- Top of page
Objective: To evaluate the prognosis of our series of patients with renal cell carcinoma (RCC) and tumor thrombus involving inferior vena cava (IVC) treated with nephrectomy and thrombectomy.
Methods: In 46 patients with unilateral RCC extending into IVC who underwent nephrectomy and thrombectomy (T3b in 38 patients, T3c in 6, T4 in 2, N+ in 15, M1 in 21), overall and cancer-specific survival rates were estimated, and the univariable and multivariable analysis were carried out to determine the prognostic factors among age, gender, performance status, fever, inflammatory laboratory parameters, nodal and distant metastasis, tumor thrombus level, pathological parameters and postoperative interferon-α administration.
Results: The median age was 66.5 (range 35–79) years. The median follow-up was 18.0 (mean 36.7 ± 38.7) months. The overall and cancer-specific 5-year survival rates were 32.9% and 40.0%, respectively. The univariate analysis revealed that fever (hazard ratio: HR 4.03), C-reactive protein (HR 4.89), grade of tumor cell (HR 3.83), and lymph node metastasis (HR 5.99) were independent prognostic factors of cause-specific survival in all patients. The multivariate analysis demonstrated that lymph node metastasis (HR 4.13) was the only independent prognostic factor of cause-specific survival. The extension level or postoperative interferon-α administration did not influence the prognosis of patients with tumor thrombus involving IVC.
Conclusions: Aggressive surgery should be considered first in RCC patients with any levels of tumor thrombus. However, patients with both IVC involvement and nodal metastasis showed significantly poor prognosis, and development of novel intensive multidisciplinary therapies will be needed.
- Top of page
Early stage renal cell carcinoma (RCC) has been increasingly among the renal incidentalomas found by convenient and efficient diagnostic imaging at health checks or when screening for other diseases. On the other hand, we still encounter a minority of patients with locally advanced RCC that developed with the extension into inferior vena cava (IVC). Extension of RCC into the IVC occurs in about 4–9% of patients newly diagnosed as having RCC.1–4 Tumor thrombus may extend as far as the right atrium in 1% of RCC patients.5 Even if the patients with locally advanced RCC had metastatic lesions, they underwent aggressive nephrectomy including tumor thrombectomy for the purpose of successful local control because the conventional conservative therapy such as immunotherapy did not seem to yield the promising outcome. Therefore, curable or imperative surgery is still supposed to be chosen in non-metastatic or metastatic patients with IVC extension until the role of molecular targeted medicines for these patients will be determined in the near future.
Accurate tumor staging with computerized tomography (CT) scanning or magnetic resonance imaging (MRI) greatly facilitated surgical planning and the safe removal of tumor thrombus even at proximal sites. In a thoracoabdominal approach by median sternotomy, extracorporeal cardiopulmonary bypass and profound hypothermal circulatory arrest have been used commonly in patients with supradiaphragmatic or atrial extension of tumor thrombus. The transesophageal echography-guided endoluminal occlusion technique is very useful in the prevention of secondary embolization of detached tumor thrombus and assistance for vascular clamp or taping. Mobilization of the liver is reliable in vascular control and large cavotomy with an en-bloc removal of the thrombus and the tumor. Liver mobilization and natural veno-venous bypass are less-invasive procedures as substitutes for cardiopulmonary bypass in patients with supradiaphragmatic tumor thrombus. Recent improvement in the surgical procedures and substantial management supporting aggressive operation for RCC patients with IVC extension have dramatically increased the ability to remove the tumor thrombus safely and decrease the perioperative mortality and morbidity.
We herein review our series of patients with RCC extending into IVC and assessed the implications for prognosis.
- Top of page
Out of a total of 1206 newly diagnosed patients who underwent nephrectomy or partial nephrectomy for unilateral localized and advanced RCC in Nara Medical University and its affiliated hospitals between January 1980 and December 2005, 46 patients (3.8%) with RCC extending into IVC were subjected to retrospective analysis. The patients consisted of 38 men and eight women, and their median age was 66.5 years (range from 35 to 79). The clinical presentation, laboratory data, level of tumor thrombus extending into IVC, operation time, estimated blood loss, transfusion, complication, pathological features, postoperative immunotherapy, overall and cancer-specific survival were evaluated.
