Abstract
Background/Aim: The Aspartate aminotransaminase/Alanine aminotransaminase ratio (AST/ALT ratio) has been identified as a prognostic marker for several malignancies. In this study, we evaluated the prognostic value of the AST/ALT ratio in a large cohort of non-metastatic colorectal cancer patients (CRC). Patients and Methods: A total of 536 patients with stage II and III CRC, as well as available AST/ALT ratio were included in this single-center retrospective analysis. Laboratory data were measured within two weeks before histological tumor diagnosis. Co-Primary endpoints for this analysis were disease-free survival (DFS) and overall survival (OS). Results: In univariate cox regression DFS was significantly shorter in patients with an elevated AST/ALT ratio (HR=1.568, 95%CI=1.10-2.23, p=0.012). In multivariable analysis, the prognostic association between an elevated AST/ALT ratio and a poor survival prevailed statistically significant (HR=1.53, 95%C=1.05-2.22, p=0.026). No statistically significant association between the AST/ALT ratio and OS was observed (HR=1.4, 95% CI=0.89-2.22, p=0.14). Conclusion: In this study, the serum AST/ALT ratio emerged as a valid prognostic marker for DFS in non-metastatic colorectal cancer patients at stage II and III.
Colorectal cancer (CRC) is the third most common form of cancer and the second leading cause of cancer-related death worldwide (1). Since the early 1980s, the mortality rate has constantly declined, mainly due to extensive screening methods and improved treatment options (2).
Surgical resection followed by adjuvant chemotherapy is considered the standard therapy for CRC UICC Stage III (3-5). In stage II patients, adjuvant chemotherapy is case-dependent, with an ongoing debate on whether benefits outweigh possible severe side-effects. Despite these treatment options, approximately 30% of all patients with UICC stage II or III CRC undergoing treatment with curative intention develop recurrent disease (6). Identifying patients at high risk of disease recurrence and adapting the treatment and follow-up regime accordingly can reduce the risk of recurrence. However, to date, only limited clinical laboratory or histopathological factors exists to predict the risk of recurrent disease in CRC patients.
It is, therefore, crucial to explore novel biomarkers that help in the identification of patients at high risk of recurrent disease, and some molecular approaches including genetic alterations, circulating tumor DNA or non-coding RNAs are promising candidates (7-9). Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are commonly used biomarkers for hepatocellular damage (10). Both enzymes are produced by malignant and non-malignant cells and are expressed in different types of cells in the body. The ratio of plasma activities of AST and ALT, the so called De Ritis ratio, has initially been described as a diagnostic marker for acute viral hepatitis. Recent studies, however, have shown that the AST/ALT ratio may be used as a prognostic tool for certain kind of malignancies (11-14). To the best of our knowledge there has been no study reporting the correlation between the AST/ALT ratio and the DFS in patients with localized CRC. Thus, we evaluated the prognostic effect of the AST/ALT ratio in a large cohort of CRC patients with UICC stage II or III.
Patients and Methods
Study design, patient cohort and clinical outcome. This retrospective single-center observational cohort study included data of patients with histologically confirmed stage II and III (UICC) CRC. The initial cohort consisted of a total of 1,018 CRC patients (UICC Stage I-VI) who were treated in our department (Division of Oncology, Department of Internal Medicine, Medical University of Graz, Austria) between March 2010 and January 2016. According to predefined criteria (Figure 1), we excluded 66 Stage I and 245 Stage IV patients. From a total of 695 stage II and III patients, AST and ALT values obtained within two weeks prior to histological diagnosis, were available from 536 patients and thus were ultimately included in this study. Baseline data was extracted from the electronic health record database of our hospital trust (includes all public hospitals in the province of Styria, Austria), our departments internal documentation system as well as paper-chart archives of our hospital. Plasma AST and ALT were measured in each patient by standard clinical testing methods in lithium heparin plasma (upper reference level AST and ALT 35 and 45 U/l, respectively). The co-primary endpoints for this analysis were DFS, defined as the time from curative surgery of the primary tumor to recurrence or death from any cause, whatever occurred first, and OS.
