Check for updates
The impact of adrenocortical carcinoma hormone secreting status as a predictor of poor survival: a systematic review and meta-analysis
Constantinos Nastos1 . Dimitrios Papaconstantinou1 . Anna Paspala1 . Nikolaos Pararas1 . Andromachi Vryonidou2 . Anastasia Pikouli1 . Eirini Chronopoulou3 . Anna Lechou1 . Melpomeni Peppa4 . Emmanouil Pikoulis1
Received: 17 March 2024 / Accepted: 10 October 2024 @ The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024
Abstract
Purpose Adrenocortical carcinoma (ACC) poses a significant challenge in healthcare due to its aggressive nature and rarity. Prior reports suggest a poorer prognosis associated with hormone-secreting neoplasms. This study aims to assess the impact of ACC hormonal status on patients’ oncologic survival.
Methods A comprehensive literature search of the Medline, Embase, Web of Science, CINAHL, CENTRAL and clinical- trials.gov databases was undertaken. Utilized data involved Hazard Ratios derived from multivariable analysis in order to minimize exposure to confounding bias. Included studies were subsequently meta-analyzed using a Random effects model. Results Twelve studies incorporating 4483 patients were included in the quantitative analysis. Hormonally active ACCs comprised 48% of the entire pooled patient cohort and were found to be associated with significantly worse Overall Survival (HR 1.57, 95% Confidence Interval 1.39-1.78, p<0.001). Disease-Free Survival was comparably impacted (HR 1.32, 95% CI 1.11-1.57, p<0.001). Furthermore, cortisol secreting ACCs specifically, were also found to be associated with a 48% increase in the hazard of death or disease recurrence. Interstudy statistical heterogeneity was minimal among evaluated outcomes.
Conclusions Hormone-producing ACCs exhibit a poorer prognosis compared to non-secreting counterparts, with a 57% increased risk of death and a 32% increased risk of recurrence. These findings support the hypothesis that hormone produc- tion signifies an adverse tumor-specific feature, particularly when leading to hypercortisolemia, indicating an aggressive disease phenotype.
Keywords Adrenocortical carcinoma . Hormone . Cortisol . Secreting . Meta-analysis
☒ Dimitrios Papaconstantinou Dimpapa7@hotmail.com
1 Third Department of Surgery, National and Kapodistrian University of Athens, Attikon University Hospital, 1 Rimini Str, Athens 12462, Greece
2 Department of Endocrinology and Diabetes Center, Hellenic Red Cross Hospital, Athens, Greece
3 Department of Pediatric Surgery, Agia Sophia Children’s Hospital, Athens, Greece
4 Endocrine Unit, Second Department of Internal Medicine Propaedeutic, Research Institute and Diabetes Center, Attikon University Hospital, National and Kapodistrian University of Athens, Athens, Greece
Introduction
Adrenocortical carcinoma (ACC) is a rare and aggressive malignancy, with an estimated incidence of 1 case per 1,000,000 people per year, a rate that has remained rela- tively stable during the past four decades [1, 2]. The wide- spread implementation of abdominal computed tomography scans (CT) in clinical practice contributed to the rise in the diagnoses of adrenal incidentalomas. However, ACC is identified in only a small minority of cases, affecting up to 8.7% of such patients [3]. Despite the surge in the diagnosis of incidental adrenal lesions, which would logically result in an increase in early-stage ACC diagnoses, the prognosis remains dismal, with a cited median overall survival of 17 months [2].
ACC presents a substantial challenge for healthcare pro- viders due to its biological aggressiveness and overall rarity.
This rarity, in turn, hampers the development of accurate management guidelines due to the lack of population-based evidence. In this context, delineating the primary tumor- specific determinants of survival can prove particularly valuable in assessing and enhancing patient prognosis in the current era of personalized medicine. While significant efforts have been dedicated to identifying disease prognosti- cators, leading to the development of composite risk scores for survival, such as the one proposed by the European Net- work for the Study of Adrenal Tumor (ENSAT) [4], their predictive capacity remains limited at present [5].
A prior meta-analysis conducted by Vanbrabant et al. [6] has recently directed attention to the significance of hor- monal production concerning long-term patient survival. This analysis indicates that hormone production is asso- ciated with a 71% increase in the observed risk of death, emphasizing the pivotal role of hormonal status in the man- agement of this disease. The current systematic review and meta-analysis serves a dual purpose: firstly, to provide an updated assessment of the impact of hormonal secretion on patient survival by evaluating emerging literature, and secondly, to offer a methodologically refined estimate of its impact by incorporating time-dependent data in the form of Hazard Ratios and thus provide a more accurate portrayal of the true significance of hormonal status in ACC prognosis.
