Check for updates
ORIGINAL RESEARCH
Prognostic Role of Ki-67 in Adrenocortical Carcinoma After Primary Resection: A Retrospective Mono- Institutional Study
Fuxun Zhang D . Fan Zhang . Zhihong Liu . Kan Wu .
Yuchun Zhu · Yiping Lu
Received: June 17, 2019 @ Springer Healthcare Ltd., part of Springer Nature 2019
ABSTRACT
Introduction: Adrenocortical carcinoma (ACC) is a rare malignancy with poor prognosis. It is vitally important to predict prognosis and restrict unnecessary adjuvant treatments for patients with ACC. This study aims to confirm the prog- nostic value of Ki-67 and provide a prognostic evaluation on ACC after primary surgery.
Methods: A total of 66 patients satisfied the inclusion criteria and their complete data were collected and reviewed. The correlation between Ki-67 index and clinicopathologic variables was analyzed using chi-square tests and Pearson’s or Spearman’s test. Survival curves were generated by Kaplan-Meier analysis and compared with the log-rank test. The Cox regression model was performed to estimate
Yuchun Zhu and Yiping Lu are co-corresponding authors of this article.
Enhanced Digital Features To view enhanced digital features for this article go to https://doi.org/10.6084/ m9.figshare.9198170.
Fuxun Zhang · Fan Zhang · Z. Liu · K. Wu . Y. Zhu ☒ Y. Lu ☒
Department of Urology, Institute of Urology, West China Hospital, Sichuan University, Chengdu, Sichuan, China e-mail: foolsoon.c@163.com
Y. Lu
e-mail: yipinglu@163.com
hazard ratios for univariate and multivariate analyses.
Results: Of the 66 patients, recurrence was observed in 30 patients (45.5%) and 26 patients (39.4%) died of progressive ACC. The evaluated median overall survival (OS) of the entire study population was 16.5 (range 1-104) months and recurrence-free survival (RFS) was 9.0 (range 0-104) months. Increased Ki-67 expression (> 20% and > 3%) was negatively correlated with OS and RFS (chi-square, P = 0.006 and 0.044, respectively). In multivariate analysis, the Ki-67 index with 20% and 3% cutoff as an independent prognostic factor for OS and RFS was validated [hazard ratio (HR) 3.289; 95% CI 1.345-8.042; P = 0.009 and HR 4.471; 95% CI 1.086-18.410; P = 0.038, respectively].
Conclusions: Ki-67 is a reliable, convenient, and independent prognostic marker for ACC. Additionally, as an indicator with a divergent prognostic role at different cutoff values (20% and 3%), Ki-67 could be used for stratifying patients with a high risk of death or rapid recurrence.
Keywords: Adrenocortical carcinoma; Cutoff value; Ki-67; Marker; Prognosis
INTRODUCTION
Adrenocortical carcinoma (ACC) is a rare malignancy with an estimated incidence of
1.0-2.0 cases per 1 million population per year and has one of the poorest prognoses among all solid tumors [1, 2]. Although complete resec- tion is the mainstay in ACC therapy and most patients with localized tumor are potentially curable, disease relapse is a common occurrence and devastatingly influences the life expectancy [3]. Mitotane, an effective drug for patients with advanced ACC, is used in an adjuvant setting to reduce recurrence on the basis of retrospective evidence. However, the efficacy of mitotane has never been assessed in a randomized trial and it has many side effects including the probable disturbance of metabolism of hormones and drugs [4, 5]. Furthermore, despite the fact that some reports have shown the relationship be- tween radiotherapy and prevention of local recurrence [6], many studies indicate the inef- fectiveness of radiotherapy and consider ACC as insensitive to radical therapy [7, 8]. Thus, the current treatment model with uncertain effi- cacy, restriction to toxic drugs, and disease recrudesce still remains a challenge for researchers worldwide. Therefore, it would be of great importance to predict prognosis and determinate high-risk patients by utilizing postoperative molecular markers from routine diagnostic workup. Meanwhile, necessary restriction on adjuvant approaches with potential adverse effects should be considered in ACC patients with insensitivity to therapy or low risk of rapid relapse.
Although histopathological scoring systems have been used for a long time to discriminate benign and malignant adrenal tumors, their associations with predictive value have not been fully investigated and established [9, 10]. Recent studies inspected the diagnostic and prognostic value of biomarkers for ACC and suggested that not all of them were informative and able to predict unwanted outcome. Mean- while, even for reportedly extensively resear- ched markers applied in prognostic tests, they are still far from being used in routine practice because of their expense and inaccessibility [10, 11].