Flowchart of the study population. CRC: Colorectal cancer; AST: aspartate aminotransferase; ALT: alanine aminotransferase; AST/ALT: aspartate aminotransferase/alanine aminotransferase.
Statistical analysis. Medians [with interquartile ranges (IQR)] and means [with standard deviations (SD)] were reported for non-normally and normally-distributed variables, respectively. Deritis ratio was calculated by dividing ALT through AST. Youden index was used to estimate the optimal cut-off value for the Deritis ratio with regards to DFS. Uni- and multivariate Cox-regression analyses were used to assess risk factors for DFS and OS, respectively. Statistical analyses were carried out using Stata Version 15.1 (StataCorp, TX, USA). A p-value of <0.05 was considered statistically significant.
Ethics approval and consent to participate. The Institutional Review Board of the Medical University of Graz approved this study before any patient-related tasks were performed. For retrospective database studies that have been approved by an ethics committee, written informed consent is not necessary and was therefore not obtained from individual patients.
Results
Analysis at baseline. A total of 536 patients were included in the study, of which 334 were males (62%) and 202 were females (38%). The median age was 65 years (IQR=64-66). A total of 196 patients (37%) were assessed as UICC stage II and 340 (63%) as stage III. Regarding tumor location, 292 tumors (56%) were located in the colon, while 236 (44%) were tumors of the rectum. A total of 260 patients (49%) received adjuvant chemotherapy. The median AST/ALT ratio was 1.16 (IQR=0.92-1.53). Further baseline characteristics of our cohort are summarized in Table I.
Baseline characteristics of the study population (n=536).
In the comparison of the low and high AST/ALT ratio groups, defined by an empirical cut-off of 0.96 suggested by the Youden Index, there were statistically significant differences in sex, age, BMI, hemoglobin and the proportion of patients who received adjuvant chemotherapy. Apart from that, the distribution of all other investigated baseline parameters was highly similar between patients with and without an elevated AST/ALT ratio (Table I).
Cohort outcome. During a median follow-up of 31 months the median DFS and OS was not reached. Out of 536 patients a total of 102 (19%) developed tumor recurrence of which 86 patients developed distant metastases, 9 had a local recurrence and 7 patients had a local recurrence and distant metastases. One-, two- and 5-year DFS rates were 89.5%, 82.3% and 76.7%, respectively. Sixty-two patients (12%) died within the follow-up period, resulting in one-, two- and five-year OS rates of 98.5%, 95.9% and 81.6%.
Uni- and multivariate analysis of clinical outcomes regarding AST/ALT ratio. DFS was significantly shorter in patients with an elevated AST/ALT ratio. In univariate cox regression, one unit increase of the AST/ALT ratio was associated with a 1.56-fold relative increase in the risk of recurrence or death (HR=1.568, 95%CI=1.10-2.23, p=0.012). Other univariate predictors of DFS were tumor stage (worse DFS in stage 3 compared to stage 2) and tumor grading (worse DFS in patients with grade 3 disease). In our univariable analysis, adjuvant chemotherapy did not emerge to be a predictive marker for DFS. However, since the application of adjuvant chemotherapy is an established prognostic marker in clinical practice, we included it in our multivariate analysis. In the multivariate model including grade, stage and adjuvant chemotherapy, the prognostic association between an elevated AST/ALT ratio and a poor survival was statistically significant (HR=1.53, 95%CI=1.05-2.22, p=0.026; Table II).
Uni- and multivariate Cox proportional hazards regression models of disease-free survival (n=536).
By treating the AST/ALT ratio as binary variable with an empirical cut-off at 0.96 suggested by the Youden index, similar results were observed. In detail, the median 12-, 24- and 36-month DFS rates were 92%, 88%, 84% in the low and 88%, 79%, 76 % in the high AST/ALT group, respectively (HR=1.63, 95%CI=0.99-2.68, p=0.052; Figure 2).