Materials and methods
Literature search
A comprehensive literature search of the Medline, Embase, Web of Science, CINAHL, CENTRAL and clinicaltrials. gov databases was undertaken, from their inception until February 2024, using the search terms “adrenocortical car- cinoma”, “functional”, “secreting”, “cortisol”, “glucocor- ticoid”, “11-deoxycorticosteron”, “17-OH-progesterone”, “pregnenolone”, “androstenedione”, “androgen”, “testos- terone”, “hydroxytestosterone”, “aldosterone”, “cortisol”, “hypercortisolism” and “hormone”, combined with the Boolean operators AND/OR, in order to construct an all- encompassing search string as appropriate for each database.
The generated abstract list was screened independently by two authors (AP, AL) for potentially relevant studies, which were then evaluated in full-text. Inclusion criteria involved studies that reported on adult patients diagnosed with adre- nocortical carcinoma, where the hormonal production status was known, and treatment was predominantly conducted with curative intent. The predetermined exclusion criteria were: (1) studies including exclusively pediatric patients (less than 18 years of age), (2) non-clinical studies or case reports, (3) studies with overlapping patient populations,
(4) studies reporting outcomes using effect sizes other than Hazard Ratios extrapolated from multivariate analysis, (5) cohort studies including exclusively metastatic ACC cases and (6) studies including patients with benign lesions and/or incidentalomas, (7) studies with exclusively non-operative management of ACC. Reference lists of potentially eligible studies were further manually checked using the snowball- ing method, a process in which references from identified studies are examined to find additional relevant research [7].
Data extraction and outcomes of interest
The main outcomes of interest investigated were the patient Overall Survival (OS) and Disease-Free Survival (DFS). Hazard Ratios were selected as the preferred measure of effect to incorporate a time component in the analysis of oncologic survival. Secondary outcomes of interest were the study characteristics, patient baseline demographics, the size of involved tumors, the presence of metastatic disease, reported resection margin status, incidence and type of hor- mone producing tumors and the use of adjuvant mitotane therapy.
Data from eligible studies were extracted by two authors (AL, EC) and were entered into standardized excel spread- sheets (Microsoft, Redmond, Washington, USA) for further tabulation. A third author (DP) oversaw the completeness and accuracy of the extracted data.
The present study was conducted according to PRISMA guidelines [8] and was registered in the “International Pro- spective Register of Systematic Review” in 2022 (PROS- PERO ID: CRD42022358972).
Methodological quality assessment
Assessment of the methodological quality of involved stud- ies was performed using the ROBINS-I tool developed by the Cochrane collaboration [9]. This risk of bias tool assesses the methodological quality of non-randomized studies across seven categories: bias due to confounding, bias due to selection of participants, bias in classification of interventions, bias due to deviations of intended interven- tions, bias due to missing data, bias in measurement of out- comes, bias in selection of the reported results. In each case, the estimated risk of bias can be low, moderate or serious. Two authors (DP, NP) independently undertook the risk of bias assessment, with a third author (CN) acting as a referee in cases of disagreement.
Statistical analysis
All statistical analyses were performed using Stata v17 (StataCorp. 2021. Stata Statistical Software: Release 17.
College Station, TX: StataCorp LLC). Hazard Ratios (HR) and corresponding 95% confidence intervals (CI) were uti- lized for the quantitative data synthesis, while descriptive statistics were used for the presentation of secondary out- comes of interest. A random-effects model (DerSimonian and Laird) was a priori selected due to expected interstudy heterogeneity in terms of important survival prognostica- tors, such as the presence of extra-adrenal spread, the rates of R0 resectability and the use of adjuvant mitotane treat- ment. Statistical heterogeneity was explored using the Hig- gin’s 12 statistic, with values below 30% representing low heterogeneity, values between 30 and 60% representing moderate heterogeneity, values between 60 and 75% rep- resenting substantial heterogeneity and values above 75% representing considerable heterogeneity. The possibility of publication bias was visually assessed by inspection of gen- erated funnel plots; statistical tests for publication bias eval- uation were not utilized due to the small number of included studies. In all analyses, a p-value equal or less than 0.05 was considered statistically significant.