Tumor staging relied on preoperative imaging studies; namely, chest computerized tomography (CT) scanning, abdominal CT or magnetic resonance imaging (MRI), echocardiography, transesophageal echocardiography, and bone scanning. Tumor thrombus staging was based on the Neves classification, that is, level I: involving the infrahepatic IVC; level II: involving the retrohepatic IVC with close proximity to the main hepatic veins; and level III: involving the supra-diaphragmatic IVC or right atrium. The pathological features assessed were the tumor grade, subtype and 2002 TNM classification edited by Union Internationale Contre le Cancer (UICC).6 The histological tumor subtypes were classified as conventional clear cell carcinoma, papillary renal carcinoma, chromophobe cell carcinoma, collecting duct carcinoma, and RCC not otherwise specified following UICC, American Joint Committee on Cancer and Heidelberg guidelines.7
Our policy of the treatment strategy for RCC with extension into IVC was basically surgical extirpation of entire tumor tissue and postoperative interferon-α (INF-α) administration to avoid sudden death and to expect survival benefit. Patients with worsening systemic conditions, such as extremely advanced metastatic disease and severe complications, were considered inoperable. Cytoreductive surgeries for metastatic lesions were selected if applicable. The retrospective nature of this study precluded a standardized follow-up protocol to identify distant metastasis and local recurrence in all patients. The majority of patients underwent and chest CT every 3 months for the first 3 years, every 6 months for the following 2 years, and every year thereafter, and abdominal CT or ultrasound sonography every 3 months for the first 2 years, every 6 months for the following 2 years, and every year thereafter, and bone scanning every 6 months for the first 2 years and every year thereafter. Brain scanning was carried out only when site-specific symptoms emerged. No patients received preoperative neoadjuvant radiotherapy or chemotherapy or immunotherapy including IFN-α or interleukin-2. Postoperative immunotherapy was proposed to all patients, and INF-α was given to the patients who consented to this therapy.
The overall and cancer-specific survival rates were estimated using the Kaplan-Meier method and the log-rank test. Survival was determined from the date of nephrectomy with thrombectomy to the date of last follow up or death. The univariable and multivariable analysis by using a Cox proportional hazard regression model was used for variables that showed a P-value of <0.05 in the univariable analysis to evaluate the multiple predictors of the outcome, and their hazard ratio (HR) with 95% confidence interval (CI) was calculated. Statistical analyses were carried out using the SPSS software package with a P-value of <0.05 considered as statistically significant.
- Top of page
Patient and tumor characteristics are listed in Table 1. The tumor was right-sided in 30 patients and left-sided in 16 patients. The maximum tumor size ranged from 4.0 to 21.0 cm with a median of 8.9 cm. Thirty-nine patients were symptomatic with tumor and seven patients had asymptomatic incidentaloma accompanying tumor thrombus at the initial presentation. Twenty patients presented with febrile signs. Erythrocyte sedimentation rate was abnormal in 31 patients and α 2-globulin was elevated in 31 patients and C-reactive protein was positive at 0.6 mg/dL or greater in 37 patients. European Cooperative Oncology Group performance status (PS) was scored as PS0 in 22 patients, PS1 in 14, PS2 in six, PS3 in three and PS4 in one patient.
Table 1. Patients and tumor characteristics (n = 46)
|Age: median (range) (years)||66.5 (35–79)|
|Symptomatic (%)||39 (84.8)|
|Tumor size: median (range) (cm)||8.9 (4–21)|
|TNM categories (%):|
| T3b||38 (82.6)|
| T3c||6 (13.0)|
| T4||2 (4.4)|
| N0M0||19 (41.3)|
| N + M0||6 (13.0)|
| N0M1||12 (26.1)|
| N + M1||9 (19.6)|
|Tumor cell grade (%):|
| G1||3 (6.5)|
| G2||20 (43.5)|
| G3||23 (50.0)|
| Clear cell||32 (69.6)|
| Non-clear cell||14 (30.4)|
|Follow up: median (mean; range) (months)||18 (36.7 ± 38.7; 2–160)|
Histological subtypes were clear cell subtype in 32 tumors and non-clear cell subtype in 14 tumors containing seven conventional granular cell carcinomas, four spindle cell carcinomas, two papillary renal carcinomas, and one chromophobe cell carcinoma. Malignancy of tumor cells were graded as grade 1 in three patients, grade 2 in 20, and grade 3 in 23 patients. Clinical T staging revealed T3b tumor in 38 patients, T3c in six and T4 in two patients. Of all patients, 32 had level I tumor thrombus, eight had level II, and six had level III tumor thrombus. Metastatic disease was found in 27 patients (lung metastasis in 17 patients, bone metastasis in three, liver metastasis in one, and lymph node metastasis in 15 patients).