Disease-free survival according to AST/ALT-ratio (n=507). DFS estimates were computed with Kaplan-Meier estimators, and the numbers below the x-axis represent a risk table.
In OS-analysis, there was a trend towards worse OS for elevated AST/ALT ratio; however, no statistically significant association was observed in univariate (HR=1.4, 95%CI= 0.89-2.22, p=0.14) and multivariate analysis (HR=1.24, 95%CI=0.78-1.96, p=0.35).
Sensitivity analysis. We further performed a sensitivity analysis investigating the association of an elevated AST/ALT ratio with DFS in our cohort stratified by adjuvant chemotherapy. We observed that in patients receiving adjuvant chemotherapy, an elevated AST/ALT ratio was associated with worse DFS (HR=1.78 per unit increase, 95%CI=1.10-2.86, p=0.017), whereas in patients not receiving adjuvant chemotherapy no statistically significant association between the AST/ALT ratio and DFS was found. (HR=1.48 per unit increase, 95%CI=0.83-2.64, p=0.177).
In a second sensitivity analysis, we investigated whether stage II and III colorectal cancer patients with missing AST/ALT ratio (n=159) differed from patients included in our primary analysis. Here, we observed that baseline characteristics between the in- and excluded patients were all grossly similar (Table III).
Distribution of baseline characteristics depending on whether the AST/ALT ratio was observed or missing (n=695).
Discussion
In the present study, we evaluated the prognostic value of the AST/ALT (De Ritis) ratio in a large cohort of stage II and III non-metastatic CRC patients. We discovered that an elevated AST/ALT ratio is significantly associated with a worse DFS. These findings indicate that the AST/ALT ratio might represent a novel, inexpensive biomarker for risk stratification in patients with localized CRC who underwent surgical resection.
AST and ALT are the most commonly used serum biomarkers for hepatocellular damage in clinical practice. While ALT is almost exclusively found in the liver, AST is additionally expressed in various other organs including the heart, skeletal muscle, kidneys, brain and erythrocytes. Thus, many pathological processes and diseases can lead to an elevated AST activity, while elevated ALT levels are mostly associated with liver diseases (15).
While the exact underlying pathophysiological processes linking an elevated AST/ALT ratio to a worse prognosis are yet to be understood, there are several possible explanations. Warburg discovered that cancer cells predominantly rely on glycolysis followed by lactic acid fermentation as their source of energy, even in the presence of abundant oxygen (16). The malate-aspartate-shuttle is a biochemical system that plays an important role in the aerobic glycolysis. It consists of four proteins, with AST being one of them. Tumor metabolism can, therefore, lead to an elevation of AST and affect the De Ritis ratio (17).
Furthermore, tumor progression is associated with high proliferation, cell turnover, tissue damage and necrosis. All these pathological processes can lead to an elevation of AST while ALT levels remain mostly unaffected (18).
Recent publications have shed new light on the well-known marker by linking an elevated AST/ALT ratio to a worse prognosis in multiple malignancies. While the majority of these studies included metastatic disease in their analysis, some specifically investigated the AST/ALT ratio in the localized setting.
Wang et al. found a correlation between an elevated AST/ALT ratio and a worse prognosis for localized prostate cancer patients in their retrospective cohort study including 438 patients. In their study, a high AST/ALT ratio was associated with a higher rate of biochemical recurrence and a predictor for a higher Gleason score. The 2-year biochemical recurrence-free survival (BCRFS) rates were 97.2% for the low ratio group (AST/ALT<1.32) and 93.7 % for the high ratio group (AST/ALT>1.32), while the 5-year BCRFS were 86.1% and 69.3%, respectively. As in our study, the maximum value of the Youden index was used to evaluate the optimum cut-off point between low and high AST/ALT ratio group (19).