Results
After screening 1097 unique abstracts and evaluating the full text of 84 studies, 12 were deemed eligible for inclusion in the final quantitative analysis (Fig. 1). The majority of included studies originated from the USA, with one study reporting on Hungarian patients and another one being a multinational cohort of European and North American patients. A continuous 42-year interval of data collection was noted amongst included studies, dating as far back as 1979 up to 2021. Overall, 4483 patients were incorporated in the meta-analysis, of which 2153 (48%) presented with hormone producing adrenocortical carcinomas. The preva- lence of functionality amongst included studies ranged from 18 to 71.6%, with a median value of 52.5%.
A high-level overview of the baseline tumor and treat- ment characteristics in evaluated patients is available for review on Table 1.
Survival in hormone-secreting ACC
Amongst nine evaluated studies [4, 10-17], hormone secret- ing ACC were associated with significantly worse OS (HR
Identification of studies via databases and registers
Identification
Records identified from: Databases (n =2432) Reference lists (n =15)
Records removed before screening: Duplicate records removed (n = 1350)
Records screened (n = 1097)
Records excluded (n = 1013)
Screening
Reports sought for retrieval (n = 84)
Reports excluded: Patient overlap (n =7) Non-operative management (n = 14) Benign lesions/incidentalomas (n =29) Outcome reporting without Hazard Ratio (n =14) Exclusively metastatic disease (n =8)
Reports assessed for eligibility (n = 84)
Included
Studies included in meta-analysis (n = 12)
| Study | Year | Country | Interval of data collection | Num- ber of patients | Hormone produc- tion n (%) | Age | Sex (M/F) n (%) | Resected tumors n (%) | R0 resec- tion n (%) | Meta- static n (%) | Adjuvant mitotane therapy n (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Sada et al. [18] | 2024 | USA | 1997-2020 | 807 | 485 | 50±15 | 302 (37.4)/ 505 | 714 (88) | 436 | 248 | 423 |
| (60.1) | (62.6) | (61) | (31) | (52.4) | |||||||
| Toke et al. [10] | 2022 | Hungary | 2000-2021 | 74 | 47 (71.6) | 49.9±14.9 | 27 (36.5)/ | 68 (91.9) | 30 | 27 | 55 (74.3) |
| 47 (63.5) | (54.5) | (58.7) | |||||||||
| Daher et al. | 2022 | USA | 1998-2007 | 621 | 336 | N/a | 238 (38.3) / 383 | 502 | 265 | 475 | 421 |
| [11] | (54.1) | (61.7) | (80.8) | (53.2) | (76.5) | (67.8) | |||||
| Sada et al. [12] | 2019 | USA | 1997-2017 | 266 | 140 | 51.6±15.5 | 105 (39.5)/ | 224 (84) | 154 | 85 | 154 |
| (52.6) | 161 (60.5) | (85) | (32) | (57.9) | |||||||
| Postlewait et | 2016 | USA | 1993-2014 | 207 | 83 (43) | 51.3±15.2 | 80 (39)/ | 207 | 127 | 59 | 88 (43) |
| al. [13] | 127 (61) | (100) | (69) | (31) | |||||||
| Margonis et al. | 2016 | USA | 1993-2014 | 234 | 95 (40.6) | 52 | 90 (38.5)/ | 234 | 143 | 43 | 78 (42.2) |
| [20] | (44-63)* | 144 (61.5) | (100) | (68.4) | (18.4) | ||||||
| Libé et al. [4] | 2015 | Multinational | 2000-2009 | 444 | 144 (32) | N/a | 173 (39)/ | 384 | 150 | 234 | N/a |
| 271 (61) | (86.5) | (39) | (52) | ||||||||
| Livhits et al. | 2014 | USA | 1999-2008 | 367 | 66 (18) | 52.9±18.5 | 155 (42.2)/ | 254 | N/a | 167 | N/a |
| [15] | 212 (57.8) | (69.2) | (45.5) | ||||||||
| Else et al. [21] | 2014 | USA | 1979-2013 | 391 | 224 (57) | 47.4 | 158 (40)/ | 241 | N/a | 100 | 106 |
| (16.0- | 253 (60) | (61.6) | (26) | (40.2) | |||||||
| 83.3)* | |||||||||||
| Berruti et al. | 2014 | Multinational | 1990-2008 | 524 | 275 | 45 | 203 (38.8)/ | 524 | 524 | 0 | 251 |
| [16] | (52.5) | (18-77)* | 321 (61.2) | (100) | (100) | (47.9) | |||||
| Ayala-Ramirez | 2013 | USA | 1998-2011 | 330 | 138 | N/a | 118 (35.8) / | 275 | 255 | 85 | 235 |
| et al. [17] | (41.8) | 212 (64.2) | (83.3) | (77.3) | (25.8) | (71.2) | |||||
| Grubbs et al. | 2010 | USA | 1991-2008 | 218 | 120 (55) | N/a | 74 (33.9)/ | 218 | 206 | 100 | 22 (10) |
| [18] | 144 (66.1) | (100) | (94.5) | (26) |
*Values presented as median (range)
1.57, 95% CI 1.39-1.78, p<0.001, Fig. 2). DFS was com- parably worse amongst patients with hormone secreting neoplasms (HR 1.32, 95% CI 1.11-1.57, p<0.001, Fig. 3), as evident from the pooled analysis of three studies [13, 15, 17]. In both cases, no statistical heterogeneity was found to be present (I2=0%).
