Fax: (904) 953-2218
High expression levels of survivin protein independently predict a poor outcome for patients who undergo surgery for clear cell renal cell carcinoma
Article first published online: 30 MAY 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 1, pages 37–45, 1 July 2006
How to Cite
Parker, A. S., Kosari, F., Lohse, C. M., Houston Thompson, R., Kwon, E. D., Murphy, L., Riehle, D. L., Blute, M. L., Leibovich, B. C., Vasmatzis, G. and Cheville, J. C. (2006), High expression levels of survivin protein independently predict a poor outcome for patients who undergo surgery for clear cell renal cell carcinoma. Cancer, 107: 37–45. doi: 10.1002/cncr.21952
- Issue published online: 16 JUN 2006
- Article first published online: 30 MAY 2006
- Manuscript Accepted: 11 JAN 2006
- Manuscript Revised: 1 JAN 2006
- Manuscript Received: 8 AUG 2005
- clear cell renal cell carcinoma;
- nuclear grade;
- prognostic biomarkers;
- protein isoforms;
- survivin expression
In a previous study of gene array data, the authors identified survivin as a candidate marker of aggressiveness in clear cell renal cell carcinoma (ccRCC). What remained in question was whether survivin expression at the protein level is an independent predictor of disease progression and cancer-specific survival.
Between 1990 and 1994, 312 patients underwent nephrectomy for ccRCC at Mayo Clinic Rochester and had paraffin tissue available. The authors performed immunohistochemistry with antisurvivin antibody, quantitated the expression by using an image-analysis system, and analyzed the association of survivin expression with disease progression and cancer-specific survival.
Within the cohort, 97 patients (31.1%) had high levels of survivin expression. Patients who had high survivin expression levels were at significantly increased risk of death from RCC compared with patients who had low expression levels (risk ratio [RR], 5.3; 95% confidence interval [95% CI], 3.5–7.9). The 5-year cancer-specific survival rate was 43.0% for patients with high survivin expression and 87.2% for patients with low survivin expression. In multivariate analysis, survivin expression remained associated with death from RCC even after adjusting for the Eastern Cooperative Oncology Group performance status; 2002 Tumor, Lymph Node, Metastases (TNM) stage groupings and nuclear grade (RR, 2.4; 95%CI, 1.5–3.8); and the Mayo Clinic composite TNM stage groupings, tumor size, nuclear grade, and tumor necrosis (SSIGN) score (RR, 1.8; 95%CI, 1.1–2.9). Among 273 patients who had localized ccRCC, survivin expression was associated significantly with cancer progression (RR, 3.9; 95%CI, 2.4–6.2).
Survivin expression is an independent predictor of ccRCC progression and death from RCC. Thus, survivin has the potential to offer additional prognostic information and to provide a novel target for the development of new adjuvant therapies. Cancer 2006. © 2006 American Cancer Society.
Incidence and mortality rates for renal cell carcinoma (RCC) have been rising steadily in the United States for nearly 3 decades.1 RCC encompasses 4 specific subtypes, but most (>80%) of RCC tumors are classified as clear cell RCC (ccRCC), and this subtype is responsible for the majority of deaths caused by the disease.2 To date, surgical excision remains the only curative therapy for patients who are diagnosed with localized ccRCC. Currently, approximately 33% of patients with ccRCC present with disseminated disease at diagnosis,3 and another 25% to 30% of patients who undergo surgery for clinically localized ccRCC will experience cancer progression (i.e., distant metastases) after excision of their tumor.4, 5 Once metastatic disease develops, the 5-year survival rate for patients with ccRCC drops from 60% to <5%.6
Although the development of prognostic scoring systems based on clinical and pathologic features has proven useful in the management of patients with RCC,7–10 there remains a tremendous need for prognostic biomarkers.11, 12 Indeed, the identification of molecular markers that are predictive of ccRCC aggressiveness and patient outcome has the potential to not only improve our ability to manage patients but also to identify molecular mechanisms of cancer progression and inform on new targets for adjuvant therapies. Unfortunately, to date, there has been only limited success in identifying biomarkers that offer predictive information beyond that provided by standard clinical and pathologic prognostic features. Moreover, many of the markers that have been identified do not appear to provide a clear opportunity for the development of new therapeutic interventions.