A cardiopulmonary bypass with hypothermia and temporary cardiac arrest procedures was used in one of 32 patients with infraheptic IVC tumor extension because of remote tumor thrombus floating in the pulmonary artery and also in five of six patients with tumor thrombus of level I. Of the eight patients with level II tumor thrombus, none needed cardiopulmonary bypass circulation. Veno–venous bypass assisted by filter manipulation was adopted for one patient with level III tumor thrombus after liver mobilization. The median operation time was 361 (range: 175–630) in level I, 503 min (range: 305–910) in level II, and 657 min (range: 570–930) in level III. The median blood loss was 3439 mL (range: 570–5972) in level I, 4546 mL (range: 600–13 685) in level II, and 5790 mL (range: 4990–8215) in level III. Transfusion was required in 25 of 32 patients with level I tumor thrombus, whereas all patients with levels II and III needed transfusion during the operation time. Pulmonary infarction did not occur in any patients. Major perioperative morbidity occurred in nine cases (19.6%) as follows: pulmonary edema in three patients, ileus in two patients, and cerebral infarction, thrombocytopenia, liver dysfunction, and venous thromboembolism in one patient each. No surgery-related death within postoperative 30 days was identified.
The median follow-up period after nephrectomy and thrombectomy was 18.0 months (mean: 36.7 ± 38.7; range: 2–160 months). The median postoperative survival time at a 50% survival rate among all patients was 34.0 months. The estimated 1-, 2- and 5-year overall survival rates were 69.7%, 54.8% and 32.9%, respectively, whereas the estimated 1-, 2- and 5-year cancer-specific survival rates were 71.7%, 61.9% and 40.0%, respectively. The tumor thrombus level and distant metastasis at presentation did not yield significant effects on the cancer-specific survival in this study (P = 0.6395 in Fig. 1 and P = 0.2795 in Fig. 2a, respectively). Lymph node metastasis had a significant influence on the cancer-specific survival (P < 0.0002, Fig. 2b).
Figure 1. Cancer-specific survival curves according to the level of tumor thrombus in the inferior vena cava. No significant differences in the cause-specific survival rates were observed among the tumor thrombus levels (P = 0.6395). n, number of patients at risk at time points 0, 1, 2, and 5 years.
Download figure to PowerPoint
Figure 2. (a) Cancer-specific survival curves by distant metastasis. There was no significant difference in the cause-specific survival between patients with and without distant metastasis (P = 0.2795). (b) Cancer-specific survival curves by lymph node metastasis. There was significant difference in the cause-specific survival between patients with and without lymph node metastasis (P < 0.0002). n, number of patients at risk at time points 0, 1, 2, and 5 years.
Download figure to PowerPoint
Thirty-three patients including 22 patients with nodal and/or distant metastasis received the postoperative IFN-α administration (300 × 104 IU/day, 2 or 3 days a week) for 22.3 ± 18.1 months of the mean administration period. The remaining 13 patient including five with nodal and/or distant metastasis did not receive or maintain IFN-α administration. The postoperative IFN-α administration was not useful among all patients with RCC extending into IVC (P = 0.8447, Fig. 3). There were no differences in age, PS, grade, stage, and thrombus level between the groups postoperatively treated with and without IFN-α (χ2 and Mann–Whitney U-tests).
Figure 3. Cancer-specific survival by postoperative interferon-α (IFN-α) therapy. There was no significant difference in the cause-specific survival between patients who received IFN-α (300 × 104 IU/day, 2 or 3 days a week, 22.3 ± 18.1 months of the mean administration period) and those who did not receive IFN-α therapy (P < 0.8447). n, number of patients at risk at time points 0, 1, 2, and 5 years.
Download figure to PowerPoint
Table 2 lists the P-values and HRs with 95% CI in all patients. Univariable analysis revealed that fever (HR 4.03), C-reactive protein (HR 4.89), tumor grade (HR 3.83), and lymph node metastasis (HR 5.99) were independent predictors of the cancer-specific survival in all patients with a P-value of less than 0.05. The multivariable analysis demonstrated that lymph node metastasis (HR 4.13) was the only independent prognostic factor of the cause-specific survival among those four independent predictors in the univariable analysis.