Survival in cortisol-secreting ACC
When the analysis was adjusted to cortisol secreting tumors specifically, reports from five studies [14, 15, 18-20] indicated worse pooled OS (HR 1.48, 95% CI 1.26-1.72, p<0.001, Fig. 4A), while pooled results from four studies similarly revealed significantly worse DFS (HR 1.48, 95% CI 1.22-1.78, p<0.001, Fig. 4B). Statistical heterogene- ity was absent in the former comparison (I2=0%) and low (I2=14%) in the latter.
Critical appraisal and risk of bias assessment
Overall, the risk of bias amongst included studies was seri- ous in two studies, moderate in nine and low in one (Table 2). In general, the risk for bias due to confounding was offset by exclusively utilizing survival outcomes derived from
multivariate analyses, which at least partially accounted for hidden confounders. Consequently, the risk of confound- ing was low in three studies that directly reported equiva- lent baseline characteristics in the hormone secreting and non-secreting groups, while it was judged to be moderate in all remaining cases in which no such baseline character- istics comparisons were made. Bias due to selection of par- ticipants was moderate in eight studies that either included unresectable cases in their patient cohorts or only evaluated cortisol secretion, ignoring other types of hormone-produc- ing ACC. A single study was found to be at serious risk for selection bias, since it only reported on cortisol-producing tumors, while also incorporating unresectable cases in its analysis. Finally, two studies were found to be at moder- ate risk for bias due to missing data, either because of los- ing more than 10% of their entire cohort to follow-up, or because they excluded patients from their final analysis for reasons unrelated to ACC hormonal status.
Visual assessment of generated funnel plots did not reveal any asymmetry in the evaluated primary and second- ary outcomes.
| Study | Hazard Ratio with 95% CI | Weight (%) | ||
|---|---|---|---|---|
| Toke et al, 2022 | 8.39 [ 1.27, | 55.32] | 0.42 | |
| Daher et al, 2022 | 1.61 [ 1.30, | 2.00] | 31.99 | |
| Sada et al, 2019 | 1.49 [ 1.04, | 2.13] | 11.55 | |
| Postlewait et al, 2016 | . | 2.30 [ 1.22, | 4.35] | 3.64 |
| Libé et al, 2015 | 1.50 [ 0.93, | 2.41] | 6.62 | |
| Livhits et al, 2014 | 1.66 [ 1.14, | 2.42] | 10.36 | |
| Berruti et al, 2014 | 1.55 [ 1.15, | 2.09] | 16.63 | |
| Ayala-Ramirez et al, 2013 | 1.40 [ 1.06, | 1.85] | 18.78 | |
| Overall | 1.57 [ 1.39, | 1.78] | ||
| Heterogeneity: 72 = 0.00, 12 = 0.00%, H2 = 1.00 | ||||
| Test of 0; = 0): Q(7) = 5.31, p = 0.62 Favors secreting | Favors non-secreting | |||
| Test of 0 = 0: z = 7.29, p = 0.00 | ||||
| 1 | 2 4 8 16 | 32 | ||
| Random-effects REML model | ||||
| Study | Hazard Ratio with 95% CI | Weight (%) | ||
|---|---|---|---|---|
| Postlewait et al, 2016 | 1.30 [ 0.82, 2.06] | 13.97 | ||
| Berruti et al, 2014 | 1.30 [ 1.04, 1.62] | 59.71 | ||
| Grubbs et al, 2010 | 1.38 [ 0.99, 1.93] | 26.32 | ||
| Overall | 1.32 [ 1.11, 1.57] | |||
| Heterogeneity: 72 = 0.00, 12 = 0.00%, H2 = 1.00 | ||||
| Test of 0; = 0): Q(2) = 0.09, p = 0.96 Favors secreting Test of 0 = 0: z = 3.18, p = 0.00 | Favors | non-secreting | ||
| 1 | 2 | |||
Random-effects DerSimonian-Laird model
Fig. 3 Forest plot of Disease-Free Survival in functional versus non-functional adrenocortical carcinomas
Discussion
The current study confirms the adverse prognostic impact of hormone production on both overall and disease-free sur- vival in patients with ACC. Specifically, tumors producing hormones were associated with a 57% increased risk in the hazard for death and a 32% increase in the hazard for devel- oping disease recurrence. These findings align with those reported in the previous meta-analysis on the topic [6], although the observed effect sizes in our study were more modest. While data on the specific type of hormone secreted by the evaluated ACCs were limited, a similar detrimental effect on survival was observed in cortisol-secreting tumors, which carried a 48% increase in the hazard for death or dis- ease recurrence.