We recently conducted genomic profiling to compare gene expression patterns between aggressive and nonaggressive forms of ccRCC.13 These gene array experiments revealed that messenger RNA (mRNA) for the human protein survivin is over-expressed in aggressive ccRCC compared with nonaggressive ccRCC. In addition, follow-up validation studies using reverse transcriptase-polymerase chain reaction (RT-PCR) analysis and immunohistochemistry confirmed that expression levels of survivin are associated significantly with more aggressive disease. Despite these significant associations of survivin expression with ccRCC aggressiveness, it remains unclear whether survivin expression is an independent predictor of ccRCC patient outcome.
For the current study, we used follow-up data from a large cohort of patients with ccRCC who underwent surgery at Mayo Clinic from 1990 to 1994 to report that patients who have tumors that show high levels of survivin expression are at markedly increased risk of cancer progression and death from RCC compared with patients who have tumors that show low survivin expression levels. More important, the association of survivin expression and patient outcome remains significant even after adjustment for well known predictors of RCC outcome.
MATERIALS AND METHODS
Upon approval from the Institutional Review Board, we identified 427 patients who underwent radical nephrectomy or nephron-sparing surgery for unilateral, sporadic ccRCC between 1990 and 1994 from the Mayo Clinic Nephrectomy Registry. Of these, 312 patients (73.1%) had archived paraffin embedded tissue available for study. There was no statistically significant difference in cancer-specific survival between patients with and without tissue available for study (P = .992; log-rank test).
Clinical and Pathologic Features
The clinical features studied included age, gender, symptomatic presentation, Eastern Cooperative Oncology Group (ECOG) performance status, tumor thrombus level, and type of surgery. Patients who had a palpable flank or abdominal mass, discomfort, macroscopic hematuria, acute onset varicocele, or constitutional symptoms (including rash, sweats, weight loss, fatigue, early satiety, and anorexia) were considered symptomatic at presentation.
The pathologic features studied included histologic subtype classified according to the Union Internationale Contre le Cancer, American Joint Committee on Cancer, and Heidelberg guidelines14, 15; tumor size; the 2002 primary tumor classification; perinephric fat invasion; regional lymph node involvement; distant metastases; the 2002 Tumor, Lymph Node, Metastasis (TNM) stage groupings; nuclear grade16; coagulative tumor necrosis17; and sarcomatoid differentiation.18 A urologic pathologist (J.C.C.) reviewed the microscopic slides from all specimens without knowledge of patient outcome or survivin expression level.
Disease status for patients in the Nephrectomy Registry is updated each year. If a patient has not been seen at the Mayo Clinic in the previous year, then the patient is sent a disease status questionnaire. If there is evidence of disease progression in this questionnaire, then the date, location, and treatment are verified in writing with the patient's local physician. Patient vital status is updated similarly on a yearly basis. If a patient has died in the previous year, then a death certificate is ordered to determine the cause of death. A recent visit to the Mayo Clinic (within 6 months of the date of death) for metastatic RCC is good documentation that RCC was the cause of death. If the death certificate does not support this cause, then the medical history is reviewed with a Mayo Clinic urologist to determine the cause of death. If a death certificate cannot be obtained, then the cause of death must be verified with the patient's family or local physician.