Ki-67, a proliferation protein and pivotal component of the mitotic chromosome periph- ery, is essential to enable independent chromo- some motility and efficient interactions with the
mitotic spindle after nuclear envelope disassem- bly [12]. As an antigen highly expressed in cycling cells but significantly downregulated in resting cells, Ki-67 was identified as an important and well-established proliferation marker for tumor grade and prognosis in many other malignances in previous studies [13]. Thus, increasing attention has been paid to prognostic research on Ki-67 in various tumors owing to its availability and long-standing clinical utility.
As a consequence, this study aimed to con- firm convenient and affordable prognostic biomarkers and additionally provide prognostic evaluation for patients with ACC after primary resection who might be at high risk of rapid recurrence or death. In this setting, we exam- ined detailed clinical and histopathological data on 66 ACC patients to determine whether Ki-67 has prognostic value in patients with this infrequent malignancy.
METHODS
Patients
We reviewed the medical records of patients diagnosed with ACC between January 1, 2009 and December 31, 2017 at the West China Hospital, Sichuan University. The patients in our study had to meet the following inclusion crite- ria: pathologically confirmed diagnosis of ACC, available laboratory results of hematologic and biochemical function, no medical history of any other tumor, no adjuvant therapy with anti- neoplastic drugs prior to the first operation, available preoperative and postoperative imag- ing, and accessible data on tumor inva- sion and surgical margin status judged on the basis of surgical and pathological report (detailed clinical information included in this study is provided in Table 1). All morphological diagno- sis accepted re-evaluation according to the Weiss criteria and Ki-67 (MIB1) score was assessed by experienced pathologists from the same depart- ment in the West China Hospital. Staging, based on imaging assessed by independent reviewers, was confirmed by the findings at surgery using the American Joint Committee on Cancer (AJCC) 7th edition. Moreover, follow-up was not
| Feature | N = 66 |
|---|---|
| Age (years) [median (range)] | 44 (2-79) |
| Gender [n (%)] | |
| Male | 29 (43.9) |
| Female | 37 (56.1) |
| Side [n (%)] | |
| Right | 38 (57.6) |
| Left | 28 (42.4) |
| Comorbidities [n (%)] | |
| Yes | 16 (24.2) |
| No | 50 (75.8) |
| Function [n (%)] | |
| Yes | 28 (42.4) |
| No | 38 (57.6) |
| Albumin (mean ± SD) (g/dL) | 40.6 ± 5.1 |
| Globulin (mean ± SD) (g/dL) | 21.4 ± 5.2 |
| Diameter of tumor (cm) [median (range)] | 8.1 (1-20) |
| Tumor stage [n (%)] | |
| I | 8 (12.1) |
| II | 35 (53.0) |
| III | 19 (28.8) |
| IV | 4 (6.1) |
| Modus operandi [n (%)] | |
| Open | 46 (69.7) |
| Laparoscopy | 20 (30.3) |
| Post-relapse adjuvant treatment [n (%)] | |
| Treated | 14 (21.2) |
| Untreated | 52 (78.8) |
| Recurrence [n (%)] | 30 (45.5) |
| Death [n (%)] | 26 (39.4) |
SD standard deviation
suspended until death occurred or to the end of December 2017.
Study Protocol and Statistical Analysis
This study was clinician-initiated, retrospective, and conducted at one of the professional med- ical centers for the treatment of adrenal tumor in China. Patients’ data were collected at the West China Hospital. No commercial entity was involved in this research. The primary endpoint was overall survival (OS) defined as the interval from the date of first surgery to the date of last follow-up or death caused by any reason, and the secondary was recurrence-free survival (RFS) calculated as the time from primary operation to diagnosis of recurrence for relapsed patients or to last follow-up for recurrence-free patients. Besides, recurrence was defined as the radio- logical detection of a neoplastic lesion follow- ing primary resection during the follow-up, which was adjudicated by independent reviewers.
Database management and all statistical analyses were performed using the SPSS 23.0 statistical analysis software (IBM, New York, USA). Mean value and standard deviation or median and ranges were computed for contin- uous variables. Rates or proportions were com- puted for categorical variables. The correlation between Ki-67 index and clinicopathologic variables was analyzed using chi-square tests for categorical variables and Pearson’s or Spear- man’s test for continuous variables. There has been little consensus in selection of Ki-67 cutoff points, and the predictions regarding OS and RFS in this study have been based on a wide range of Ki-67 score from 0% to 90%. In light of this, Ki-67 was modelled as a binary variable and exploratively grouped by cutoff 1-5% with equidistant increases of 5% or 10%. Survival curves were generated by Kaplan-Meier analysis and compared with the log-rank test. The Cox regression model was used to estimate hazard ratios for univariate and multivariate analyses. Levels of statistical significance were set at P < 0.05. All P values were two-sided and 95% CIs and P values were based on X2 statistics with one degree of freedom.