When investigating the causal relationship between hor- monal secretion and poor survival, it becomes imperative to consider potential confounding factors. Many reports have associated hormone-secreting ACCs with larger tumor sizes and a higher propensity for metastatic spread [18, 19], both
well-known risk factors linked to poor survival [21]. To mitigate the impact of confounding bias, our study exclu- sively employed data derived from multivariate analyses that accounted for such factors. This approach minimizes the exposure of the observed effect sizes to confounding influences, suggesting that hormone production, be it corti- sol or otherwise, independently contributes to a detriment in oncologic survival.
In the case of cortisol-secreting ACCs, the causal rela- tionship with poor survival is more evident. Excess corti- sol not only increases the risk of morbidity [19] but also induces a generalized state of immunosuppression, thereby favoring tumor spread [22]. However, the underlying patho- physiological mechanism explaining poor survival in ACCs secreting other types of hormones is less clear. Although recent years have highlighted several key genetic mutations in driver genes [22] however, none convincingly explain the observed disease aggressiveness.
Given the scarcity of available data, the influence of hor- mone secretion, apart from cortisol, on prognosis remains
| A Study | Hazard Ratio with 95% CI | Weight (%) | eight %) | ||
|---|---|---|---|---|---|
| Sada et al, 2024 | 1.47 [ 0.91, 2.38] | 10.46 | 1.75 | ||
| Margonis et al, 2016 | 1.20 [ 0.74, 1.94] | 10.52 | 3.85 | ||
| Livhits et al, 2014 | 1.66 [ 1.14, 2.42] | 16.88 | ).26 | ||
| Berruti et al, 2014 | 1.55 [ 1.15, 2.09] | 27.09 | 9.14 | ||
| Else et al, 2014 | 1.43 [ 1.10, 1.86] | 35.05 | |||
| Overall | 1.48 [ 1.26, 1.72] | ||||
| Heterogeneity: 12 = 0.00, 12 = 0.00%, H2 = 1.00 | |||||
| Test of 0; = 0): Q(4) = 1.25, p = 0.87 Favors cortisol-secreting Test of 0 = 0: z = 4.91, p = 0.00 1 | |Favors non-secreting 2 | ||||
Random-effects REML model
2
4
| B Study | Hazard Ratio with 95% CI | Weight (%) | ||
|---|---|---|---|---|
| Sada et al, 2024 | 2.09 [ 1.07, 4.07] | 7.61 | ||
| Margonis et al, 2016 | 2.05 [ 1.16, 3.61] | 10.36 | ||
| Berruti et al, 2014 | 1.30 [ 1.04, 1.62] | 49.52 | ||
| Else et al, 2014 | 1.49 [ 1.11, 2.00] | 32.51 | ||
| Overall | 1.48 [ 1.22, 1.78] | |||
| Heterogeneity: 12 = 0.01, 12 = 14.36%, H2 = 1.17 | ||||
| Test of 01 = 0; Q(3) = 3.56, p = 0.31 Favors cortisol-secreting | Favors non-secreting | |||
| Test of 0 = 0: z = 4.05, p = 0.00 | ||||
Random-effects REML model
Fig. 4 Forest plots of Overall Survival (A) and Disease-Free Survival (B) in cortisol secreting versus non-functional adrenocortical carcinomas
| Study | R1 | R2 | R3 | R4 | R5 | R6 | R7 | Overall |
|---|---|---|---|---|---|---|---|---|
| Ayala-Ramirez et al. [17] | Moderate | Moderate | Low | Low | Low | Low | Low | Moderate |
| Berruti et al. [16] | Low | Moderate | Low | Low | Low | Low | Low | Moderate |
| Daher et al. [11] | Moderate | Moderate | Low | Low | Low | Low | Low | Moderate |
| Else et al. [21] | Moderate | Low | Low | Low | Low | Low | Low | Moderate |
| Grubbs et al. [18] | Moderate | Low | Low | Low | Low | Low | Low | Moderate |
| Livhits et al. [15] | Moderate | Serious | Low | Low | Low | Low | Low | Serious |
| Margonis et al. [20 | Low | Low | Low | Low | Low | Low | Low | Low |
| Postlewait et al. [13] | Moderate | Low | Low | Low | Moderate | Low | Low | Moderate |
| Sada et al. [18] | Low | Moderate | Low | Low | Low | Low | Low | Moderate |