Survivin Immunoperoxidase Staining and Evaluation
A histotechnologist in the Mayo Clinic Tissue Acquisition and Cellular/Molecular Analysis (TACMA) facility dissected 5-μm sections from representative paraffin embedded tissue blocks for each member of the cohort. The study pathologist selected the block with the highest grade tumor for dissection. The slides were stained with antisurvivin (Dako, Carpenteria, CA; 1/100 dilution) by using standard techniques. This antibody recognizes an epitope on the N-terminal portion of the survivin molecule and reacts with wild-type survivin as well as recently identified rare variant forms of survivin. The study pathologist reviewed the stained slides and circled the area of greatest staining for further analysis (from 1 mm to 4 mm in greatest dimension). In every section, there was at least 1 positive-stained cell to indicate that the staining process was successful and that the antigen was preserved in the tissue section. The staining pattern was intense and nuclear, allowing for the separation from nonspecific cytoplasmic or membrane staining. A cytotechnologist and imaging specialist in the TACMA facility scanned each slide using a Slide Scanner (Bacus Laboratories, Inc.). The BLISS system is capable of digitally capturing images at 480 × 752 pixel resolution and at × 40 magnification. The entire slide is composed of multiple tiles to create a mosaic or composite picture. Computer-assisted analysis was performed using the IHC (Immunohistochemistry) Score software (Bacus Laboratories, Inc.) to obtain measurements of total area and IHC area. The percentage of total area that stained positive for the antisurvivin antibody was used as an overall measure of survivin expression.
We selected a dichotomized indicator variable (low vs. high) as the primary analysis variable for survivin expression. To determine the best cut-off point for the low and high expression categories, we used scatter plots of survivin expression against the difference in observed survival and the survival expected from a Cox proportional hazards regression model (formally known as the Martingale residuals). Martingale residuals can be used to evaluate the functional form of a continuous variable in the setting of a Cox model and, thus, can be helpful when selecting a logical cut-off point for illustrative purposes and further analysis.19 These analyses indicated a cut-off point at 2% survivin expression (Fig. 1). In secondary analysis, we also evaluated survivin expression using categories based on tertile and quartile cut-off points but found no evidence of a dose response.
We compared the distribution of clinical and pathologic features between patients with low and high survivin expression levels by using chi-square and Fisher exact tests. To visualize the association of survivin expression with progression-free and cancer-specific survival, we constructed Kaplan–Meier curves and evaluated any differences in outcome by expression level with log-rank tests. The duration of follow-up was calculated from the date of surgery to the date of cancer progression (i.e., distant metastases), the date of death from RCC, or the date of last known follow-up. We then fit univariate and multivariate Cox proportional hazards regression models to estimate the magnitude of the association of survivin expression and patient outcome (progression and cancer-specific survival). These associations were summarized by using risk ratios (RR) and 95% confidence intervals (95%CI).
Each of the clinical and pathologic covariates studied was tested first for interaction with the dichotomous survivin expression variable by 1) conducting a stratified analysis of survivin expression by the levels of the covariate in question and 2) formally testing the interaction term between survivin expression and the covariate. If a covariate was not deemed to interact with survivin expression, then we adjusted for the confounding effects of that covariate by including it in a model with the variable for survivin expression. The covariates that were considered simultaneously in our multivariate model included ECOG performance status, the TNM stage groupings, and nuclear grade. We also constructed an additional multivariate model adjusting for the Mayo Clinic SSIGN Score, which is a composite scoring system that was developed specifically for patients with ccRCC.9 The SSIGN Score combines TNM stage groupings, tumor size, nuclear grade, and tumor necrosis into a single summary score with a high degree of accuracy for predicting death from RCC. All statistical analyses were performed using the SAS software package (SAS Institute, Inc., Cary, NC), and P values <.05 were considered statistically significant.
At last follow-up, 179 of 312 patients studied had died, including 99 patients who died from RCC at a median of 2.2 years after nephrectomy (range, 0–13 years). Among the 133 patients who remained alive at last follow-up, the median duration of follow-up was 11.3 years (range, 0–15 years). Cancer-specific survival rates (the number of patients still at risk) at 1 year, 5 years, and 10 years after nephrectomy were 90.6% (264 patients), 73.4% (192 patients), and 67.2% (129 patients), respectively. Among the subset of 273 patients who had clinically localized ccRCC at the time of surgery, 68 patients (25%) progressed to distant metastases at a median of 1.6 years after nephrectomy (range, 0–12 years). Progression-free survival rates (the number of patients still at risk) at 1 year, 5 years, and 10 years after nephrectomy were 90.1% (226 patients), 79.5% (174 patients), and 72.6% (114 patients), respectively.
