Check for updates
Revisiting the AJCC staging system of adrenocortical carcinoma
O. Abdel-Rahman1 (D
Received: 13 March 2021 / Accepted: 16 June 2021 @ Italian Society of Endocrinology (SIE) 2021
Abstract
Objective To evaluate the performance characteristics of AJCC 7th and 8th staging systems among patients with adrenal cortical carcinoma.
Methods Surveillance, Epidemiology, and End Results (SEER) 18-registry was accessed and patients with adrenocortical carcinoma who were diagnosed 2010-2015 with complete information about AJCC 7th staging system were included. AJCC 8th staging system information was then reconstructed for each patient using available TNM staging variables. Kaplan-Meier overall survival estimates, multivariable Cox regression analysis, and concordance index (C-statistic) were used to examine the performance characteristics of both staging systems.
Results A total of 574 patients with a diagnosis of adrenocortical carcinoma were included in the current analysis. Using Kaplan-Meier survival estimates, overall survival was compared among different AJCC stages for both versions; and the P value was significant (<0.001) for both comparisons. C-statistic was then calculated for both staging systems and the results were as follows: for AJCC 7th version: 0.726 (95% CI 0.683-0.769); and for AJCC 8th version: 0.745 (95% CI 0.704-0.786). Patients with M1 disease (stage IV according to AJCC 8th edition) were then divided according to the extent of distant metas- tases into single versus multiple sites of metastases. Using Kaplan-Meier survival estimates, patients with a single site of metastases have better overall survival (P=0.006). A C-statistic for a hypothetical modification of AJCC 8th staging system subdividing stage IV patients into IVA and IVB based on the number of metastatic sites was: 0.753 (95% CI 0.713-0.794). Conclusions There is a minimal difference in the prognostic performance between both versions of the AJCC staging system. Subdivision of stage IV cancer into stage IVA and IVB (according to the number of organs with metastatic deposits) should be considered in subsequent versions of adrenocortical carcinoma staging.
Keywords Staging · AJCC . Prognosis · Outcomes
Introduction
Adrenocortical carcinoma is a rare disease entity with lim- ited information about prognostic variables and treatment strategies [1, 2]. Factors shown to play a prognostic role among those patients include anatomic stage, grade (through Weiss classification and other grading systems), type of surgical resection (R0 versus R1), Ki 67, and hormonal hypersecretion (particularly cortisol hypersecretion) [3-6]. The rarity of this disease entity has limited our ability to obtain confirmatory studies regarding the prognostic value
of different prognostic variables or reaching out a clear con- sensus about the efficacy of different treatments.
American Joint Committee on Cancer (AJCC) first intro- duced a staging system for adrenocortical carcinoma with the 7th staging system (2009). This staging system was designed along the lines of the 2004 international union against cancer staging classification [7]. The stage group- ing was later modified in 2017 with the publication of the AJCC 8th system. The main differences between both stag- ing systems are detailed in Table 1, but the main change has been in the regrouping of stage IV disease to include only patients with M1 disease in the 8th version (after it included some patients with locally advanced M0 disease in the 7th version). This re-organization was done along the lines of the previously published ENSAT (European network for the study of adrenal tumors) staging system [8]. It is unclear if these changes in the AJCC staging system will be
☒ O. Abdel-Rahman omar.abdelsalam@ahs.ca
1 Department of Oncology, Cross Cancer Institute, University of Alberta, Edmonton, AB T6G 1Z2, Canada
| Stage | AJCC 7th system | AJCC 8th system |
|---|---|---|
| I | T1N0M0 | T1N0M0 |
| II | T2N0M0 | T2N0M0 |
| III | T1-2N1M0 | T1-2N1M0 |
| OR T3N0M0 | OR T3-4N0-1,X M0 | |
| IV | T4,N0,M0 OR T3-4,N1,M0 OR: Any T Any N M1 | Any T Any N M1 |
T1 tumor 5 cm or less in size; invasion absent, T2 tumor greater than 5 cm in size; invasion absent, T3 any size with local invasion but not invading adjacent organs, T4 any size that invades adjacent organs or large blood vessels
associated with improved prognostic performance. Moreo- ver, the ENSAT staging system was first proposed in 2009 based on a retrospective study from the German adrenocorti- cal carcinoma registry. This registry included patients who were diagnosed more than 30 years ago (between 1986 and 2007). Thus, there is a need to re-assess this system in the context of the current diagnostic and therapeutic landscape of adrenocortical carcinoma.
