Expression and prognostic significance of LIVIN, SURVIVIN and other apoptosis-related genes in the progression of superficial bladder cancer

P. Gazzaniga1, A. Gradilone1, L. Giuliani1, O. Gandini1, I. Silvestri1, I. Nofroni1, G. Saccani2, L. Frati1 and A. M. Aglianò1,+

1 Dipartimento di Medicina Sperimentale e Patologia, Università degli Studi di Roma ‘La Sapienza’, Rome; 2 Dipartimento di Patologia e Medicina di Laboratorio, Università degli Studi di Parma, Parma, Italy

Received 28 March 2002; revised 27 May 2002. accepted 8 July 2002


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

It has been suggested that progression of superficial bladder cancer may be regulated at the molecular level by a typical pattern of expression of genes involved in apoptosis. Recently LIVIN, belonging to the inhibitors of apoptosis (IAP) family, has been found to be expressed in most solid tumors, where its expression is suggested to have prognostic significance. No data are available concerning the significance of LIVIN in the progression of bladder tumors.

Patients and methods:

In the present paper we used RT–PCR to investigate the expression of LIVIN isoforms {alpha} and ß, SURVIVIN, BCL-X and BCL-2/BAX expression ratio both in normal and tumoral bladder tissues, and correlated their expression with the emergence of early relapses in a follow-up of 4 years. This study shows that only the {alpha} isoform of LIVIN, which is not expressed in normal bladder tissue, is expressed in a proportion of tumors with a high risk of relapse.

Results:

LIVIN was found in 7/30 patients (23%), SURVIVIN in 9/30 (30%), BCL-2/BAX ratio >1 in 16/30 (53%), BCL-2/BAX expression ratio <1 in 14/30 (46.6%) and BCL-X, only in isoform BCL-XL, in 11/30 (36.6%). When we evaluated the dependence between each gene expression and relapse free time of patients, we found that LIVIN, high BCL-2/BAX ratio and BCL-XL, but not SURVIVIN, reached statistical significance in order to predict relapses.

Conclusions:

Our findings suggest that LIVIN may be involved in the progression of superficial bladder cancer and used as a marker of early recurrence; while the expression of SURVIVIN cannot be used to identify patients with high risk of relapse.

Key words: BCL-2/BAX, BCL-X, bladder cancer, LIVIN, relapse, SURVIVIN


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
It has been claimed that superficial tumors of the urinary bladder represent a disease with variable clinical behavior, which shows a clear tendency to early relapse in almost 60% of patients, independent of clinical prognostic variables [1]. An increasing number of studies have focused on the identification of urinary and circulating tumor markers that may represent an adjunct to traditional diagnostic techniques [2]. It has been also suggested that these markers may provide information allowing the appropriate selection of candidate patients for adjuvant therapy. The progression of superficial bladder cancer is known to be regulated at the molecular level by the amplification or rearrangement of some genes; in this context c-erbB1, c-erbB2, p53 and c-myc are thought to be involved [3]. Nevertheless, these markers seem unable to predict the outcome of disease. In recent years, the hypothesis that altered pathways of cell death may contribute to the progression of disease has been put forward. After the first report on BCL-2 gene expression in a proportion of bladder tumors [4], the expression ratio between BCL-2 and BAX, the first inhibiting and the second inducing apoptotic cell death, has been found to determine the potential of cancer progression and in part predict the clinical progression of superficial disease [5]. Recently, novel proteins which suppress apoptosis through caspase-dependent and caspase-independent mechanisms have been characterized, and named inhibitors of apoptosis (IAPs). In humans, six members of the IAP family have been described: HIAP1, HIAP2, XIAP, NIAP, SURVIVIN and, more recently, LIVIN. Two splicing variants have been described for LIVIN, which are almost identical except for a 54 bp truncation in exon 6. Despite their similarity, the two isoforms, termed {alpha} and ß, have been shown to have different antiapoptotic properties in vitro [6]. In fact, while both the {alpha} and ß isoforms seem to block apoptosis induced by TNF-{alpha} and anti-CD45 antibody, LIVIN ß, but not {alpha}, has a strong proapoptotic effect after treatment with etoposide.