Survivin expression was predominantly nuclear among the members of this cohort. The mean survivin expression level for the entire cohort was 2.3% (median, 1.0%; range, 0.01% to 35.8%). There were 97 patients (31.1%) with tumors that were classified with high survivin expression levels (≥2%). Figure 2 shows representative stains from a tumor with low survivin expression (Fig. 2A) and a tumor with high survivin expression (Fig. 2B). Table 1 provides a comparison of clinical and pathologic features between patients who had tumors with low and high survivin expression. The table shows that survivin expression was associated significantly with several adverse pathologic features. Specifically, patients who had tumors with high survivin expression were more likely to have tumor thrombus, larger tumor size, advanced tumor stage, higher grade, tumor necrosis, and an increased likelihood of regional lymph node involvement and distant metastases at the time they underwent nephrectomy compared with patients who had tumors with low survivin expression.
|Survivin Expression Level: No. of patients (%)|
|Feature||Low (<2%), n = 215||High (≥2%), n = 97||P|
|Age at surgery|
|<65 y||114 (53.0)||41 (42.3)||.079|
|≥65 y||101 (47.0)||56 (57.7)|
|Female||91 (42.3)||36 (37.1)||.386|
|Male||124 (57.7)||61 (62.9)|
|Symptoms at presentation|
|No||79 (36.7)||26 (26.8)||.086|
|Yes||136 (63.3)||71 (73.2)|
|Constitutional symptoms at presentation|
|No||162 (75.4)||65 (67.0)||.126|
|Yes||53 (24.6)||32 (33.0)|
|ECOG performance status|
|0||191 (88.8)||91 (93.8)||.168|
|≥1||24 (11.2)||6 (6.2)|
|None||181 (84.2)||59 (60.8)||<.001|
|Level 0||18 (8.4)||25 (25.8)|
|Level I–IV||16 (7.4)||13 (13.4)|
|Type of surgery|
|Nephron-sparing surgery||36 (16.7)||6 (6.2)||.011|
|Radical nephrectomy||179 (83.3)||91 (93.8)|
|Primary tumor size|
|<5 cm||101 (47.0)||22 (22.7)||<.001|
|From 5 cm to <7 cm||50 (23.3)||18 (18.6)|
|From 7 cm to <10 cm||31 (14.4)||29 (29.9)|
|≥10 cm||33 (15.3)||28 (28.9)|
|2002 Primary tumor classification|
|pT1a||83 (38.6)||13 (13.4)||<.001|
|pT1b||71 (33.0)||23 (23.7)|
|pT2||17 (7.9)||11 (11.3)|
|pT3a||9 (4.2)||12 (12.4)|
|pT3b||29 (13.5)||35 (36.1)|
|pT3c||4 (1.9)||2 (2.1)|
|pT4||2 (0.9)||1 (1.0)|
|Perinephric fat invasion|
|No||184 (85.6)||67 (69.1)||<.001|
|Yes||31 (14.4)||30 (30.9)|
|Regional lymph node involvement|
|pNX and pN0||210 (97.7)||88 (90.7)||.014|
|pN1 and pN2||5 (2.3)||9 (9.3)|
|pM0||203 (94.4)||79 (81.4)||<.001|
|pM1||12 (5.6)||18 (18.6)|
|2002 TNM stage groupings|
|Stage I||150 (69.8)||33 (34.0)||<.001|
|Stage II||16 (7.4)||5 (5.2)|
|Stage III||34 (15.8)||38 (39.2)|
|Stage IV||15 (7.0)||21 (21.7)|
|Grade 1||36 (16.7)||1 (1.0)||<.001|
|Grade 2||120 (55.8)||26 (26.8)|
|Grade 3||53 (24.7)||43 (44.3)|
|Grade 4||6 (2.8)||27 (27.8)|
|No||183 (85.1)||34 (35.0)||<.001|