78 Eligible patients
12 excluded 6 Lost to follow-up or unknown status 2 Incomplete clinicopathological data 2 Unknown ki67 score
Eligibility
2 Incomplete imaging data
66 patients included in this study
30 Recurrence 36 Non-recurrence
Grouping by Ki67 score with different cutoff of 3%, 5%, 10%, 20%, 30% respectively.
Method of diagnosis
40 survival 26 Death
Ki67 score with cutoff of 3%
Ki67 score with cutoff of 20%
Ki67≤3%
Ki67>3%
Ki67≤20%
Ki67>20%
Analysis
Non-recurrence 10 Recurrence 2
Non-recurrence 26 Recurrence 28
Survival 36 Death 18
Survival 4 Death 8
Compliance with Ethics Guidelines
This article is based on previously collected data and does not contain any new studies with human participants or animals performed by any of the authors. The institutional review board of West China Hospital had approved this study and all patients included in this study provided informed consent for the use and publication of their data.
RESULTS
Patients and Clinical Prognostic Factors
A total of 78 patients were screened for this study. After the exclusion of 12 patients who were lost to follow-up, provided incomplete medical records, or provided incomplete pathological data or imaging assessment, 66 patients satisfied the inclusion criteria and their
medical data were collected and reviewed (Fig. 1). During the thorough follow-up of all the patients, relapse was observed in 30 patients (45.5%) while 26 patients (39.4%) died of pro- gressive ACC. The evaluated median OS of the entire study population was 16.5 (range 1-104) months and RFS was 9.0 (range 0-104) months (characteristics of these patients including recurrence and death are provided in Table 1).
Outcome-affecting clinical characteristics were preoperative albumin, modus operandi, diameter of tumor, and Ki-67 index analyzed as a binary variable. However, only the Ki-67 index with 20% cutoff as an independent prognostic factor for OS and 3% cutoff for RFS had been confirmed utilizing multivariate the Cox model (P = 0.009 and 0.038, respectively). Interest- ingly, Ki-67 index was notably related to OS with the cutoff of 20% (chi-square, P = 0.006) but irrelevant to it with the cutoff of 3% (chi- square, P = 0.254). Similarly, significant corre- lation was observed between Ki-67 index with
| Variable | Ki-67 index | Chi-square | P value | |
|---|---|---|---|---|
| ≤ 20% | > 20% | |||
| Total | 54 (81.8) | 12 (18.2) | ||
| Age (years) [n (%)] | ||||
| ≤ 45 | 28 (42.4) | 8 (12.1) | 0.869 | 0.351 |
| > 45 | 26 (39.4) | 4 (6.1) | ||
| Gender [n (%)] | ||||
| Male | 23 (34.8) | 6 (9.1) | 0.219 | 0.640 |
| Female | 31 (47.0) | 6 (9.1) | ||
| Side [n (%)] | ||||
| Right | 30 (45.5) | 8 (12.1) | 0.496 | 0.481 |
| Left | 24 (36.4) | 4 (6.1) | ||
| Comorbidities [n (%)] | ||||
| Yes | 15 (22.7) | 1 (1.5) | 2.021 | 0.155 |
| No | 39 (59.1) | 11 (16.7) | ||
| Function [n (%)] | ||||
| Yes | 24 (36.4) | 4 (6.1) | 0.496 | 0.481 |
| No | 30 (45.5) | 8 (12.1) | ||
| Tumor stage [n (%)] | ||||
| I/II | 38 (57.6) | 5 (7.6) | 3.563 | 0.059 |
| III/IV | 16 (24.2) | 7 (10.6) | ||
| Modus operandi [n (%)] | ||||
| Open | 36 (54.5) | 10 (15.2) | 1.291 | 0.256 |
| Laparoscopy | 18 (27.3) | 2 (3.0) | ||
| Post-relapse adjuvant treatment [n (%)] | ||||
| Treated | 12 (18.2) | 2 (3.0) | 0.181 | 0.670 |
| Untreated | 42 (63.6) | 10 (15.2) | ||
| Recurrence [n (%)] | ||||
| Yes | 26 (39.4) | 4 (6.1) | 0.869 | 0.351 |
| No | 28 (42.4) | 8 (12.1) | ||
| Survival [n (%)] | ||||
| Yes | 36 (54.5) | 4 (6.1) | 4.569 | 0.033 |
| No | 18 (27.3) | 8 (12.1) | ||
| Albumin | 0.771 | |||
| Variable | Ki-67 index | Chi-square | P value | |
|---|---|---|---|---|
| ≤ 20% | > 20% | |||
| Globulin | 0.884 | |||
| Diameter of tumor | 0.474 | |||
Bold figures indicate statistical significance at P < 0.05
the cutoff of 3% and RFS (chi-square, P = 0.044), but irrelevance between Ki-67 index with the cutoff of 20% and RFS (chi-square, P = 0.985) (the origin of the material used for analysis is outlined in Table 3).