Table 2. Univariable and multivariable analysis (n = 46)
|Parameter||Univariable analysis||Multivariable analysis|
|Hazard ratio||95% CI||P-value||Hazard ratio||95% CI||P-value|
|Age (<65/65≤)||1.085||0.956–1.130||0.5020|| || || |
|Sex (men/women)||2.243||0.995–6.957||0.0603|| || || |
|PS (0/1 or 2 or 3 or 4)||0.711||0.289–1.896||0.5001|| || || |
|ESR (pos/neg)||2.357||0.789–6.892||0.1505|| || || |
|α 2-globulin (pos/neg)||1.997||0.678–5.502||0.1507|| || || |
|Tumor side (right/left)||1.298||0.433–3.700||0.5171|| || || |
|Tumor size (<10/10≤)||0.889||0.788–1.008||0.1280|| || || |
|Grade (3/1or 2)||3.833||1.501–10.432||0.0180||2.356||0.789–7.124||0.1231|
|Cell subtype (clear/non-clear)||0.771||0.389–1.699||0.3500|| || || |
|Thrombus level (I/II or III)||0.995||0.634–1.723||0.7885|| || || |
|LN metastases (yes/no)||5.994||2.003–14.112||0.0007||4.125||1.638–12.998||0.0054|
|Distant metastases (yes/no)||2.004||0.758–3.543||0.2986|| || || |
|Postoperative IFN-α (yes/no)||0.916||0.515–1.805||0.5640|| || || |
- Top of page
Renal cell carcinoma occasionally silently invades the venous system in 4–10% of cases.1–4 Many reports suggest that tumor thrombus extending into IVC is not necessarily associated with a worse prognosis and demonstrats that patients with tumor thrombus but no metastasis can achieve long-term survival after complete surgical extirpation.3,8–11 In the patients with non-metastatic RCC and IVC involvement, the 5-year survival rate ranges between 18% and 68% after complete surgical extirpation.3,9–12 The estimated 5-year overall and cancer-specific survival rates (32.9% and 40.0%, respectively) in our series of patients with tumor thrombus extending into IVC were similar to those of early reports.9,11,13
Extracorporeal cardiopulmonary bypass and profound hypothermal circulatory arrest have been used commonly as a standard procedure in patients with supradiaphragmatic or atrial extension of tumor thrombus. Recently a less invasive and more convenient procedure of liver mobilization and natural veno-venous bypass has been developed and used electively in patients with intrahepatic IVC extension. The improvement of these surgical procedures and substantial management supporting aggressive operation for RCC patients with IVC extension have dramatically increased the ability to remove the tumor thrombus safely.14,15 However, the surgical management of RCC with tumor thrombus extension still remains a technically challenging endeavor. Major morbidities occurred in nine cases (19.6%) of our patients. There were no perioperative deaths and none of the patients underwent a second operation due to these complications. The long-term survival may be due to success of complete surgical excision of RCC and safe and high-quality perioperative management and care of these patients.
There is a controversy concerning the significant prognostic factors of the tumor thrombus level in IVC.3,9–12 Most early reports concluded that tumor thrombus level, lymph node involvement, and distant metastasis were considered as important prognostic factors while a few reports concluded that tumor grade, PS, and stage (particularly perinephric fat invasion) were prognostic factors.14,15 In the present study, the tumor thrombus level and distant metastasis at presentation did not have significant effects on the cancer-specific survival. As is often reported by early authors, we also emphasize that lymph node metastasis had a significant effect on the cancer-specific survival (P < 0.0002). The prognostic factors reported in the 13 early studies were distant metastasis in seven reports, thrombus level in three reports, lymph node metastasis in three reports including ours, tumor grade in three reports, and incomplete tumor-thrombus resection or T stage or performance status in one each.3,5,8–12,15–19
In our patients with no metastasis, the adjuvant INF-α showed no significant influence on the cause-specific survival when compared with the surgery alone (P < 0.6572). On the other hand, in the patients with nodal and/or distant metastasis, the postoperative INF-α administration seemed to yield some benefit for the cause-specific survival (P < 0.028, data not shown). However, interpretation of our retrospective data had limits because of the small number of patients and short period of follow up.
The only promising curative approach to RCC with IVC involvement is aggressive surgery and immunotherapy of the tumor and thrombus because no standard agent or combination program has yet been recognized as the best effective therapy. Our data may support the notion that patients with RCC involving IVC should be initially considered to undergo aggressive surgery even in a high level of IVC extension, the difference of which did not influence postoperative survival. However, the prognosis of advanced RCC patients with both IVC involvement and nodal metastasis was very poor. Therefore, the development of novel intensive multidisciplinary therapies including molecular targeting, against cytokine-resistant or inoperable RCC with IVC involvement will be a pressing need.