|Yes||32 (14.9)||63 (65.0)|
|No||212 (98.6)||81 (83.5)||<.001|
|Yes||3 (1.4)||16 (16.5)|
In a univariate setting, patients who had tumors with high survivin expression (≥2%) were >5 times more likely to die from RCC compared with patients who had tumors with low survivin expression (RR, 5.3; 95%CI, 3.5–7.9; P<.001) (Table 2). Visualization of the survival experience of the 2 groups confirmed the poorer survival for patients who had tumors with high survivin expression and showed that this difference was apparent very early during follow-up (Fig. 3) (P<.001; log-rank test). Cancer-specific survival rates (the number of patients still at risk) at 1 year, 5 years, and 10 years after nephrectomy were 77.8% (71 patients), 43.0% (34 patients), and 36.2% (18 patients), respectively, for patients who had tumors with high survivin expression compared with 96.6% (193 patients), 87.2% (158 patients), and 81.2% (111 patients), respectively, for patients who had tumors with low survivin expression.
|High Survivin Expression (≥2%)||RR (95% CI)*||P|
|Univariate model||5.27 (3.51–7.93)||<.001|
|Age at surgery||5.76 (3.81–8.72)||<.001|
|Symptoms at presentation||5.16 (3.43–7.76)||<.001|
|Constitutional symptoms at presentation||5.02 (3.34–7.56)||<.001|
|ECOG performance status||5.31 (3.53–7.98)||<.001|
|Tumor thrombus||4.20 (2.76–6.39)||<.001|
|Type of surgery||5.01 (3.33–7.55)||<.001|
|Primary tumor size||4.40 (2.88–6.74)||<.001|
|2002 Primary tumor classification||3.59 (2.35–5.49)||<.001|
|Perinephric fat invasion||4.46 (2.94–6.77)||<.001|
|Regional lymph node involvement||5.03 (3.33–7.59)||<.001|
|Distant metastases||3.99 (2.60–6.12)||<.001|
|2002 TNM stage groupings||3.05 (1.98–4.68)||<.001|
|Nuclear grade||2.67 (1.71–4.17)||<.001|
|Tumor necrosis||2.38 (1.52–3.73)||<.001|
|Sarcomatoid differentiation||4.54 (2.97–6.94)||<.001|
|ECOG, TNM stage groupings, nuclear grade†||2.42 (1.54–3.82)||<.001|
|Mayo Clinic SSIGN score‡||1.78 (1.12–2.85)||.015|
|All unique covariates§||1.86 (1.14–3.02)||.013|
Stratified analyses and formal tests did not demonstrate any significant interactions between survivin expression and other features that were important for predicting outcome. Therefore, we evaluated the association of survivin expression with outcome after adjusting for these features and for the Mayo Clinic SSIGN Score (Table 2). In a multivariate setting, survivin expression remained significantly associated with death from RCC after adjusting for ECOG performance status, 2002 TNM stage groupings, and nuclear grade (RR, 2.4; 95%CI, 1.5–3.8; P<.001) and after adjusting for the Mayo Clinic SSIGN Score (RR, 1.8; 95%CI, 1.1–2.9; P = .015). Further adjustment for additional covariates did not alter substantially the risk associated with survivin expression (Table 2).
