Intravesical gemcitabine versus mitomycin for non-muscle invasive bladder cancer: a systematic review and meta-analysis of randomized controlled trial

Bladder cancer has become a common cancer worldwide, and 430,000 new cases and over 165,000 deaths are reported in 2012 [1]. Transitional cell carcinoma is dominant in bladder cancer, in addition, adenocarcinoma (cancer originating from the mucus-making and releasing cells) and squamous cell carcinoma (cancer originating from the thin and flat cells) are also observed [2]. Generally, bladder cancer is associated with the following symptoms, frequent urination, bloody urine (rendering the slight rusty to deep red color of urine), urination pain, or urination sensation with no urine indeed [3].

The incidences of bladder cancer in male and female are reported to be 3.4 and 1.2%, respectively, however, that in patients aged over 70 years is twice to thrice higher than that among patients aged 55–65 years, and 15 to 20 times greater than for patients aged 30–54 years [4]. As estimated by the World Health Organization, there were 132,432 bladder cancer-related deaths in the world in 2000 [5].

About 80% bladder cancers are non-muscle invasive bladder cancers (NMIBC) restricted to the urothelium (clinical stage Ta) or lamina propria (stage T1) at first, which is featured by the in-situ flat carcinoma (stage Tis) [6]. Tumor relapse following transurethral resection (TUR) has been identified as a main issue in treating non-muscle invasive bladder cancer. Specifically, the mechanisms regarding NMIBC relapse following TUR are shown as follows, (1) residual tumor originated from the incomplete resection; (2) floating cancer cell implantation in traumatized bladder sites; (3) incidence of new neoplasm due to high cancer aggressiveness; (4) relapses at the urothelial instability sites (atypia, hyperplasia and dysplasia) [7]; (5) over 50–70% tumor relapse since no adjuvant treatment is given, and approximately 15% cases develop muscle-invasive cancers eventually [8].

The intravesical chemotherapy has been considered as the standard treatment for patients receiving following TUR to eradicate underlying disorders, inhibit cancer relapse, prevent tumor development and extend patient survival [9]. Mitomycin (MMC) is a frequently used compound in intravesical treatment. In addition, gemcitabine (GEM), the relatively new pyrimidine analog, displays anticancer effect on various solid cancers, such as the advanced bladder cancer [10]. MMC and GEM have been classically used as the cytotoxic agents for affecting the DNA integrity, but these two display distinctly different mechanisms of action. Among them, GEM, the difluoro-2, 2-deoxy-cytidine, can be activated upon the stimulation of deoxycytidine kinase, meanwhile, its phosphorylated metabolites may impact the synthesis of deoxy-nucleotide through resulting in DNA injuries and interfering with DNA repair [11]. On the other hand, MMC can be triggered within cancer cells through forming the reducing equivalents, thereby affecting cancer cell replication through forming the damage-induced DNA adducts [12]. Nonetheless, it remains unclear about the efficacy and side effects between the two agents. In this regard, the current systematic review was carried out aiming to evaluate the therapeutic effects and safety between GEM versus MMC on treating NMIBC patients.

Meta-analysis is an approach adopted to statistically pooled and examined the findings from several independent RCTs [21]. NMIBC is generally associated with favorable clinical outcomes, but it is refractory due to the increased tumor relapse rate following TUR. As a matter of fact, 40–85% cases develop tumor relapse in 2–5 years following standardized TUR, while approximately 10% among such relapse patients are at the more advanced grade and stage [22]. In clinical practice, TUR is performed prior to intravesical injection with chemotherapeutic or the agents for modulating immunity, so as to lower the bladder cancer relapse and development.

Various anti-cancer agents are administered intravesically to prevent tumor relapse, which is validated as effective. The classical intravesical agents, Bacille Calmette-Guérin (BCG) as well as MMC are effective on postponing and lowering the postoperative tumor relapse rate; nonetheless, they are also associated with obvious adverse reactions and may be ineffective for some cases. Based on clinical research, GEM is effective on reducing the relapse of non-muscle invasive bladder cancer with lower toxicity, which deserves more investigation [23]. In this study, the meta-analysis was carried out to examine the intravesical GEM and MMC safety and efficacy among the NMIBC patients on the basis of the five enrolled clinical trials including 335 bladder cancer cases. Three of those 5 articles mentioned disease grade and stage [17, 19, 20]. Our results showed that the recurrence rate of MMC group increased compared with that of GEM group (OR = 0.44 95% CI [0.24, 0.78]), and the difference was statistically significant between the two groups; besides, GEM led to a lower incidence of chemical cystitis than MMC (OR = 0.23 95% CI [0.12, 0.44]). But differences in the incidences of hematuria (OR = 0.46 95% CI [0.16, 1.31]), skin reaction (OR = 0.49 95% CI [0.14, 1.70]) and liver and kidney function damage (OR = 0.51 95% CI [0.09, 2.85]) were not statistically significant between the two groups., suggesting that GEM had lower toxicity than MMC. Overall, these two doses were well-tolerable among a majority of cases with lower adverse reactions. Nonetheless, the above conclusions must be interpreted carefully and validated in more articles.