| Sada et al. (2019) | Low | Moderate | Low | Low | Low | Low | Low | Moderate |
| Toke et al. [10] | Moderate | Moderate | Low | Low | Serious | Low | Low | Serious |
| Libé et al. [4] | Moderate | Moderate | Low | Low | Low | Low | Low | Moderate |
uncertain and could not be presently evaluated within the scope of the present meta-analysis due to the same litera- ture constraints. Although cumulative analysis suggests an increased risk of death regardless of the hormone type produced, it is important to note that cortisol-secreting neo- plasms predominate in most of the included patient series.
Margonis et al. [19] specifically linked cortisol production to increased length of hospital stay, elevated postoperative adrenal insufficiency, and higher R1 resection rates, effects not observed with other functional types. In another recent multicentric trial involving 807 patients, cortisol secretion was confirmed to be associated with reduced R0 resection
rates, while androgen and mineralocorticoid-secreting neo- plasms exhibited resection rates equivalent to non-secretory ACCs [18]. Interestingly, the same study reported that mixed androgen/cortisol-producing tumors were associ- ated with the highest hazard of death. However, when only R0-resected cases were considered, cortisol-secreting ACCs were the sole hormonally active subtype significantly linked to an increased hazard for disease recurrence. As of now, evidence on which secreting subtype is worse in terms of patient survival remains inconclusive.
Another important consideration is that hormone-secret- ing ACCs are often symptomatic, and as a result, they may be diagnosed at an earlier disease stage compared to their hormonally silent counterparts. In the report by Sada et al. [18], secreting tumors demonstrated an increased propensity for lymphovascular invasion and metastatic spread, indicat- ing an aggressive biological behavior. This was observed despite equivalent Ki67% levels compared to non-secreting neoplasms. Such findings provide further support for the consideration of complementary lymphadenectomy in cases of ACC, including hormone-secreting tumors, to reduce the risk of incomplete disease clearance.
Undoubtedly, one of the most established staging sys- tems employed for ACC is the one proposed by the Euro- pean Network for the Study of Adrenal Tumors (ENSAT). This classification system categorizes adrenocortical malig- nancies based on size, adjacent organ invasion, and the presence of lymphatic or distant metastases [23]. Although externally validated [5], efforts for further refinement have been ongoing. Some initiatives aimed to optimize patient prognostic stratification within the ENSAT/TNM systems [24]. However, these staging systems fall short in account- ing for tumor-specific risk factors that significantly influ- ence survival [25]. In 2015, Libé et al. [4] introduced a set of five tumor and patient-specific parameters known as the GRAS parameters (Grade, Resection completeness, Age, and Symptoms). These parameters were utilized to enhance the prognostic accuracy of the ENSAT classification, par- ticularly for stage III and IV patients. This represents a shift in approach by applying the principles of personalized medicine in the management of ACC. The integration of the GRAS and ENSAT staging systems appears to have height- ened the prognostic efficacy of existing frameworks, under- scoring the importance of incorporating parameters such as ACC secretory activity into standardized patient staging.