Within the 8th edition of the AJCC staging system, there has been increasing recognition of the heterogeneity of stage IV among many solid tumors; and the concept of the oligo- metastatic disease seems to be affecting sub-divisions of stage IV [9]. That being said, this was not reflected in the AJCC 8th version of adrenocortical carcinoma. Highlight- ing the differences between different prognostic subgroups within stage IV cancer can play an important role in guid- ing potentially radical approaches to oligo-metastatic disease and in standardizing the reporting within clinical research. This is an aspect that needs to be examined (and potentially developed) within the current AJCC staging system for adrenocortical carcinoma.
Surveillance, Reporting, and End Results (SEER) data- base represents an appropriate venue to study such a rare disease entity. It does represent approximately 28% of the US cancer patient population and it enjoys high credibility through rigorous quality assurance and continuous updates. Thus, the SEER database has been employed in this study to evaluate the performance characteristics of AJCC 7th and 8th staging systems for adrenal cortical carcinoma.
Methods
Data sources
The current study is derived from SEER 18-registry database [10]; with the following inclusion criteria: (1) diagnosis with adrenocortical carcinoma; (2) diagnosis year 2010-2015
(as AJCC 7th staging was available for patients diagnosed during this period); (3) complete information about AJCC 7th staging system. Data cutoff for this study cohort was up to December 2017. For patients with duplicate records of adrenocortical carcinoma, only the first record was included in the current study.
Data collection
The following data were collected where available: age at diagnosis, sex, race, laterality, histology, AJCC 7th stage dis- tribution, AJCC 8th stage distribution (reconstructed from available TNM staging variables), and surgical treatment(s). Sites of metastatic disease (including lung, liver, bone, brain, and distant lymph nodes) were also reported for patients who have M1 disease at the time of diagnosis. Mitotic index, systemic therapy, and radiation therapy information was not available in the current SEER registry.
The main endpoints of the current study included overall survival (defined as the time from adrenocortical carcinoma diagnosis till death because of any reason) and cancer-spe- cific survival (defined as the time from adrenocortical car- cinoma diagnosis till death from adrenocortical carcinoma). Cancer-specific survival was assessed through the cause- specific death classification variable.
Statistical analysis
Descriptive statistics were initially conducted to evaluate the frequency and distribution of different baseline characteris- tics within the study cohort. Kaplan-Meier survival estimates and log-rank testing were then used to evaluate overall sur- vival according to both versions of the AJCC staging sys- tem. Multivariable Cox regression analyses were then used to examine pair-wise cancer-specific survival estimates accord- ing to both versions of the AJCC staging system. Factors adjusted for in these multivariable Cox regression models (in addition to AJCC stage) included age at diagnosis, sex, race, laterality, and surgical treatment. Repeated variable coding was used for each of the AJCC staging system variables to identify pair-wise comparisons between different stages.
C-statistic (concordance index) was then calculated for both staging systems using death attributed to adrenocortical carcinoma as the dependent variable. All statistical analyses were conducted by SPSS software (version 23.0; IBM; NY).
