Impact of increased aspartate aminotransferase to alanine aminotransferase (De Ritis) ratio in prognosis of testicular cancer

Testicular cancer (TCa) is responsible for 1% of solid organ neoplasms in men, and its incidence is increasing in Western countries [1, 2, 3, 4]. Bilateral TCa is seen in 1% to 2% of cases at the time of diagnosis, and most of these cases are germ cell tumors [1].

The process of diagnosing TCa is begun and continued by determination of the level of tumor markers before and after surgical treatment, respectively. Follow-up regimens are also configured on the basis of these markers [5, 6]. Alpha-fetoprotein (AFP) and human chorionic gonadotropin (hCG) are increased in almost 50% of patients with nonseminomatous germ cell tumors. Moreover, most nonseminomatous germ cell tumors present with an increase in one or both markers. Nearly one-third of seminomas can have increased hCG along the course of TCa [7]. Surgical treatment is the backbone of TCa treatment. Radical orchiectomy is one aspect of definitive treatment of TCa [8, 9].

Amino transaminases, including aspartate aminotransferase (AST) and alanine aminotransferase (ALT), are enzymes produced in liver cells and released into the circulatory system. Because they can indicate hepatocellular damage, these enzymes are routinely included in the commonly requested enzyme panel used to evaluate liver function [10]. The ratio of serum activities of AST to ALT has been known as the De Ritis ratio (DRR) since it was first described by Fernando De Ritis [11]. This ratio (AST/ALT) is also increased in liver diseases, specifically in viral hepatitis [12, 13]. Published studies point to changing levels of amino transaminases [14]. Therefore, the DRR may be a prognostic biomarker in patients with malignant tumors [14]. In fact, these enzymes and the DRR have been reported as a significant prognostic biomarker in some malignancies [12, 13, 14]. However, the role of the DRR in TCa has not been presented in the published literature before.

In the present study, we evaluated the prognostic significance of preoperatively assessed DRR in the course of patients with TCa. Additionally, we described the threshold DRR for TCa metastasis.

Mean follow-up was 37±9.7 months. Five patients were excluded because of liver metastasis from TCa. Thus, 123 patients with TCa were enrolled in the present study. The patients' mean age was 39.3±10.3 years. The minimum levels of AST and ALT were 19 U/L and 8 U/L, respectively. The maximum levels of AST and ALT were 71 U/L and 57 U/L, respectively. Demographic and clinical data are summarized in Table 1. There were 45 and 73 TCa patients with any metastasis and without lymph node involvement, respectively. For both presence of retroperitoneal lymph node involvement and metastasis, an AST/ALT ratio of 1.30 was determined as an optimal threshold (Fig. 1, AUC: 0.674; Fig. 2, AUC: 0.705). Fig. 1 shows 45 metastatic TCa patients. Thus, a variety of tests were used to evaluate the risk of any metastasis.

Fig. 1
The risk of retroperitoneal lymph node involvement in testicular cancer patients at a De Ritis ratio level of 1.30 (area under the curve, 0.674; 95% confidence interval, 0.563–0.786). ROC curve, receiver operating characteristic curve.

• Click for larger image
• Download as PowerPoint slide

Fig. 2
The risk of metastasis in testicular cancer patients at a De Ritis ratio level of 1.30 (area under the curve, 0.705; 95% confidence interval, 0.535–0.875). ROC curve, receiver operating characteristic curve.

• Click for larger image
• Download as PowerPoint slide

Table 1
Demographic and clinical data of patients with testicular cancer (n=123)

• Click for larger image
• Click for full table
• Download as Excel file

Univariate analyses were done to evaluate the presence of retroperitoneal lymph node involvement and rete testis involvement. These analyses suggested 1.30 for DRR as the threshold for lymph node involvement and/or metastases. In the multivariate logistic regression model, age had a protective effect (OR, 0.946; 95% CI, 0.907–0.986), although a preoperative hCG level of 5,000 mU/mL and above and a DRR of 1.30 and above were risk factors (OR, 6.170; 95% CI, 1.122–33.934; OR, 3.830; 95% CI, 1.478–9.924) in the assessment of metastasis.

Tumor size, preoperative hCG, lymphovascular invasion, rete testis involvement, and a DRR of 1.30 and above were risk factors in the univariate analyses for metastasis (OR, 1.222; 95% CI, 1.039–1.438; OR, 12.240; 95% CI, 3.164–47.352; OR, 3.429; 95% CI, 1.153–10.195; OR, 8.571; 95% CI, 2.408–30.506; OR, 6.441; 95% CI, 1.935–21.440). In the multivariate regression model, only a preoperative hCG level of 5,000 mU/mL and above and a DRR of 1.30 and above were risk factors (OR, 16.253; 95% CI, 2.783–94.906; OR, 4.561; 95% CI, 1.021–20.375) (Tables 2, 3). We also showed survival in Kaplan–Meier curves for retroperitoneal lymph node involvement (Fig. 3A) and metastasis (Fig. 3B).

Fig. 3
Survival functions in Kaplan–Meier curves. (A) Retroperitoneal lymph node involvement. (B) Metastasis.