Prognostic Role of Ki-67 Index with Diverse Cutoffs
Increased Ki-67 expression (>20% and > 3%) was negatively correlated with OS and RFS in patients with ACC (chi-square, P = 0.006 and 0.044, respectively). No prognostic effect of age, gender, side, globulin, secretion, stage, post-re- lapse adjuvant treatment, and comorbidity including diabetes and hypertension was observed in univariate analysis (Table 3). A total of 54 cases were in the low Ki-67 group (≤ 20%) with 18 deaths (33.3%) and 12 cases in the high Ki-67 group (>20%) with 8 deaths (66.7%) (Table 2), while the estimated median OS in two
groups was 18 months (range 1-104 months) and 14 months (range 2-73 months) respec- tively. Kaplan-Meier survival curves for Ki-67 in two groups (≤ 20% vs > 20%) showed a remarkable difference (P = 0.006, Fig. 2a).
On the other hand, two of 12 patients with Ki-67 score ≤ 3% suffered recurrence and the median RFS for this group was 21.5 months (range 1-104 months); 28 of 54 patients with Ki-67 score >3% relapsed and median RFS was 8 months (range 0-76 months). Statistical cor- relation was found between the Ki-67 index with a 3% cutoff and disease recurrence in univariate and multivariate analyses using the Cox regression model (P = 0.044 and 0.038, respectively) (Tables 3, 4; Fig. 2b). We also assessed Ki-67 score as a continuous variable and categorical variable with diverse cutoff points (2%, 4%, 10%, 15%, 25%, and 30%); however, no significant difference was observed in corresponding groups (Table 3).
A
Overall survival
B
Recurrence-free survival
Survival from primary surgery (%)
Non-recurrence from primary surgury(%)
100
- Ki67 > 20
100
-Ki67d ≤ 3
Ki67d ≤ 20
Ki67>3
50
50
HR 3.32(95%CI 1.402-7.861)
HR 4.263(95%CI 1.037-17.531)
0
P=0.006
0
P=0.044
0
50
100
150
0
50
100
150
Time since entry (months)
Time since entry (months)
| Variable | Univariate analysis for OS | Univariate analysis for RFS | ||
|---|---|---|---|---|
| HR (95%CI) 1 (ref) | P value | HR (95%CI) 1 (ref) | P value | |
| Age | ||||
| ≤ 45 | 0.917 (0.420-2.002) | 0.828 | 0.781 (0.375-1.624) | 0.508 |
| > 45 | ||||
| Gender | ||||
| Male | 1.177 (0.539-2.573) | 0.682 | 1.661 (0.788-3.500) | 0.182 |
| Female | ||||
| Side | ||||
| Right | 0.740 (0.332-1.648) | 0.461 | 0.971 (0.461-2.045) | 0.938 |
| Left | ||||
| Comorbidities | ||||
| Yes | 0.918 (0.366-2.301) | 0.855 | 1.296 (0.572-2.938) | 0.534 |
| No | ||||
| Function | ||||
| Yes | 0.979 (0.449-2.134) | 0.958 | 0.808 (0.384-1.701) | 0.575 |
| No | ||||
| Tumor stage | ||||
| I/II | 1.609 (0.733-3.533) | 0.236 | 1.539 (0.724-3.268) | 0.262 |
| III/IV | ||||
| Diameter of tumor | 1.108 (1.007-1.219) | 0.035 | 0.973 (0.882-1.074) | 0.594 |
| Modus operandi | ||||
| Open | 0.337 (0.116-0.979) | 0.046 | 1.121 (0.527-2.383) | 0.767 |
| Laparoscopy | ||||
| Postoperative adjuvant treatment | ||||
| Treated | 0.839 (0.335-2.098) | 0.707 | – | – |
| Untreated | ||||
| Recurrence | ||||
| Yes | 0.921 (0.424-2.000) | 0.835 | – | – |
| No | ||||
| Ki-67 index | ||||
| 3% cutoff | 1.652 (0.698-3.911) | 0.254 | 4.263 (1.037-17.531) | 0.044 |
| 5% cutoff | 0.765 (0.326-1.793) | 0.538 | 2.979 (0.899-9.868) | 0.074* |
| Variable | Univariate analysis for OS | Univariate analysis for RFS | ||
|---|---|---|---|---|
| HR (95%CI)1 (ref) | P value | HR (95%CI)1 (ref) | P value | |
| 10% cutoff | 1.980 (0.879-4.460) | 0.099* | 1.364 (0.631-2.947) | 0.430 |
| 20% cutoff | 3.320 (1.402-7.861) | 0.006 | 0.990 (0.342-2.863) | 0.985 |
| 30% cutoff | 2.162 (0.803-5.821) | 0.127 | 0.903 (0.272-2.997) | 0.868 |
| Albumin | 0.953 (0.88-1.031) | 0.228 | 0.940 (0.875-1.011) | 0.094* |
| Globulin | 0.994 (0.928-1.066) | 0.874 | 0.970 (0.908-1.037) | 0.378 |
Bold figures indicate statistical significance at P < 0.05, *could be treated as statistically significant adjusted significant level (x = 0.10) OS overall survival, RFS recurrence-free survival, HR hazard ratio, 1 (ref) 1 degree of freedom, CI confidence interval