In the subset of 273 patients who had clinically localized ccRCC at the time of nephrectomy, survivin expression was associated significantly with progression to distant metastases (RR, 3.9; 95%CI, 2.4–6.2; P<.001) (Table 3) and death from RCC (RR, 5.1; 95%CI, 3.1–8.4; P<.001). Multivariate adjustment for ECOG performance status, TNM stage groupings, and nuclear grade attenuated the risk of progression; however, the evidence of a significant association remained (RR, 1.9; 95%CI, 1.1–3.3; P = .015). Visualization of progression-free survival using the Kaplan–Meier method confirmed that high survivin expression was associated with greater risk of progression and also suggested that this increased risk was apparent within 1 year of follow-up (Fig. 4) (P<.001; log-rank test). Progression-free survival rates (the number of patients still at risk) at 1 year, 5 years, and 10 years after nephrectomy were 75.6% (49 patients), 58.8% (28 patients), and 45.9% (12 patients), respectively, for patients who had tumors with high survivin expression compared with 85.3% (177 patients), 86.8% (146 patients), and 81.2% (102 patients), respectively, for patients who had tumors with low survivin expression.
|High Survivin Expression (≥2%)||RR (95% CI)*||P|
|Univariate model||3.85 (2.38–6.22)||<.001|
|Age at surgery||3.91 (2.38–6.42)||<.001|
|Symptoms at presentation||3.73 (2.31–6.04)||<.001|
|Constitutional symptoms at presentation||3.65 (2.26–5.90)||<.001|
|ECOG performance status||3.92 (2.42–6.35)||<.001|
|Tumor thrombus||2.91 (1.78–4.77)||<.001|
|Type of surgery||3.71 (2.29–6.00)||<.001|
|Primary tumor size||3.13 (1.91–5.13)||<.001|
|2002 Primary tumor classification||2.44 (1.49–4.00)||<.001|
|Perinephric fat invasion||3.01 (1.84–4.93)||<.001|
|2002 TNM stage groupings||2.57 (1.57–4.21)||<.001|
|Nuclear grade||1.94 (1.15–3.26)||.013|
|Tumor necrosis||1.89 (1.12–3.21)||.018|
|Sarcomatoid differentiation||3.58 (2.18–5.87)||<.001|
|ECOG, TNM stage groupings, nuclear grade†||1.93 (1.14–3.28)||.015|
Abnormal inhibition of apoptosis during cellular homeostasis is a critical step for the development and progression of cancer. Survivin is an antiapoptotic protein that belongs to the inhibitor of apoptosis protein (IAP) family.20 Although the role of survivin as an inhibitor of apoptosis is established,21, 22 elucidation of the exact mechanism(s) by which this occurs has not been as straightforward. Currently, it is believed that survivin functions as an inhibitor of cellular apoptosis primarily through a preferential blocking of mitochondrial-dependent apoptosis by targeting of caspase 9 and second mitochondria-derived activator of caspases/direct inhibitory apoptotic protein-binding protein with a low isoleletric point. In addition to its well known antiapoptotic activity, it has been shown that survivin plays a critical role in mitosis and microtubule stability.22, 23 It is noteworthy that survivin is undetectable in most normal adult tissues but is expressed at a high level in virtually every human cancer studied to date,21, 22 including RCC.24
Data from other investigators on survivin expression and RCC aggressiveness are limited. Mahotka et al.25 recently evaluated survivin mRNA expression in 57 RCCs (47 were ccRCC) using quantitative RT-PCR and found no association of survivin expression and primary tumor classification (pT1, pT2 vs. pT3; P = .220). However, conclusions from that investigation were limited by the small sample size, the lack of data on protein expression, and the fact that the investigators did not examine the association of survivin expression with patient outcome, which is the most critical endpoint for biomarker analysis. To the best of our knowledge, the current investigation is the first study to demonstrate an independent association between survivin protein expression and cancer-specific survival in patients with ccRCC.