Several limitations inherent in the present analysis need acknowledgment. Firstly, the use of hazard ratios derived from multivariable analyses, while necessary to mitigate the impact of confounders, exposes the analysis to potential selective reporting bias. In the evaluated studies, investiga- tors were more likely to conduct a multivariable analysis if the univariable results were statistically significant. As a
result, studies with potentially non-significant results might have been omitted due to a lack of reported data, while oth- ers were excluded due to a lack of data reporting in the form of HR, thus potentially exposing our results to publication bias. Additionally, the absence of high-quality prospective or propensity-matched studies, stemming from the overall rarity of ACC and the associated problem of small sample sizes, makes the obtained results susceptible to selection and confounding biases that cannot be currently addressed. Another notable limitation of this study is that it primar- ily includes patients who were treated with curative intent, thereby excluding non-operable patients, including those with and without Cushing’s syndrome. While our inclusion criteria specified that studies must involve a majority of patients undergoing curative intent treatment, several of the included studies also reported a minority of patients receiv- ing non-operative treatment. This variation may introduce confounding factors that could affect the overall conclu- sions regarding the efficacy of curative intent interventions. Consequently, these limitations may impact the applica- bility of our findings to the broader ACC population and should be considered when interpreting the results. Finally, the scarcity of data regarding oncologic survival for differ- ent secretory subtypes precludes any meaningful subgroup analysis on this topic.
In conclusion, hormone-producing ACCs demonstrate a poorer prognosis compared to their non-secreting coun- terparts. Hormone production is associated with a 57% increased risk of death and a 32% increased risk of recur- rence. Specifically, cortisol secretion is linked to a 48% increase in the hazard for either death or disease recurrence following surgery. These findings support the hypothesis that hormone production represents an adverse tumor-spe- cific feature, likely indicating an aggressive disease pheno- type, especially when leading to hypercortisolemia. In this context, the secretory status of ACCs may serve as a valu- able prognostic indicator within existing staging systems.
Author contributions C.N .: Conception and design, manuscript revi- sion and overall supervisionD.P .: Writing of the original draft, statis- tical analyses, supervision, risk of bias assessmentA.Pa .: Writing of the original draft, statistical analysesN.P .: Designed the search algo- rithm, risk of bias assessment, manuscript revisionA. V .: Study design, manuscript revisionA.Pi: Literature search, figure editingE.C .: Writ- ing of the original draft, data extractionA.L .: Literature search, data extractionM.P .: Study design, manuscript revisionE.P .: Study design, supervision and correction of the manuscript.
Funding None.
Data availability Data is provided within the manuscript.
Declarations
COI/disclosures None.
Competing interests The authors declare no competing interests.
References
1. Sharma E, Dahal S, Sharma P et al (2018) The characteristics and trends in Adrenocortical Carcinoma: a United States Population based study. J Clin Med Res 10:636. https://doi.org/10.14740/ JOCMR3503W
2. Kerkhofs TMA, Verhoeven RHA, Van Der Zwan JM et al (2013) Adrenocortical carcinoma: a population-based study on inci- dence and survival in the Netherlands since 1993. Eur J Cancer 49:2579-2586. https://doi.org/10.1016/J.EJCA.2013.02.034
3. Ebbehoj A, Li D, Kaur RJ et al (2020) Epidemiology of adrenal tumours in Olmsted County, Minnesota, USA: a population-based cohort study. Lancet Diabetes Endocrinol 8:894-902. https://doi. org/10.1016/S2213-8587(20)30314-4
4. Libé R, Borget I, Ronchi CL et al (2015) Prognostic factors in stage III-IV adrenocortical carcinomas (ACC): an European Net- work for the study of adrenal tumor (ENSAT) study. Ann Oncol 26:2119-2125. https://doi.org/10.1093/ANNONC/MDV329
5. Lughezzani G, Sun M, Perrotte P et al (2010) The European Net- work for the Study of Adrenal Tumors staging system is prognos- tically superior to the international union against cancer-staging system: a north American validation. Eur J Cancer 46:713-719. https://doi.org/10.1016/J.EJCA.2009.12.007