Detection of LIVIN ß in fetal tissues, and in placenta in particular, seems to indicate that it may play a role during fetal development. In adult tissues, high expression of both LIVIN {alpha} and ß were detected in heart, placenta, spleen, ovary and lung, while low levels of just LIVIN {alpha} were found in lymphocytes, brain and skeletal muscle. Furthermore, high levels of both LIVIN {alpha} and ß were detected in melanoma, colon cancer and prostate cancer cell lines. In addition, high levels of just isoform {alpha} were detected in some melanoma and lymphoma cell lines. Although extensive studies have been performed concerning the expression of LIVIN in most human tissues, no data were available in the literature concerning the expression of LIVIN isoforms in bladder tissues and in bladder cancer.

Within the IAP family SURVIVIN has also been recently described. Despite its limited expression in normal tissues, SURVIVIN seems overexpressed in a variety of human tumors, including breast, colon, pancreas and prostate carcinoma, neuroblastoma, melanoma and non- Hodgkin’s lymphoma [7]. Studies performed by immunohistochemistry described presence of SURVIVIN in a variable percentage of tumors, ranging from 30% of gastric cancers to 90% of melanomas [8, 9]. Most of these studies found a positive correlation between SURVIVIN expression and prognosis of disease, which is more evident in neuroblastoma and in colorectal cancer, where a multivariate statistical analysis revealed that SURVIVIN expression is an independent prognostic factor for disease progression [10].

Expression of SURVIVIN has been evaluated in bladder cancer by immunohistochemistry and found to be associated with short disease-free interval [11]. Furthermore, SURVIVIN has been detected in the urine of patients with bladder cancer, leading to the hypothesis that it could be used as a marker of early diagnosis [12].

The first goal of this study was to evaluate the expression of LIVIN isoforms in 24 normal bladder tissues as well as in 30 primary superficial bladder cancer specimens. The same tissue samples were also analyzed for the presence of SURVIVIN, BCL-2, BAX and BCL-X mRNA. Furthermore, we evaluated dependence between expression of all these genes and relapse-free time of patients in 4 years of follow-up.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients and tumoral samples
This study included a total of 30 patients affected by primary superficial transitional cell carcinoma of the bladder, who underwent transurethral resection (TUR-B) between January 1996 and January 1998.

The post-surgical pathological stage of each tumor was classified according to the revised tumor–node–metastasis (TNM) staging system. Tumor extension limited to the mucosa (pTa), or the lamina propria (pT1) of the bladder wall was defined as superficial.

Tumor specimens were immediately frozen in liquid nitrogen after surgery and stored at –80°C until use. Informed consent was obtained from all patients.

All patients were subjected to periodical follow-up studies. Patients at high risk for relapse, with multiple lesions or high tumor grade, were treated by TUR-B followed by mitomycin C or BCG. The schedule of the treatment post-TUR was one instillation/week for 8 weeks followed by one instillation/month.

These patients were followed every 3 months with urinary cytology and pelvic ultrasonography for a period of 1 year. If no recurrence was observed during this period, patients were then followed every 4 months for the second year, and every 6 months thereafter.

The mean follow-up period for the study was 39 months (27–51 months). Clinical features of patients are summarized in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Clinical, pathological and molecular features of superficial bladder cancer patients
 
In addition, since no data were available concerning LIVIN expression in bladder tissue, we evaluated the presence of LIVIN isoforms {alpha} and ß in 24 normal bladder tissues, 18 adjacent to the tumor and six taken from autoptic procedures. The same samples were used as controls for the expression of BCL-2, BAX, BCL-X and SURVIVIN.

RT–PCR
Total RNA (1 µg) extracted from frozen tissues was reverse transcribed in a final volume of 20 µl with 100 pmol of random examer and 50 U MuLV reverse transcriptase (Perkin Elmer Cetus, Norwalk, CT, USA), according to the manufacturer’s guidelines.

Aliquots corresponding to 100 ng RNA were then amplified in PCR buffer containing 25 pmol each primer and 1 U Taq polymerase in a final volume of 50 µl.

Aliquots of the same cDNA were amplified with ß-actin, LIVIN, SURVIVIN, BCL-2, BAX and BCL-X primers. Each amplification, except for LIVIN, was performed for 30 cycles; a cycle profile consisted of denaturation at 94°C for 30 s, annealing at 60°C (62°C for BAX and BCL-X) for 30 s and extension at 72°C for 30 s. A sample without RNA was included in each RT–PCR as a negative control; for positive controls, RNA extracted from M14 cell line for SURVIVIN, HeLa cell line for BCL-2 and CaSki cell line for BAX and BCL-X were used. LIVIN primer sequences and amplification conditions were as described [6].