| Parameter | HR [1 (ref)] | 95% CI | P value |
|---|---|---|---|
| Analysis for OS | |||
| Diameter of | 1.056 | 0.949-1.174 | 0.319 |
| tumor | |||
| Modus operandi | 0.403 | 0.126-1.297 | 0.128 |
| 20%-cutoff | 3.289 | 1.345-8.042 | 0.009 |
| Ki-67 | |||
| Analysis for RFS | |||
| Albumin | 0.930 | 0.861-1.005 | 0.065* |
| Ki-67 with 3% | 4.471 | 1.086-18.410 | 0.038 |
| cutoff | |||
| Ki-67 with 5% cutoff | 2.993 | 0.903-9.923 | 0.073* |
Bold figures indicate statistical significance at P < 0.05, *could be treated as statistically significant at adjusted sig- nificant level (x = 0.10)
OS overall survival, RFS recurrence-free survival, HR haz ard ratio, 1 (ref) 1 degree of freedom, CI confidence interval
In Cox multivariate analysis, the dichoto- mous Ki-67 index with the cutoff of 20% and 3% as independent prognostic factors for OS and RFS was validated (HR 3.289; 95% CI 1.345-8.042; P = 0.009 and HR 4.471; 95% CI 1.086-18.410; P = 0.038) (Fig. 3). The covariate extent of disease was excluded from the final
model on account of inadequacy of statisti- cal power and predictive value (Table 4; Fig. 3).
DISCUSSION
Treatment and prognosis for ACC have not significantly improved in recent years. In this disappointing scenario, the potential prognos- tic role of molecular biomarkers should be fur- ther investigated and immunotherapy should be considered one of the most promising approaches in the future [14]. At the same time, although great progress has been made in exploration on altered genes and emerging protein markers for ACC over the last decade, recent guidelines still emphasize the combina- tion of clinical parameters and molecular markers to ameliorate the prognosis of this rare malignancy [15]. However, easy and cost-effec- tive biomarkers related to prognosis have been recognized as a rarity [15, 16].
Ki-67, the most promising immunohisto- chemical proliferation marker identified so far, has been shown to play a prognostic role in several malignancies with various tissue derivation [17]. In this setting, higher Ki-67 expression was considered to be associated with unfavorable clinical outcomes in many cancers [18]. Moreover, recent studies compared Ki-67 with other predictors at prognostic value and regarded Ki-67 as a more powerful biomarker carrying significantly prognostic information [18, 19]. High expression of Ki-67 also has a
prognostic indicators
Hazard ratio (95%CI)
Modus operation
0.403 (0.126-1.297)
Diameter of tumor
1.056 (0.949-1.174)
Albumin
0.930 (0.861-1.005)
20%cutoff Ki67
3.289 (1.345-8.042)
5%cutoff Ki67
2.993 (0.903-9.923)
3%cutoff Ki67
4.471 (1.086-18.410)
-5.0
0.0
5.0
10.0
15.0
20.0
direct correlation with tumor size, grade, and increased mortality [20, 21].
As a result of the rarity of ACC, the associa- tion between molecular biomarkers and recur- rence or overall survival has not been defined adequately compared with other cancers in clinical practice. To our knowledge, few studies analyzed the prognostic role of Ki-67 in ACC and little data on clinical risk of tumor progres- sion were available in China. Assessment of Ki- 67 in previous studies has prompted specula- tion that Ki-67 should be considered the most important predictor in patients following com- plete resection [22]. However, the evi- dence in recent studies was flawed by several limitations from heterogeneous data sources, such as uniformly defined OS and RFS, diverse treatment measures in different patients, vari- ability of Ki-67 index evaluated by different pathologists in different medical institutions, and obvious variation in baseline characteristics on account of the large time span in patients included [22]. At the same time, varied cutoff values of Ki-67 in previous studies may be indicative of the deviation in threshold-value confirmation and histomorphological hetero- geneity of ACC. Therefore, an optimized cutoff
value of Ki-67 score applied in prognostic study on ACC should be determined as well.