6. Vanbrabant T, Fassnacht M, Assie G, Dekkers OM (2018) Influ- ence of hormonal functional status on survival in adrenocortical carcinoma: systematic review and meta-analysis. Eur J Endocri- nol 179:429-436. https://doi.org/10.1530/EJE-18-0450
7. Wohlin C, Kalinowski M, Romero Felizardo K, Mendes E (2022) Successful combination of database search and snowballing for identification of primary studies in systematic literature stud- ies. Inf Softw Technol 147:106908. https://doi.org/10.1016/J. INFSOF.2022.106908
8. Page MJ, Mckenzie JE, Bossuyt PM et al (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. PLoS Med 18:e1003583
9. Sterne JA, Hernán MA, Reeves BC et al (2016) ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interven- tions. BMJ 355. https://doi.org/10.1136/BMJ.I4919
10. Tőke J, Uhlyarik A, Lohinszky J et al (2022) Prognostic factors and mitotane treatment of adrenocortical cancer. Two decades of experience from an institutional case series. Front Endocrinol (Lausanne) 13. https://doi.org/10.3389/FENDO.2022.952418
11. Daher M, Varghese J, Gruschkus SK et al (2022) Temporal trends in outcomes in patients with Adrenocortical Carcinoma: a multi- disciplinary referral-center experience. J Clin Endocrinol Metab 107:1239-1246. https://doi.org/10.1210/CLINEM/DGAC046
12. Sada A, Asaad M, Bews KA et al (2020) Comparison between functional and non-functional adrenocortical carcinoma. Surgery 167:216-223. https://doi.org/10.1016/J.SURG.2019.04.066
13. Postlewait LM, Ethun CG, Tran TB et al (2016) Outcomes of adjuvant mitotane after resection of Adrenocortical Carcinoma: a 13-Institution study by the US Adrenocortical Carcinoma Group. J Am Coll Surg 222:480-490. https://doi.org/10.1016/J. JAMCOLLSURG.2015.12.013
14. Livhits M, Li N, Yeh MW, Harari A (2014) Surgery is associated with improved survival for adrenocortical cancer, even in meta- static disease. Surgery 156:1531-1541. https://doi.org/10.1016/J. SURG.2014.08.047
15. Berruti A, Fassnacht M, Haak H et al (2014) Prognostic role of overt hypercortisolism in completely operated patients with adre- nocortical cancer. Eur Urol 65:832-838. https://doi.org/10.1016/J. EURURO.2013.11.006
16. Ayala-Ramirez M, Jasim S, Feng L et al (2013) Adrenocortical carcinoma: clinical outcomes and prognosis of 330 patients at a tertiary care center. Eur J Endocrinol 169:891-899. https://doi. org/10.1530/EJE-13-0519
17. Grubbs EG, Callender GG, Xing Y et al (2010) Recurrence of adrenal cortical carcinoma following resection: surgery alone can achieve results equal to surgery plus mitotane. Ann Surg Oncol 17:263-270. https://doi.org/10.1245/S10434-009-0716-X
18. Sada A, Foster TR, Al-Ward R et al (2024) The effect of hormonal secretion on survival in adrenocortical carcinoma: a multi-center study. Surgery 175. https://doi.org/10.1016/J.SURG.2023.04.070
19. Margonis GA, Kim Y, Tran TB et al (2016) Outcomes after resection of cortisol-secreting adrenocortical carci- noma. Am J Surg 211:1106-1113. https://doi.org/10.1016/J. AMJSURG.2015.09.020
20. Else T, Williams AR, Sabolch A et al (2014) Adjuvant therapies and patient and tumor characteristics associated with survival of adult patients with adrenocortical carcinoma. J Clin Endocrinol Metab 99:455-461. https://doi.org/10.1210/JC.2013-2856
21. Scollo C, Russo M, Trovato MA et al (2016) Prognostic factors for Adrenocortical Carcinoma outcomes. Front Endocrinol (Lau- sanne) 7:25. https://doi.org/10.3389/FENDO.2016.00099
22. Baechle JJ, Hanna DN, Sekhar KR et al (2021) Integrative com- putational immunogenomic profiling of cortisol-secreting adre- nocortical carcinoma. J Cell Mol Med 25:10061-10072. https:// doi.org/10.1111/JCMM.16936
23. Fassnacht M, Dekkers OM, Else T et al (2018) European Soci- ety of Endocrinology Clinical Practice Guidelines on the man- agement of adrenocortical carcinoma in adults, in collaboration with the European Network for the study of adrenal tumors. Eur J Endocrinol 179:G1-G46. https://doi.org/10.1530/EJE-18-0608
4. David SO, Krieg S, Esposito I et al (2023) A revised version of the TNM classification leads to optimized predictive performance in patients with Adrenocortical Carcinoma. Horm Metab Res 55:227-235. https://doi.org/10.1055/A-2042-2431
25. Asare EA, Wang TS, Winchester DP et al (2014) A novel staging system for adrenocortical carcinoma better predicts survival in patients with stage I/II disease. Surgery 156:1378-1386. https:// doi.org/10.1016/J.SURG.2014.08.018
Publisher’s note Springer Nature remains neutral with regard to juris- dictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.