Sequences of the other primers used are as follows: ß-actin upstream, 5'-TTAGCTGTGCTCGCGCTACTCTCTC-3'; ß-actin downstream, 5'-GTCGGATTGATGAAACCCAGACACA-3'; SURVIVIN upstream, 5'-CAGATTTGAATCGCGGGACCC-3'; SURVIVIN downstream, 5'- CCAAGTCTGGCTCGTTCTCAG-3'; BCL-2 upstream, 5'-GTGGAGGAGCTCTTCAGGGA-3'; BCL-2 downstream, 5'-AGGCACCCAGGGTGATGCAA-3'; BAX upstream, 5'-GGCCCACCAGCTCTGAGCAGA-3'; BAX downstream, 5'-GCCACGTGGGCGGTCCCAAAGT-3'; BCL-X upstream, 5'-TTGGACAATGGACTGGTTGA-3'; BCL-X downstream, 5'-GTAGAGTGGATGGTCAGTG-3'. The primers for BCL-X were designed to recognize both isoforms, BCL-XL and BCL-XS.

The size of the amplified products were 168, 206, 304, 479, 780 and 591 bp for ß-actin, SURVIVIN, BCL-2, BAX, BCL-XL and BCL-XS, respectively. The size of LIVIN isoforms {alpha} and ß were 368 and 314 bp, respectively.

We then performed a semi-quantitative analysis of BCL-2/BAX ratio in each sample, as described [5]. In order to verify the specificity of the amplified products, we performed a hybridization with specific oligonucleotide probes, as previously described [13].

Statistical analysis
Statistical analysis was performed using BMDP statistical software, version 7 [14].

Time to first recurrence was calculated from the date of initial surgery and used as the endpoint of the study. Relapse-free time was estimated using the product-limit method of Kaplan–Meier; the difference between the curves was evaluated using the log-rank test (Mantel–Cox method). Chi-square test was used in order to assess the association between expression of LIVIN, SURVIVIN, BCL-X and BCL-2/BAX ratio and clinical prognostic variables, such as stage, grade, age and multicentricity of tumors. A value of P <0.05 was considered statistically significant.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
In the present study for the first time we evaluated by RT–PCR the presence of LIVIN expression in 24 normal and in 30 cancerous bladder tissues. In normal tissues, no detectable levels of either mRNA isoforms were detected. Among tumor tissues, 7/30 (23%) showed expression of isoform {alpha}, while none had detectable isoform ß expression (Figure 1). LIVIN expression does not correlate with any of the known prognostic variables, such as stage, grade and multicentricity.



View larger version (23K):
[in this window]
[in a new window]
 
Figure 1. Expression of LIVIN isoforms {alpha} and ß in neoplastic and normal bladder tissues. (A) Lanes 1–7, bladder cancer specimens; lane 8, positive control (M14 cell line); (B) lanes 1–7, normal bladder tissues; lane 8, positive control (M14 cell line).

 
Moreover, in the same patients SURVIVIN, BCL-2, BAX and BCL-X expression was analyzed. Because the experimental evidence suggests that the balance between antiapoptotic and proapoptotic members of the BCL-2 family is a much better determinant of the sensitivity to apoptosis, we evaluated the BCL-2/BAX expression ratio, instead of the expression of individual members. In normal tissues, while BCL-2 and SURVIVIN were not found to be expressed, BAX was present in 100% and BCL-X in 37% of samples (data not shown).

In tumor samples the presence of SURVIVIN expression was found in 9/30 (30%) patients, BCL-2/BAX ratio >1 in 16/30 (53%) and BCL-X expression in 11/30 (36%) (Table 1). BCL-X gene was found to be expressed only in the BCL-XL isoform (Figure 2).



View larger version (44K):
[in this window]
[in a new window]
 
Figure 2. Expression of SURVIVIN, BCL-2, BAX and BCL-X in superficial bladder cancer. Agarose gel electrophoresis of RT–PCR products: lanes 1–5, bladder cancer specimens. Positive control: RNA from M14 cell line for ß-actin and SURVIVIN; RNA from HeLa cells for BCL-2; RNA from Caski cell line for BAX and BCL-X amplification. Negative control: sample with all the reagents except template.