In this study, we evaluated the clinical- pathological associations and the prognostic role of Ki-67 in a consecutive mono-institu- tional series of ACCs and found that high Ki-67 expression was associated with worse outcomes. In the current study, Ki-67 score was analyzed as a dichotomous variable for its prognostic role at diverse cutoffs and served as a prognostic dis- criminator to distinguish patients with different outcomes. Our work showed that ACC patients with Ki-67 expression less than or equal to 20% had better OS compared those who maintained Ki-67 expression greater than 20%; analogously, patients whose tumors had Ki-67 expression less than or equal to 3% had better RFS and delayed recurrence. In this sense, we confirmed that the Ki-67 index with varied cutoffs has different prognostic value for mortality and rapid recur- rence (Fig. 4). Besides, in univariate analysis for RFS, the association between variables and relapse was treated as statistically significant at an adjusted significant level (x = 0.10) because of the rarity of ACC and insufficient sample size (Table 3). Although out results were not as strong as those in prior reports, we believe that
Ki67 > 20
Death
Ki67d ≤ 20
Survival
Recurrence
Ki67>3
Ki67d ≤ 3
Non-recurrence
0
10
20
30
40
Cases (n)
Ki-67 score with the 3% cutoff could be an independent predictor of recurrence and patients with Ki-67 expression greater than 3% may relapse more rapidly (Fig. 2b).
In addition to the Ki-67 index, the prognosis of ACC had proven to be driven partially by other factors, including the presence of age, functionality, tumor stage, metastasis, invasion of adjacent structures, and resection status [23]. Age was defined as a dichotomous variable according to the median in this study which may be optimal. Although some studies have indicated that cortisol-secreting ACCs might be associated with unwanted prognosis [24, 25], we did not find that functional status was a signif- icant prognostic factor. Thus, the functionality of ACC should not preclude surgery and adju- vant treatment. At this point, further investi- gations should be conducted on the correlation between the magnitude and subtypes of secre- tions and the clinical outcome. In recent stud- ies, the tumor stage was recognized as the basis of prognostic stratification of ACC, underscoring the importance of the T-staging system [26]. However, the T-stage classification system for ACC is still being revised and the
best approach to evaluate the prognosis for ACC continues to fuel debate [23, 27]. Furthermore, although some prognostic predictors such as surgical margin status, invasion of adjacent structures, and distant metastasis could be considered as indicators for tumor progression, these predictors seem to be more intuitive and subjective for experience-varied surgeons and also lead to uncertainty. Thus, it is not easy to assess resection status objectively and define the optimal time to perform R0 resection accurately [23, 28].
In the present study, there was no significant correlation between Ki-67 index and age, gen- der, side, comorbidity, adjuvant treatment, functionality, and tumor stage (Table 2). High Ki-67 index (> 20%) was only correlated with overall survival (Table 2). Our study also dis- covered several factors that might portend worse prognosis, such as diameter of tumor, modus operandi, and preoperative albumin. However, no independent prognostic marker among them was validated in multivariate analysis (Table 4).
A recent study described Ki-67 as an uncer- tain predictive marker and indicated that the
deficiency of uniform quantification and pro- liferative heterogeneity of ACC might place a substantial limitation on its utility [29]. However, in spite of its small simple size, our study suggested that the Ki-67 index could be deemed as a prognostic parameter in accor- dance with recent data [22, 23]. Meanwhile, mono-institutional assessment of Ki-67 may minimize the adverse impact of the confound- ing factors (also known as interobserver or interlaboratory variation) such as tissue prepa- ration, interpretation, scoring, and data analysis which varied at different laboratories and clin- ical centers.
Discrimination between ACC and benign adrenocortical tumors is difficult and can be accomplished using multiple histological scores, such as the Weiss system or Weiss revisited index. Our research was limited to evaluate the associ- ation between the Weiss score, Ki-67 expression, and the clinical outcome. However, the Weiss score had limited prognostic potential and failed to predict all outcomes [10]. This uncertainty is probably ascribed to interobserver or interlabo- ratory variation mentioned above and stan- dardization of applying certain scoring systems requires further studies.
