Statistical analysis
All data were stored using an Excel database, and analyses were performed using SPSS v. 21.0 (IBM Corporation, NY, USA). The distribution of OS and DFS was evaluated using the Kaplan-Meier method. The differences between the subgroups were assessed by the log-rank test. The parameters found to be statistically significant in the univariate analysis were further evaluated using multivariable models. The Multivariate Cox proportional risk regression model was fitted to the data to estimate the independent prognostic importance of survival. Statistical significance was accepted as a p-value of <0.05.
Results
Radical nephrectomy was performed on 703 patients with histologically confirmed renal carcinoma diagnosed from 2006 to 2019. Renal carcinoma and associated venous thrombosis were detected in 38 of 703 (5.4 %) patients. Two patients were excluded from the study due to perioperative death, and a further three due to missing data. Thirty-three patients diagnosed with renal carcinoma and associated level I to IV venous thrombosis, according to the Mayo Clinic classification system, were enrolled in the study.
There were 22 male (66.7%) and 11 female (33.3%) patients. The median age was 57.6 years (27 to 77 years), with a median follow-up of 35.3 months (8 to 108 months). A total of 15 patients (45.5%) died during the follow-up. There were 15 patients (45.4 %) with level I, five (15.2%) with level II, eight (24.2%) with level III, and five (15.2%) with level IV venous thrombosis. A high Mayo Clinic thrombus level was associated with poor DFS in the univariate analysis (p = 0.002); however, according to the multivariate analysis, the thrombus level was not independently associated with DFS (p = 0.066).
While 24 patients (72.8%) had no distant metastasis, nine (27.2%) had clinical M1 disease at the time of diagnosis. Preoperative clinical M1 disease was associated with poor OS (56 months vs. 25 months) in the multivariate analysis (p = 0.008). There were eight (24.2%) patients with clinical N1 disease, and lymph node dissection was performed in 21 (63.6%) patients. Preoperative clinical N1 disease was not associated with OS (p = 0.973). Eight patients (24.2%) required sternotomy due to intra-atrial thrombus excision, coronary artery bypass grafting, or valvuloplasty. Other baseline clinical findings, tumor characteristics, and perioperative and postoperative findings are shown in Table 1a and 1b .
The most common histopathological type was renal cell carcinoma (RCC) that presented in 30 patients (90.7%). Poor median OS and DFS were found in patients with non-RCC histopathologies (11.7 months vs. 35.2 months). Sarcomatoid differentiation was seen in eight patients (24.2%). Tumors that coexisted with sarcomatoid differentiation were related to poor median OS (34 months vs. 58 months). However, non-RCC histopathologies and sarcomatoid differentiation were not found statistically associated with median OS and DFS. Postoperative N1 disease was not associated with OS (p = 0.066) and DFS (p= 0.437). The remaining pathological findings are shown in Table 1c .
The median pathological tumor size was 10.9 cm (5 to 20 cm). The median overall survival was 32.1 months in patients with a tumor of 10.1 cm or greater in size, and 47.8 months in those with a tumor sized below 10.1 cm. Increased tumor size (10.1 cm or greater) was associated with poor OS (p = 0.046) and DFS (p = 0.005) according to the univariate analysis. In the multivariate Cox regression analysis conducted by adjusting the remaining clinical and pathological variables, tumor size was independently correlated with poor OS (p = 0.02) but not correlated with DFS (p = 0.129). The multivariate analyses of the parameters calculated to be statistically significant in the univariate analysis are given in Table 2 .
Discussion
In this study, it was found that tumor size and preoperative clinical M1 disease were correlated with poor OS. Although poor DFS was observed in patients with a higher thrombus level and greater tumor size (10.1 cm and over), these parameters did not have a statistically significant effect on DFS in the multivariate analysis.