There are limitations to the current investigation that should be mentioned. Our choice of a designation for high survivin expression was based on an analysis of Martingale residuals, which are designed to select optimal cut-off points for categorical analysis. Although this approach is reasonable for exploratory investigations, prospective analyses are warranted now for validation of our results. Other limitations include low generalizability because of an overwhelmingly Caucasian population (97%), the use of a referral-based practice, and assessment of survivin expression using a single detection method. Nevertheless, the larger sample size is an improvement over many similar molecular investigations of ccRCC prognosis that are limited by low statistical power. Additional strengths of the current study include centralized pathology review by a single urologic pathologist, long-term follow-up information, the use of imaging software for survivin expression quantitation, and adjustment for known predictors of ccRCC outcome, including recently published prognostic scoring systems.8, 9
The evidence in support of a role of survivin in determining ccRCC aggressiveness, coupled with the lack of survivin expression in normal adult tissues, makes this pathway particularly attractive for therapeutic intervention. Indeed, the sharp differential expression in cancer versus normal tissues is among the most intriguing features of survivin and sets it apart from other members of the IAP family.21–23 To date, 4 independent strategies for targeting survivin expression in patients with cancer have gained considerable attention in preclinical testing. The first approach relies on the generation of an antigen-specific immune response against survivin-bearing tumor cells for potential cancer vaccination strategies.26 In addition, the antibody response to survivin has the added potential to offer surveillance strategies for patient follow-up.27 Another approach focuses on molecular antagonists, including antisense, ribozymes, small-interfering RNA, and dominant-negative survivin mutants. In studies along those lines, tumor cell apoptosis and suppression of tumor growth in various cancer models was achieved through interference with survivin expression/function, either alone or in combination with other anticancer strategies.28–31 A third strategy involves pharmacologic inhibition of mitotic phosphorylation of survivin on Thr34. Mutation of Thr34 in survivin abolishes an important phosphorylation site for the mitotic kinase p34cdc2.32 This blocking of phosphorylation results in a dominant-negative phenotype with the induction of apoptosis and anticancer activity in vivo.29 Finally, Pisarev et al.33 recently used a group of patients with prostate cancer and healthy volunteers to investigate the possible utility of dendritic cells transduced with the human full-length dominant-negative survivin as a therapeutic treatment for cancer. From their results, the authors concluded that dendritic cells transduced with dominant-negative survivin induced a potent, survivin-specific, cytotoxic T-lymphocyte response that was able to recognize and effectively kill tumor cells. Without question, there already is compelling precedent for the development of effective therapeutic interventions designed specifically to block the cancer-promoting effects of survivin.34 Although independent validation certainly is required, our data provide the first clear indication that such direct targeting of survivin expression has the potential, either alone or in combination with other therapies, to improve outcomes significantly for patients with ccRCC.
The potential for therapeutic implications notwithstanding, future studies examining survivin expression and ccRCC patient outcome are needed to confirm our results, improve on limitations, and address potential mechanistic aspects of this association. For instance, alternative splicing of survivin mRNA gives rise to multiple, distinct protein isoforms (survivin, survivn-2B, survivin ΔEx3, and survivin 3B), and recent reports in the literature suggest that these 4 splice variants have unique functions and subcellular localization patterns.35, 36 To date, there has been no compelling evidence that survivin splicing variants are involved in tumorigenesis. However, there is some preliminary indication that expression of survivin 2B is decreased significantly in more advanced-stage tumors, suggesting a possible unfavorable role of this variant in cancer progression. With specific regard to RCC, Mahotka et al.25 reported that, although the expression levels of survivin and survivin ΔEx3 were not associated with tumor stage among 47 clinical ccRCC samples, survivin 2B expression was decreased significantly in the late-stage ccRCC compared with early-stage and intermediate-stage ccRCC (P = .036). These data suggest that alternative splice variants of survivin may be involved in the fine tuning of survivin actions during ccRCC progression. Therefore, future investigations should be conducted to validate our results in independent populations and to examine the association of the various survivin variants with the outcome of patients with ccRCC.
- 19Modeling survival data: extending the Cox model. In: DietzK, GailM, KrickebergK, TsiatisA, SametJ, eds. Statistics for biology and health, 1st ed. Ann Arbor: Springer-Verlag, 2000: 87–92., .
- 20Inhibitor of apoptosis proteins and their relatives: IAPs and other BIRPs. Genome Biol. 2001; 2: 1–10., , .