 
When we evaluated the dependence between SURVIVIN expression and known prognostic variables, we found that SURVIVIN expression segregates well with histological grade (5% in G1 versus 89% in G2; P <0.05), but not to clinical stage or multicentricity of the tumors. On the contrary, our analysis showed that higher BCL-2 levels segregates well only with multicentricity (92% in multicentric forms versus 18% in solitary; P <0.05), but not with clinical stage and histological grade of the tumors. BCL-XL expression segregates well with multicentricity, but not with stage or grade of the tumors (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Dependence between clinicopathological features and gene expression (chi-square test)
 
When we evaluated dependence between each gene expression and relapse-free time of patients in the first 4 years of follow-up (mean follow-up period 39 months), we found that LIVIN expression, high BCL-2/BAX expression ratio and BCL-XL, but not SURVIVIN expression, reached statistical significance in the prediction of relapses. For instance, median relapse-free time of LIVIN-positive patients was 3.5 versus 27.2 months in LIVIN-negative patients (P <0.0001). When BCL-2/BAX ratio was taken into account, patients with a high BCL-2/BAX ratio had significantly shorter recurrence-free intervals than did patients with a low BCL-2/BAX ratio. In fact, mean period of relapse-free time of patients with a high BCL-2/BAX expression ratio was 3 months compared with 48 months for those with a BCL-2/BAX expression ratio <1 (P <0.0001). In patients with a BCL-2/BAX expression ratio <1 the percentage of patients relapse-free was 100%, compared to 0% observed in those with BCL-2/BAX ratio >1. Moreover, the outcome in patients with BCL-XL expression was worse compared with those with no BCL-XL expression. Regarding BCL-XL, the percentage of relapse-free patients was 0% in the group with BCL-XL expression, compared with 73.7% observed in the group with no BCL-XL expression (P <0.0005). In contrast, SURVIVIN expression does not correlate with the emergence of relapses (P >0.0005). In patients who relapsed during treatment (mitomycin C or BCG) we found 100% were positive for BCL-2, 66% for BCL-X, 33% for LIVIN, and 16% for SURVIVIN. Results of the survival distribution evaluated by the Kaplan–Meier method are summarized in Table 3.


View this table:
[in this window]
[in a new window]
 
Table 3. Estimation of relapse-free time by the Kaplan–Meier methoda
 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The main purpose of this study was to investigate the expression of some genes involved in apoptotic pathways in neoplastic tissues of patients affected by superficial bladder cancer, and to correlate their expression to the emergence of relapses. The most interesting data was the finding that LIVIN, a new member of the IAP family, was found to be not expressed in any of the normal bladder tissues, and expressed only as the isoform {alpha} in a proportion of superficial bladder cancer patients (23%). LIVIN has been described in two isoforms, {alpha} and ß, showing different antiapoptotic properties in vitro. In fact, while both isoforms are involved in blocking apoptosis induced by TNF-{alpha} and anti-CD95, LIVIN ß seems to be more effective than LIVIN {alpha} in blocking apoptosis induced by DNA damaging agents, such as etoposide [6].

Tissue distribution of LIVIN has been recently described: elevated levels of both LIVIN isoforms {alpha} and ß have been detected in heart, placenta, lung, spleen and ovary, while LIVIN ß alone has been detected specifically in fetal tissues. Adult kidney seems to express only isoform ß. LIVIN {alpha} alone has been detected in brain, skeletal muscle and peripheral blood lymphocytes.

Since no data were available concerning LIVIN expression in bladder, our study demonstrates for the first time that expression of both LIVIN isoforms {alpha} and ß are not detectable in normal bladder tissues, as previously described for BCL-2, a gene with antiapoptotic properties.

Furthermore, while LIVIN expression was detected in a variety of cancerous cell lines, no data were available concerning the expression of LIVIN isoforms in tumor tissues. Our study for the first time demonstrates that LIVIN isoform {alpha} is expressed in a proportion of superficial bladder tumors, all of which had a clear tendency to early relapse, showing an unexpected short relapse-free time. Data from the literature have demonstrated that, whether both LIVIN isoforms are involved in blocking apoptosis induced by anti-CD95 and TNF-{alpha}, isoform {alpha} is less efficient than isoform ß in blocking apoptosis induced by DNA damaging agents. Nevertheless, it is conceivable that in bladder tumors the expression of LIVIN isoform {alpha}, together with an overexpression of BCL-2 upon BAX, may be sufficient to give cells a strong resistance to chemotherapy-induced apoptosis.

In the same group of tumor samples we investigated the presence of SURVIVIN, another member of the IAP family, which other authors described as being absent in normal bladder tissue and expressed in bladder tumors, and other genes known to be involved in the pathogenesis of bladder cancer, specifically BCL-2, BAX and BCL-X. LIVIN expression does not seem to be correlated with known prognostic variables, such as grade, stage and multicentricity; furthermore, no significant association was found between LIVIN and SURVIVIN expression, or between LIVIN and BCL-2 or BCL-X expression. These results may reflect the different transcriptional pathways of these genes.