For the low risk disorder, post-operative intravesical MMC treatment is still the vital part of intravesical therapy, meanwhile, there is level-one evidence to prove the effectiveness of GEM on bladder cancer, yet it has not been recommended in guidelines [24]. Many therapeutic methods are effective on the treatment of NMIBC cases. The intravesical chemotherapy using MMC and GEM has achieved certain favorable results [25]. Gemcitabine is an effective pyrimidine antimetabolite, with only mild toxicity relative to other chemotherapeutic agents [26]. According to Messing EM et al. [27], for the low-grade NMIBC cases (n = 215), those receiving intravesical GEM treatment following TUR were associated with the 4-year tumor relapse rate of 34%, while that in the saline-alone control group was 54%. In addition, 5 cases of GEM group and 10 in control group progressed to muscle invasive bladder cancer, and 17 of GEM group as well as 25 of saline group died. Based on Dalbagni et al., for the 30 BCG-refractory NMIBC cases (including uncontrollable Ta, multiple unresected T1, and refractory cancer in situ) who underwent intravesical GEM treatment regularly, 50% achieved a complete response (CR), and adverse effects were reported among 23% cases with a very low toxicity [28]. Moreover, according to Bartoletti et al., no 1-year tumor relapse was reported among 18 out of the 24 medium-risk patients, as well as among 9 out of the 16 high-risk BCG-refractory cases (pTa ~ pT1) treated with weekly intravesical GEM therapy. In that study, no side effect was reported in 94 of the 116 patients in the 1-year treatment process [29]. Jones et al. reviewed 6 GEM treatment-related RCTs in their Cochrane review, and indicated that, intravesical GEM treatment potentially exerted an important part in managing NMIBC cases with moderate and high risk, particularly when it was adopted to be the alternative to MMC for high-risk cases [30].

Auxiliary intravesical treatment can lower the relapse and development rates of NMIBC in the meantime of inducing local and systemic adverse reactions. In terms of the toxic effects, chemical cystitis, hematuria, skin reaction, and liver and kidney function damage were observed in 0–30% cases of each group. The frequency of these side effects of GEM arm markedly decreased compared with that of MMC arm. Such findings validated that, chemical cystitis, the irritable symptom in lower urinary tract, stands for the typical MMC side reaction [31]. GEM led to minimal local toxicity, which was self-resolved rapidly. By contrast, MMC treatment resulted in the more serious local toxicity that required treatment delay. On the other hand, economic influence is also a factor potentially affecting the treatment decision-making. Recently, the mean sales price for 40 mg MMC markedly increases compared with that for 2 g GEM, suggesting that the GEM-based intravesical chemotherapy is advantageous over MMC in price [32].

Some limitations should be noted in the current meta-analysis: 1) the risk of tumor relapse was not compared between GEM and MMC of Ta or T1 group due to the limited total sample size, so more high-quality RCTs should be carried out. 2) The administration does and schedules of these two agents were similar among those five enrolled studies, but the conclusion might be disturbed by the clinical heterogeneity because of the differences in operating methods, like whether the positional changes were allowed, as well as the duration of each infusion. 3) All included RCTs in the current review did not apply the double-blinding method, which might potentially impact our results. 4) The publication bias was not totally ruled out, which possibly led to conclusion distortion.

In conclusion, our meta-analysis supports that GEM can be a potential agent of intravesical therapy with better efficiency than MMC in preventing recurrence of patients with non-muscle invasive bladder cancer after TUR. At the same time, we observed that GEM produce lower local toxic effects than MMC during intravesical therapy. In addition, more studies are needed to examine the GEM safety, and more high-quality RCTs are warranted to validate our conclusions due to the limitations.