A venous thrombus is often associated with metastasis at presentation. In most of the series reported in the literature, up to 30% of patients with renal cancer and associated venous thrombosis have clinical M1 disease [7,16]. While patients presenting without distant metastasis have a good prognosis when treated with successful resection, preoperative clinical M1 disease is one of the poor prognostic factor for OS in patients with renal carcinoma and associated venous thrombosis [17,18]. In a study with 87 patients with preoperative clinical M1 disease, Ciancio et al. reported that the median OS was reduced to eight months [19]. Our study also showed a decrease in OS (56 months vs. 25 months) in patients with preoperative distant metastasis (p = 0.008) (Figure 2a ), which is consistent with the literature. A significant decrease in OS despite surgical treatment in preoperative clinical M1 patients led researchers to seek new treatment alternatives. In this context, new oncological treatment options are considered before and after surgery. In addition, systemic treatments are investigated as an alternative to surgery. In a study conducted by Mejean et al., it was reported that sunitinib treatment without nephrectomy in the patients with metastatic RCC with moderate- or high-risk disease was not inferior to the standard treatment of nephrectomy plus sunitinib [20]. In another study by Bex et al., it was reported that sunitinib treatment prior to surgery in patients with metastatic RCC provided an OS advantage compared to early surgery with no additional sunitinib treatment [21]. Therefore, advanced clinical trials may offer better treatment alternatives in patients with metastatic RCC [22,23].
There are a few studies concerning the association between tumor size and survival outcomes [19,25]. In these studies, the reduction of OS and DFS was emphasized. In our study, tumor size was found to be independently related to poor OS in the multivariate analysis (Figure 2b ). Additionally, the most significant cut-off value in terms of the reduction of survival was calculated as 10.1 cm, while the literature does not contain an exact cut-off value. Although patients with venous thrombosis are directly classified as T3 according to the TNM staging system, tumor size should also be taken into account for optimal disease management.
In the literature, the OS outcomes of patients with tumor thrombosis are inconsistent. Although some studies reported poor survival in patients with a more cephalad tumor thrombosis [15,26,27], others did not find such differences [24,28]. A critical research investigating this issue was conducted with 1,192 patients over a median follow-up of 61.4 months [24] and determined the median OS was 52 months for renal vein thrombosis, 26 months for subdiaphragmatic IVC thrombosis, and 18 months for supradiaphragmatic IVC thrombosis. However, these differences in survival were not found to be related to the thrombus levels. Similarly, the association between the thrombus level and OS was not statistically significant in the current study (Figure 2c ).
In the literature, the presence of non-RCC histopathology was found to be related to poor prognosis [28,29]. In a retrospective study, statistically significant poor OS was shown in patients presenting with sarcomatoid differentiation [29]. Although our study revealed a difference in OS between the patients that presented with RCC and non-RCC histopathology (35.2 months vs. 11.7 months), this difference was not found statistically significant (p = 0.860). In addition, we did not find any statistically significant difference in the OS of patients that had sarcomatoid differentiation (58 months vs. 34 months,p = 0.810). This finding may be due to sampling bias since 90.7% of the tumors in our series showed clear cell histology, while 24.2% had sarcomatoid differentiation.
We acknowledge that the current study had certain limitations. It was conducted in a single center with a retrospective design. In addition, the size of our cohort was relatively small, with a population size of 33 patients. Furthermore, the study population underwent surgery performed by multiple surgeons, and we did not attempt to differentiate venous wall infiltration from the venous thrombus alone. Lastly, the length of follow-up was relatively poor. However, we consider that this study has a potential role in contributing to the literature in terms of the effect of tumor size on the survival of patients with renal carcinoma and associated venous thrombosis. Prospective studies with a long-term follow-up and larger population are needed to validate our findings.
Conclusions
Renal carcinoma presenting with associated venous thrombosis is a potentially curable condition that offers reasonable survival. Although there remains controversy regarding the prognostic significance of tumor thrombus involvement and other clinic parameters, there are no prospective studies for predicting mortality, morbidity, and survival findings. The results from our study demonstrate a significant decrease in the OS of patients with clinical M1 and larger tumor size (>10.1 cm). However, there is no statistically significant association between thrombus level and survival (OS and DFS).