Results
Patients’ characteristics
A total of 574 patients with a diagnosis of adrenocorti- cal carcinoma were included in the current analysis. The
mean age was 53.39 (SD: 19.18). The majority of included patients were females (60.3%), white (83.1%), and left- sided (51.7%). According to the AJCC 7th stage, the most common stage is stage IV (51%); and according to the 8th edition, the most common stage is stage IV (41.2%). The most common site of metastatic disease is in the lung (23.3%), and the majority of patients have complete surgi- cal resection (64.6%) (Table 2). Mean survival follow-up is 26.1 months; SD 25.64.
| Variables | N (%) |
|---|---|
| Age (mean; SD) | 53.39; 19.18 |
| Sex | |
| Males | 228 (39.7) |
| Females | 346 (60.3) |
| Race | |
| White | 477 (83.1) |
| Black | 50 (8.7) |
| American-Indian | 3 (0.5) |
| Asian/Pacific Islander | 40 (7) |
| Others | 4 (0.7) |
| Laterality | |
| Right | 259 (45.1) |
| Left | 297 (51.7) |
| Bilateral/unknown | 18 (3.2) |
| AJCC 7th stages | |
| Stage I | 27 (4.7) |
| Stage II | 173 (30.1) |
| Stage III | 81 (14.2) |
| Stage IV | 293 (51) |
| AJCC 8th stages | |
| Stage I | 27 (4.7) |
| Stage II | 173 (30.1) |
| Stage III | 138 (24) |
| Stage IV | 236 (41.2) |
| Sites of metastases (for M1 patients) | |
| Lung | 134 (23.3) |
| Liver | 126 (22) |
| Bone | 50 (8.7) |
| Brain | 5 (0.9) |
| Distant lymph nodes | 37 (6.4) |
| Surgery to the primary | |
| None | 177 (30.8) |
| Local destruction/excision | 16 (2.8) |
| Total resection | 371 (64.6) |
| Unknown | 10 (1.7) |
Survival outcomes according to the two AJCC versions (7th and 8th)
Using Kaplan-Meier survival estimates, overall survival was compared among different AJCC stages for both versions; and the P value was significant (<0.001) for both compari- sons (Fig. 1a, b). Overall survival for stages I and II (which were identical in definition for both staging systems) was overlapping.
Using multivariable Cox regression analysis, cancer- specific death hazard (using death from adrenocortical car- cinoma as the dependent variable and adjusted for age, sex, race, laterality, and surgical treatment) was evaluated for both staging systems (Fig. 2a, b). For the 7th AJCC staging
a
Overall survival according to AJCC 7th stages
1.0-
7th stage
7|
I
IV
0.8-
I-censored
Il-censored
Ill-censored
IV-censored
Survival
0.6-
P<0.001
0.4-
0.2-
0.0
0
20
40
60
80
100
Months since diagnosis
b
Overall survival according to AJCC 8th stage
1.0-
8th stage
1
711
III
IV
0.8-
I-censored
Il-censored
III-censored
IV-censored
Survival
0.6
P<0.001
0.4-
0.2-
0.0
0
20
40
60
80
100
Months since diagnosis
a
Adjusted cancer-specific death hazard according to AJCC 7th edition
3.0
7th
stage
2.5
II
III
IV
2.0
P values:
Hazard
Stage I versus II: 0.992
Stage Il versus III: < 0.001
1.5
Stage III versus IV: 0.011
1.0
0.5
0.0
0
20
40
60
80
100
Months since diagnosis
b
Adjusted cancer-specific death hazard according to AJCC 8th edition
4
8th stage
Il
III
3
IV
Hazard
P values:
Stage I versus II: 0.985
Stage Il versus III: < 0.001
N
Stage III versus IV: < 0.001
1
0
0
20
40
60
80
100
Months since diagnosis
system, the following comparisons were significant: Stage II versus stage III (HR 0.48; 95% CI 0.32-0.71; P <0.001) and stage III versus stage IV (HR 0.62; 95% CI 0.43-0.89; P=0.011). For the 8th AJCC staging system, the follow- ing comparisons were significant: Stage II versus stage III (HR 0.45; 95% CI 0.32-0.64; P <0.001) and stage III versus stage IV (HR 0.56; 95% CI 0.41-0.77; P <0.001). For both staging systems, stage I versus stage II comparison was not significant (P=0.992 for 7th stage; and P=0.985 for 8th stage). Within Fig. 2a and b, stage I and II are overlapping, so that it is hard to distinguish both stages from each other.