In a study by Bezan et al. a high AST/ALT was associated with poor survival in non-metastatic RCC. A preoperative increase in AST/ALT ratio above 1.26 was an an independent prognostic factor for both metastasis-free survival (HR=1.61) and overall survival (HR=1.76) (11). Yun-Sok Ha et al. retrospectively investigated 118 non-metastatic bladder cancer patients. In that study, patients with an elevated AST/ALT ratio had inferior metastasis-free survival (HR=2.38), cancer-specific survival (HR=2.77) and OS (HR=2.76) outcomes (20). Nishikawa et al. investigated the AST/ALT ratio in localized upper urinary tract urothelial carcinoma patients following nephroureterectomy. In their retrospective study, an elevated AST/ALT ratio was shown to be significantly correlated with a worse extravesical recurrence-free survival in both uni- and multivariate analysis. In addition, the relationship between the AST/ALT ratio and the OS as well as the DSS of the included patients were analyzed, although no correlation was found (21).
Following a comparable approach to these studies, we investigated the AST/ALT ratio in stage II and III disease, where risk stratification and choice of treatment are especially crucial. Although we are the first to investigate the association between the AST/ALT ratio and the prognosis of CRC patients, Lindmark et al. has already reported an association of elevated AST levels to a worse disease-specific survival in CRC patients in the year 1994 (22). Interestingly, we found no association between elevated AST levels and worse survival outcome indicated by DFS and OS in our cohort. In our analysis, the De Ritis ratio emerged as an independent prognostic indicator for DFS. Furthermore, the association was stronger in patients receiving adjuvant chemotherapy, which might indicate a potential predictive value of this biomarker. In a prospective study by Riedl et al. the predictive properties of the AST/ALT ratio have been shown for patients with advanced pancreatic cancer receiving combination chemotherapy with gemcitabine and nab-paclitaxel. A treatment response analysis revealed a 19% lower absolute ORR in patients with an elevated AST/ALT ratio (23).
Finally, this study contains several limitations that should be discussed. First, the retrospective design of this study may have introduced a selection bias to our cohort. We tried to address this issue by excluding patients of UICC stage I, since the majority of this patient cohort are not routinely referred to our Cancer Center.
Secondly, in our study an elevated AST/ALT ratio was only associated with a worse DFS, but not with worse OS. However, the hazard ratio of 1.4 indicates a similar trend towards worse OS in patients with elevated AST/ALT ratio. The lack of statistical significance might be due to the low event rate caused by the relatively short follow up period of 31 months.
Furthermore, the absence of an independent validation cohort for our biomarker analysis proposes an additional limitation. We therefore encourage other study groups to validate our findings in comparable patient cohorts as a further step to clinical significance
Conclusion
In this study, the plasma AST/ALT ratio emerged as a valid prognostic marker for DFS in non-metastatic colorectal cancer patients at stage II and III. Within the limitations of a retrospective study we conclude that the serum AST/ALT ratio might represent a novel and inexpensive prognostic tool to aid in the identification of patients at high risk of recurrent disease.
Footnotes
Authors’ Contributions
Study conception design: Scheipner, Stöger, Gerger, Riedl. Data acquisition: Scheipner, Smolle, Barth, Posch, Riedl. Data analysis and interpretation: Scheipner, Smolle, Posch, Stotz, Pichler, Gerger, Riedl. Drafting the article: Scheipner, Riedl. Critical revision: Scheipner, Smolle, Barth, Posch, Stotz, Pichler, Stöger, Gerger, Riedl. Final approval: Scheipner, Smolle, Barth, Posch, Stotz, Pichler, Stöger, Gerger, Riedl.
Conflicts of Interest
The Authors declare that there are no conflicts of interest.
- Received October 31, 2020.
- Revision received November 16, 2020.
- Accepted November 17, 2020.
- Copyright© 2021, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