Follow-up data obtained in the first 4 years after surgery revealed that patients whose tumors showed expression of LIVIN isoform {alpha} had a very short relapse-free time (3.5 months), similar to that observed in patients whose tumors expressed a high BCL-2/BAX ratio. In all patients who relapsed during the course of treatment we found a high BCL-2/BAX expression ratio. This further stresses the role of high BCL-2 expression as a mechanism by which tumor cells escape chemotherapy-induced apoptosis.

On the contrary, SURVIVIN expression in our samples does not correlate with the emergence of early relapse, although its expression seems to be associated with histological grade of tumors.

The only data available concerning SURVIVIN in bladder tumors seem to indicate that SURVIVIN, but not BCL-2, detected by immunohistochemical methods, is significantly associated with time of recurrence and histological grade [11]. We cannot really explain the discrepancy between the data of Swana et al. [11] and ours, mainly due to the lack of sufficient clinical data in that series of patients (size of tumors, modalities of treatment after surgery). Moreover, it is known that BCL-2 alone in bladder cancer, as well as in other solid malignancies, is not sufficient to predict disease recurrence. Thus, the lack of association between BCL-2 expression and poor prognosis in that series of patients may be due to altered expression levels of other family members, like BAX, whose presence was not investigated by Swana et al. [11].

The emergence of early relapses in patients with high levels of BCL-2 expression may also depend on the known ability of BCL-2 to inhibit chemotherapy-induced apoptosis, as previously described in bladder, lung and breast cancer [15–17]. It is thus conceivable that SURVIVIN does not share with BCL-2 the ability to block mitomycin-C-induced apoptosis. In fact, few data are available concerning the efficacy of SURVIVIN in blocking chemotherapy-induced apoptosis. To date, SURVIVIN has been shown to inhibit apoptosis of NIH3T3 transfectants induced by paclitaxel, and apoptosis of 293 cell line induced by etoposide, but it seems ineffective against microtubule depolymerizing agents such as vincristine [18]. Due to the small number of patients whose results were positive for LIVIN, it is still difficult to establish whether LIVIN expression may interfere with chemotherapy-induced apoptosis.

Unlike other solid malignancies, such as breast, gastric and colorectal cancer, the presence of SURVIVIN in our series of bladder tumor specimens does not seem to be associated with high levels of BCL-2. This seems unsurprising, due to the fact that SURVIVIN and BCL-2 do not completely share common mechanisms of transcriptional activation [6]. In addition, similar to what has been observed in other tissues, no correlation was found between LIVIN isoform {alpha} and SURVIVIN expression in bladder cancer.

The emergence of local relapses in superficial bladder cancer represents one of the major problems in the clinical management of this tumor. In fact, it is widely acknowledged that stage and grade are often unable to predict the local progression of Ta–T1 low grade bladder tumors [1], and that 50% of patients affected by superficial disease have local recurrence in the first year of follow-up. This is the reason why in 1996 the European Organization for Research and Treatment of Cancer and Medical Research Council defined the utility of prophylactic treatment also in stage Ta–T1–G1 bladder cancer patients, since it gives a clear advantage in terms of duration of disease-free interval [19]. In view of this, the search for a panel of molecular markers could be useful in the identification of patients with a higher risk of relapse. In our series of patients, the combination of a high BCL-2/BAX ratio, high BCL-X expression and LIVIN isoform {alpha} seems to identify a subset of patients with shorter relapse-free time.

Our findings, which suggest for the first time a role of LIVIN isoform {alpha} in the progression of superficial bladder cancer, also indicate that the expression of SURVIVIN cannot be used to identify patients with a higher risk of early relapse.


    Acknowledgements
 
This work was partially supported by MURST (ministero dell’Università e della Ricerca Scientifica) and AIRC (Associazione Italiana Ricerca sul Cancro).