The limitations of this study include the retrospective nature of its design, 12 patients who provided incomplete data were not inclu- ded, and the small sample size with inadequate statistical power and inherent biases. On the other hand, the result of our study might be influenced by those patients that died from other causes. Additionally, the recently pub- lished experience shows that adjuvant treat- ments including radiation therapy and mitotane are associated with improvement of clinical outcome [30, 31]. In the present study, however, only the impact of post-relapse adju- vant treatment was incorporated into analysis because of the different perspectives on adju- vant strategies, the paucity of accurate data on treated cases, and extremely low propor- tion of patients who accepted adjuvant treat- ment after primary resection. All of these factors may influence the result and confidence. Therefore, additional cases should be accumu- lated to elucidate these associations and sup- plementary investigations are mandatory.
Though our study is retrospective and requires further validation, the rarity of adrenal cancer remains the biggest challenge to con- ducting a prospective study. As mentioned above, several studies had aggregated and pro- cessed a massive amount of information about ACC patients who underwent treatment at various clinical centers, and it seems that Ki-67 is well researched on the basis of the large number of patients. However, it is crucial to highlight that both diversity of demographic characteristics and heterogeneity of merging data from different medical institutions should be taken into account judiciously. After all, statistical robustness depends not only on the number of included patients but also homogeneity of data. Furthermore, though Ki- 67 has been considered the most cogent pre- dictor for ACC thus far [22], it is necessary to perform confirmatory research and continu- ously collect primary data for clinical and prognostic utilization.
CONCLUSIONS
Ki-67 is a reliable, convenient, and independent prognostic marker for patients with ACC. In this study, Ki-67 score carried a divergent prognostic role at different cutoffs and should be drawn into treatment flow charts. Additionally, Ki-67 score with different cutoffs (20% and 3%) could serve as a potential discriminator in prognosti- cation models for stratifying patients with high risk of death or rapid recurrence who should be provided appropriate treatment options and prudent follow-up evaluation. Besides, future efforts should focus not merely on searching for a probable malignancy marker but also original investigations and confirmatory studies suit- able for meta-analysis, aiming at generating more convincing evidence to improve the therapeutic strategies and prognosis in patients with this rare disease.
ACKNOWLEDGEMENTS
We thank the participants of the study in addition to Chuan Zhou and Jiayu Liang for
their contributions to the collection of data and revision of the manuscript.
Funding. This study and its publishing expenses including Rapid Service fee were fun- ded by the 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University and the Science and Technology Foundation of the Sichuan Province (2017SZ0123 to Zhihong Liu).
Authorship. All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published. Fan Zhang and Zhihong Liu contributed equally to this study.
Disclosures. All named authors, Fuxun Zhang, Fan Zhang, Zhihong Liu, Kan Wu, Yuchun Zhu, and Yiping Lu, have nothing to disclose.
Compliance with Ethics Guidelines. This article is based on previously collected data and does not contain any new studies with human participants or animals performed by any of the authors. The institutional review board of West China Hospital had approved this study and all patients included in this study provided informed consent for the use and publication of their data.
Data Availability. The datasets analyzed during the current study are available from the corresponding author on reasonable request.
REFERENCES
1. Fassnacht M, Terzolo M, Allolio B, et al. Combina- tion chemotherapy in advanced adrenocortical carcinoma. N Engl J Med. 2012;366(23):2189-97.
2. Sharma E, Dahal S, Sharma P, et al. The character- istics and trends in adrenocortical carcinoma: a United States population based study. J Clin Med Res. 2018;10(8):636-40.
3. Kim Y, Margonis GA, Prescott JD, et al. Curative surgical resection of adrenocortical carcinoma: determining long-term outcome based on condi- tional disease-free probability. Ann Surg. 2017; 265:197-204.
4. Kroiss M, Quinkler M, Lutz WK, et al. Drug inter- actions with mitotane by induction of CYP3A4 metabolism in the clinical management of adrenocortical carcinoma. Clin Endocrinol (Oxf). 2011;75:585-91.
5. Postlewait LM, Ethun CG, Tran TB, et al. Outcomes of adjuvant mitotane after resection of adrenocor- tical carcinoma: a 13-institution study by the US Adrenocortical Carcinoma Group. J Am Coll Surg. 2016;222:480-90.
6. Sabolch A, Feng M, Griffith K, et al. Adjuvant and definitive radiotherapy for adrenocortical carci- noma. Int J Radiat Oncol Biol Phys. 2011;80: 1477-84.
7. Habra MA, Ejaz S, Feng L, et al. A retrospective cohort analysis of the efficacy of adjuvant radio- therapy after primary surgical resection in patients with adrenocortical carcinoma. J Clin Endocrinol Metab. 2013;98:192-7.
8. Atallah S, Al-Assaf H, Xu Y, et al. Adrenocortical carcinoma: patterns of care and role of adjuvant radiation therapy-a population-based study and review of the literature. Curr Oncol. 2017;24: e316-22.