C-statistic was then calculated for both staging systems and the results were as follows: for AJCC 7th version: 0.726 (95% CI 0.683-0.769); and for AJCC 8th version: 0.745 (95% CI 0.704-0.786).
Subdivision of patients with M1 disease (AJCC 8th stage IV) according to the number of organs with metastatic deposits
Reviewing the reported sites of distant metastases in the cohort (lung, liver, bone, brain, and distant lymph nodes), patients were classified into two categories: patients with a single site of distant metastases (proposed to be stage IVA) and patients with more than one site of distant metas- tases (proposed to be stage IVB). Using Kaplan-Meier survival estimates, patients with the proposed stage IVA have better overall survival (P=0.006) (supplementary Fig. 1). Within multivariable cox regression analysis lim- ited to patients with M1 disease (adjusted for age, sex, race, and laterality), single site of metastatic disease was
associated with better cancer-specific survival (HR 0.63; 95% CI 0.47-0.85).
A C-statistic for a hypothetical modification of AJCC 8th staging system subdividing stage IV patients into IVA and IVB (as described above) was: 0.753 (95% CI 0.713-0.794).
Discussion
The current study evaluates the prognostic performance of the AJCC 8th edition of adrenocortical carcinoma staging compared to the 7th edition in a real-world, population- based study from the US. It suggested that there is a minimal difference in the prognostic performance between both ver- sions of the AJCC staging system. Subdivision of stage IV patients into those with stage IVA versus stage IVB (accord- ing to the number of organs with metastatic deposits) should be considered in subsequent versions of adrenocortical car- cinoma staging.
The almost identical overall and cancer-specific survival outcomes of stage I and II patients in the current study (con- trary to the previous German study upon which the ENSAT staging system was proposed) deserve more analysis. Pos- sible reasons included (1) the relatively small size of patients labeled as stage I in the current cohort; therefore, it might have been difficult to discern their survival outcomes com- pared to stage II patients; (2) more widespread use of sal- vage locoregional and systemic therapies among patients with relapsed stage II (which would have compensated for previously observed prognostic differences in older studies). In any case, this point deserves further consideration within other datasets; if this observation is confirmed, we should redefine the distinction between stage I and II to be more prognostically relevant to current day practice.
The findings of the current study confirm the heteroge- neity of stage IV adrenocortical carcinoma. This was rec- ognized already in current management guidelines for this disease which suggests consideration of aggressive cytore- ductive surgery among patients with limited resectable met- astatic disease [11]. This clinical recognition should also be reflected in the staging system for this disease to ensure standardization of disease reporting.
It is important to acknowledge some limitations in the current analysis; first: the sample size of the current study is small. While this is understandable given the rarity of this disease entity; this can still be a potential source of bias. It has to be remembered, however, that the landmark Ger- man study based on which ENSAT (and AJCC 8th) staging system was adopted included a fewer number of patients (416 patients) than the number of patients in the current study. Second, other relevant prognostic information is not available within the SEER database. These include resec- tion (R) status, hormonal secretion status, and systemic and
locoregional treatments (beyond surgery) administered to recruited patients. That said, non-surgical treatments for adrenocortical carcinoma have been associated with mod- est improvement in patients’ outcomes, and they are not expected to markedly impact the outcomes of those patients [12]. Third, information about performance status and comorbidity is unfortunately absent from the SEER data- base. This is adding another layer of potential bias in the current study. These limitations need to be weighed against the strengths of the current study; most importantly, the rela- tively current nature of data collection (compared to limited prior studies in the same field).