    Footnotes
 
+ Correspondence to: Professor Anna Maria Aglianò, Dipartimento di Medicina Sperimentale e Patologia, Viale Regina Elena, 324-00161 Rome, Italy. Tel: +39-06-49973011; Fax: +39-06-4454820; E-mail: annamaria.agliano{at}uniroma1.it Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Ozen H, Hall MC. Bladder cancer. Curr Opin Oncol 2000; 12: 255–259.[CrossRef][ISI][Medline]

2. Stein JP, Grossfeld GD, Ginsberg DA et al. Prognostic markers in bladder cancer: a contemporary review of the literature. J Urol 1998; 160: 645–659.[ISI][Medline]

3. Lee E, Park I, Lee C. Prognostic markers of intravesical bacillus Calmette–Guerìn therapy for multiple, high grade, stage T1 bladder cancer. Int J Urol 1997; 4: 552–556.[Medline]

4. Gazzaniga P, Gallucci M, Gradilone A et al. Detection of BCL-2 RNA in low grade tumours of the urinary bladder. Eur J Cancer 1995; 131A: 2119–2120.[CrossRef]

5. Gazzaniga P, Gradilone A, Vercillo R et al. BCL-2/BAX mRNA expression ratio as prognostic factor in low grade urinary bladder cancer. Int J Cancer 1996; 69: 100–104.[CrossRef][ISI][Medline]

6. Ashhab Y, Alian A, Polliack A et al. Two splicing variants of a new inhibitor of apoptosis gene with different biological properties and tissue distribution pattern. FEBS Lett 2001; 495: 56–60.[CrossRef][ISI][Medline]

7. Ambrosini G, Adida C, Altieri DC. A novel anti-apoptosis gene, SURVIVIN, expressed in cancer and lymphoma. Nature Med 1997; 3: 917–921.[ISI][Medline]

8. Lu CD, Altieri DC, Tanigawa N. Expression of a novel antiapoptosis gene, SURVIVIN, correlates with tumor cell apoptosis and p53 accumulation in gastric carcinoma. Cancer Res 1998; 58; 1808–1812.[Abstract]

9. Grossman D, McNiff JM, Li F, Altieri DC. Expression and targeting of the apoptosis inhibitor SURVIVIN in human melanoma. J Invest Dermatol 1999; 113: 1076–1081.[Abstract/Free Full Text]

10. Kawasaki H, Altieri DC, Lu CD et al. Inhibition of apoptosis by SURVIVIN predicts shorter survival rates in colorectal cancer. Cancer Res 1998; 58: 5071–5074.[Abstract]

11. Swana HS, Grossman D, Anthony JN et al. Tumor content of the antiapoptotic molecule SURVIVIN and recurrence of bladder cancer. N Engl J Med 1999; 341: 452–453.[Free Full Text]

12. Smith SD, Wheeler MA, Plescia J et al. Urine detection of SURVIVIN and diagnosis of bladder cancer. JAMA 2001; 285: 324–328.[Abstract/Free Full Text]

13. Gazzaniga P, Gradilone A, Silvestri I et al. Variable levels of BCL-2, BCL-X and BAX mRNA in bladder cancer progression. Oncol Rep 1998; 5: 901–904.[ISI][Medline]

14. Pompeo E, Sergiacomi G, Nofroni I et al. Morphologic grading of emphysema is useful in the selection of candidates for unilateral or bilateral reduction pneumoplasty. Eur J Card Thor Surg 2000; 17: 680–686.

15. Miyake H, Hara I, Yamanaka K et al. Synergistic enhancement of resistance to cisplatin in human bladder cancer cells by overexpression of mutant type p53 and BCL-2. J Urol 1999; 162: 2167–2181.

16. Zhang Y, Fujita N, Tsuruo T. p21Waf1/Cip1 acts in synergy with BCL-2 to confer multidrug resistance in a camptothecin-selected human lung cancer cell line. Int J Cancer 1999; 83: 790–797.[CrossRef][ISI][Medline]

17. Yang OF, Sakurai T, Yashimura G et al. Expression of BCL-2 but not BAX or p53 correlates with in vitro resistance to a series of anticancer drugs in breast carcinoma. Breast Cancer Res Treat 2000; 61: 211–216.[CrossRef][ISI][Medline]

18. Buolamwini JK. Novel anticancer drug discovery. Curr Opin Chem Biol 1999; 3: 500–509.[CrossRef][ISI][Medline]

19. Pawinski A, Sylvester R, Kurth KH et al. A combined analysis of European Organization for Research and Treatment of Cancer and Medical Research Council randomized clinical trials for the prophylactic treatment of stage Ta T1 bladder cancer. European Organization for Research and Treatment of Cancer Genitourinary Tract Cancer Cooperative Group and the Medical Research Council Working Party on Superficial Bladder Cancer. J Urol 1996; 156: 1934–1940.[ISI][Medline]