;31:2676-2686. https://doi.org/10.1002/sim.4509
32. Saygili ES, Elhassan YS, Prete A, Lippert J, Altieri B, Ronchi CL. Machine learning-based survival prediction tool for adreno- cortical carcinoma. J Clin Endocrinol Metab. 2025;110:e3185- e3192. https://doi.org/10.1210/clinem/dgaf096
33. Fassnacht M, Puglisi S, Kimpel O, Terzolo M. Adrenocortical carcinoma: a practical guide for clinicians. Lancet Diabetes Endocrinol. 2025;13:438-452. https://doi.org/10.1016/S2213- 8587(24)00378-4
34. Puglisi S, Calabrese A, Basile V, et al. Mitotane concentrations influence the risk of recurrence in adrenocortical carcinoma patients on adjuvant treatment. J Clin Med. 2019;8:1850. https://doi.org/10.3390/jcm8111850
35. Boilève A, Mathy E, Roux C, et al. Combination of mitotane and locoregional treatments in low-volume metastatic adrenocort- ical carcinoma. J Clin Endocrinol Metab. 2021;106:e4698-e4707. https://doi.org/10.1210/clinem/dgab449