AJCC 8th staging systems have witnessed a considerable move towards incorporating biomarkers within the staging system of many solid tumors [13, 14]. Although our under- standing of the biological complexity of adrenocortical car- cinomas is still in its infancy, the incorporation of promising biological candidates should be considered in subsequent versions of AJCC staging systems as this would improve our clinical approach to those patients.
In conclusion, there is a minimal difference in prognostic performance between both versions of the AJCC staging system. Subdivision of stage IV cancer into stage IVA and IVB (according to the number of organs with metastatic deposits) should be considered in subsequent versions of adrenocortical carcinoma staging. Moreover, incorporation of information about resection status, ki67 index, and hor- monal secretion status could be helpful in predicting the outcomes of patients with adrenocortical carcinoma.
Supplementary Information The online version contains supplemen- tary material available at https://doi.org/10.1007/s40618-021-01618-0.
Funding None.
Declarations
Conflict of interest Advisory board; Eisai Canada, Lilly Canada, and Roche Canada.
Informed consent As this study is based on a publicly available data- base without identifying patient information, informed consent was not needed.
References
1. Else T, Kim AC, Sabolch A, Raymond VM, Kandathil A, Caoili EM et al (2014) Adrenocortical carcinoma. Endocr Rev 35(2):282-326
2. Fassnacht M, Kroiss M, Allolio B (2013) Update in adrenocortical carcinoma. J Clin Endocrinol Metab 98(12):4551-4564
3. Fassnacht M, Dekkers OM, Else T, Baudin E, Berruti A, de Krijger R et al (2018) European society of endocrinology clini- cal 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 179(4):G1-g46
4. Owen DH, Patel S, Wei L, Phay JE, Shirley LA, Kirschner LS et al (2019) Metastatic adrenocortical carcinoma: a single institutional experience. Horm Cancer 10(4-6):161-167
5. Papotti M, Libè R, Duregon E, Volante M, Bertherat J, Tissier F (2011) The Weiss score and beyond-histopathology for adreno- cortical carcinoma. Horm Cancer 2(6):333-340
6. Libé R, Borget I, Ronchi CL, Zaggia B, Kroiss M, Kerkhofs T et al (2015) Prognostic factors in stage III-IV adrenocortical carcino- mas (ACC): an European network for the study of adrenal tumor (ENSAT) study. Ann Oncol 26(10):2119-2125
7. DeLellis RA (2004) Pathology and genetics of tumours of endo- crine organs. IARC
8. Fassnacht M, Johanssen S, Quinkler M, Bucsky P, Willenberg HS, Beuschlein F et al (2009) Limited prognostic value of the 2004 International Union against cancer staging classification for adrenocortical carcinoma. Cancer 115(2):243-250
9. Amin MB, Edge SB (2017) AJCC cancer staging manual. Springer
10. Surveillance, Epidemiology, and End Results (SEER) Program (2020) SEER*Stat Database: incidence-SEER research data, 18 registries, Nov 2019 sub (2000-2017)-linked to county attrib- utes-time dependent (1990-2017) income/rurality, 1969-2018
counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2020, based on the November 2019 submission. Available at: www.seer.cancer.gov. Accessed 24 Nov 2020
11. Fassnacht M, Assie G, Baudin E, Eisenhofer G, de la Fouchardiere C, Haak HR et al (2020) Adrenocortical carcinomas and malig- nant phaeochromocytomas: ESMO-EURACAN clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 31(11):1476-1490
12. Vaidya A, Nehs M, Kilbridge K (2019) Treatment of adrenocorti- cal carcinoma. Surg Pathol Clin 12(4):997-1006
13. Abdel-Rahman O (2018) Validation of American joint committee on cancer eighth staging system among prostate cancer patients treated with radical prostatectomy. Ther Adv Urol 10(2):35-42
14. Abdel-Rahman O (2018) Validation of the 8th AJCC prognostic staging system for breast cancer in a population-based setting. Breast Cancer Res Treat 168(1):269-275
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.