9. van’t Sant HP, Bouvy ND, Kazemier G, et al. The prognostic value of two different histopathological scoring systems for adrenocortical carcinomas. Histopathology. 2007;51:239-45.
10. Mete O, Gucer H, Kefeli M, et al. Diagnostic and prognostic biomarkers of adrenalcortical carci- noma. Am J Surg Pathol. 2017;48:201-13.
11. Zheng S, Cherniack AD, Dewal N, et al. Compre- hensive pan-genomic characterization of adreno- cortical carcinoma. Cancer Cell. 2016;30:363.
12. Cuylen S, Blaukopf C, Politi AZ, et al. Ki-67 acts as a biological surfactant to disperse mitotic chromo- somes. Nature. 2016;535(7611):308-12.
13. Sun X, Kaufman PD. Ki-67: more than a prolifera- tion marker. Chromosoma. 2018;127(2):175-86.
14. Varghese J, Habra MA. Update on adrenocortical carcinoma management and future directions. Curr Opin Endocrinol Diabetes Obes. 2017;24:208-14.
15. Fassnacht M, Dekkers OM, Else T, et al. European Society of Endocrinology Clinical Practice Guideli- nes on the management of adrenocortical
carcinoma in adults, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol. 2018;179(4):G1-46.
16. Poli G, Ruggiero C, Cantini G, et al (2018) Fascin-1 is a novel prognostic biomarker associated with tumor invasiveness in adrenocortical carcinoma. J Clin Endocrinol Metab. 2019;104(5):1712-24.
17. Kreipe H. Ki67: biological intertumor variance ver- sus variance of assay. Pathologe. 2018;39(Suppl 2):271-7.
18. Fisher G, Yang ZH, Kudahetti S, et al. Prognostic value of Ki-67 for prostate cancer death in a con- servatively managed cohort. Br J Cancer. 2013;108(2):271-7.
19. Goltz D, Montani M, Braun M, et al. Prognostic relevance of proliferation markers (Ki-67, PHH3) within the cross-relation of ERG translocation and androgen receptor expression in prostate cancer. Pathology. 2015;47(7):629-36.
20. Yoshioka T, Hosoda M, Yamamoto M, et al. Prog- nostic significance of pathologic complete response and Ki67 expression after neoadjuvant chemother- apy in breast cancer. Breast Cancer. 2015;22(2): 185-91.
21. Constantinou C, Papadopoulos S, Karyda E, et al. Expression and clinical significance of Claudin-7, PDL-1, PTEN, c-Kit, c-Met, c-Myc, ALK, CK5/6, CK17, p53, EGFR, Ki67, p63 in triple-negative breast cancer-a single centre prospective observational study. In Vivo. 2018;32(2):303-11.
22. Beuschlein F, Weigel J, Saeger W, et al. Major prognostic role of Ki67 in localized adrenocortical carcinoma after complete resection. J Clin Endo- crinol Metab. 2015;100:841-9.
23. 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(10):2119-25.
24. Abiven G, Coste J, Groussin L, et al. Clinical and biological features in the prognosis of adrenocorti- cal cancer: poor outcome of cortisol-secreting tumors in a series of 202 consecutive patients. J Clin Endocrinol Metab. 2006;91:2650-5.
25. Berruti A, Fassnacht M, Haak H, et al. Prognostic role of overt hypercortisolism in completely oper- ated patients with adrenocortical cancer. Eur Urol. 2014;65:832-8.
26. Poorman CE, Ethun CG, Postlewait LM, et al. A novel T-stage classification system for adrenocorti- cal carcinoma: proposal from the US Adrenocortical Carcinoma Study Group. Ann Surg Oncol. 2018;25(2):520-7.
27. Lughezzani G, Sun M, Perrotte P, et al. The Euro- pean Network for the Study of Adrenal Tumors staging system is prognostically superior to the International Union Against Cancer-staging sys- tem: a North American validation. Eur J Cancer. 2010;46(4):713-9.
28. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocor- tical carcinoma in the United States: treatment utilization and prognostic factors. Cancer. 2008;113(11):3130-6.
29. Papathomas TG, Pucci E, Giordano TJ, et al. An international Ki67 reproducibility study in adrenal cortical carcinoma. Am J Surg Pathol. 2016; 40(4):569-76.
30. Sabolch A, Else T, Griffith KA, et al. Adjuvant radi- ation therapy improves local control after surgical resection in patients with localized adrenocortical carcinoma. Int J Radiat Oncol Biol Phys. 2015;92:252-9.
31. Berruti A, Grisanti S, Pulzer A, et al. Long-term outcomes of adjuvant mitotane therapy in patients with radically resected adrenocortical carcinoma. J Clin Endocrinol Metab. 2017;102(4):1358-65.