• Click for larger image
• Download as PowerPoint slide

Table 2
ORs on univariate analyses for presence of retroperitoneal lymph node involvement and metastasis in testis tumor patients

• Click for larger image
• Click for full table
• Download as Excel file

Table 3
Adjusted OR and 95% CIs for clinical outcomes in testis tumor patients in multivariate analyses

• Click for larger image
• Click for full table
• Download as Excel file

We investigated the relationship between the DRR and prognosis of TCa. Although we had a small patient population, we recorded 45 patients with TCa (36.6%) who had metastases. According to these findings, we determined the threshold DRR to be 1.30. Thus, our findings suggest that a preoperative DRR higher than 1.30 may be a sign of organ metastases or retroperitoneal lymph node involvement.

Diagnosis of TCa depends on clinical, laboratory, and radiological features and tumor markers [18]. From another point of view, however, none of these can accurately predict with certainty the presence of future metastases. If retroperitoneal lymph node involvement or metastasis is not detected in radiologic examinations, markers cannot provide information for metastases or retroperitoneal lymph node involvement. Here we showed evidence of the opposite: that an increased preoperative DRR might be an indicator for retroperitoneal lymph node involvement or organ metastases in TCa. In this way, the DRR can provide clinicians a means of predicting retroperitoneal lymph node involvement and metastases before inguinal orchiectomy.

Serum tumor markers are important in preoperative and postoperative follow-up for TCa. It is well known that the mean serum half-lives of hCG and AFP are 3 days and 5 days, respectively [7]. Thus, TCa tumor markers should be reevaluated after surgery, taking into account the half-life of tumor markers. In view of this, patients with clinical stage 1 TCa should undergo reassessment until normal values are measured. Markers are essential for categorizing TCa patients before starting chemotherapy [19]. The detection of elevated serum tumor markers after orchiectomy may be a sign of metastasis (macro- or micro). On the other hand, clinicians should be cognizant that normal values of a marker after orchiectomy do not exclude tumor metastases. Moreover, tumor markers should be followed-up, and persistence of elevated marker levels indicates a poor prognosis during chemotherapy [20, 21].

AST is produced in different tissues, whereas ALT is regarded as liver-specific [22]. In addition, some conditions like pathologic diseases with high proliferative properties, organ damage, and tumors with high cell turnover can elevate AST rather than ALT [22]. All these variations in levels of ALT and AST make DRR a candidate as a possible biomarker. The use of DRR as a biomarker has been reported previously in the published literature. Tan et al. concluded that a DRR of more than 2.0 might be a useful biomarker for determining survival in distal cholangiocarcinoma [21, 22, 23]. Dorward et al. [24] reported abnormal anaerobic glycolysis in cancer cells to produce adenosine triphosphate (ATP). Rapid tumor growth is related to increased ATP production, and this is used to determine mechanisms in tumor-specific imaging and antitumor therapy [25]. Increased glycolysis is also related to mitochondrial activities involved with nicotinamide adenine dinucleotide enzymes and glucose transporters [26]. Because tumor cells produce ATP faster, they prefer anaerobic glycolysis to oxidative phosphorylation [27]. In view of these data, an elevated DRR can be used as one indicator of tumor activity. In the literature, only a few studies have shown a connection between DRR and urologic cancers [28, 29, 30]. In these studies, the increased DRR was associated with poor prognosis. However, no published study has shown a relationship between DRR and TCa. In the present paper, we showed that an elevated DRR may predict possible metastasis (such as lymph node involvement and metastasis) of TCa. Either our results are parallel to the literature or this is the first study giving eligible proof of metastases in TCa patients with elevated DRR. Additionally, this is related to poor prognostic factors.

Although the current study reported new findings, it had several limitations. First, the study was performed in a retrospective manner and had a small patient population. Second, the DRR can be changed by undetected liver diseases, other diseases, and alcohol consumption, which can affect the levels of ALT and AST. This may suggest a large selection bias. However, this is the first study in the literature on this issue. Additionally, we did not compare perioperative and postoperative DRR. High levels of DRR were related to the presence of TCa. Surgical removal of the tumor, of course, leads to decreased DRR. Third, drug interactions could affect liver functions, leading to increased AST and ALT. Besides, retroperitoneal lymph node involvement and metastases were evaluated in computed tomography scans. Additionally, these were compared after treatments during follow-up. This might suggest a confounding bias. Moreover, we did not perform any molecular study. Furthermore, our results can be considered a pathfinder study in the way of testis-sparing surgery. However, testis-sparing surgery is still under debate. More series with higher patient volume are needed.

This is the first study in the literature that investigated the impact of amino transaminase on the prognosis of patients with TCa before surgery. In addition, AST and ALT are used in clinicians' daily practice. We identified that an increased DRR may be related to worsening prognosis in patients with TCa. Twelve patients died because of TCa; for this reason, we could not perform any survival analysis. Survival analysis should be performed for a larger series of patients in future studies.

An elevated DRR may be an independent risk factor for retroperitoneal lymph node involvement or metastases in patients with TCa at the time diagnosis, before surgery. If the DRR is higher than 1.30, clinicians should reevaluate to detect possible retroperitoneal lymph node involvement and metastases in TCa. Therefore, clinicians should inform patients before the operation concerning metastases in the case of a DRR <1.30, independent of radiological examinations. Our results need to be confirmed with prospective studies including larger sample sizes, including molecular studies.