Influence of Antibiotic Administration on the Urinary Bladder Cancer Early Recurrence Rate

Bladder cancer tends to recur, making treatment one of the most expensive in oncology. The limited efficacy and high cost of adjuvant therapies in the treatment of bladder cancer prompt research on new drugs which could replace them. In vitro studies have established that antibiotics can have a cytostatic and cytotoxic effect on urinary bladder cancer cells. The objective of the study was to investigate the influence of antibiotics on the recurrence rate of bladder cancer. In a retrospective study, we analyzed a group of 199 patients with urinary bladder cancer from four urological centers. The study groups consisted of 40 patients who received ciprofloxacin and 83 patients who received beta-lactams as perioperative antimicrobial prophylaxis. The control group included 76 patients who did not get perioperative antimicrobial prophylaxis. The groups were analyzed for risk stratification, degree of malignancy, and size of the primary tumor. The average follow-up time was 24 months. The main focus of the study was to investigate the early recurrence rate of bladder cancer among studied groups, which could correlate with the effectiveness of currently used intravesical instillations. Additionally, cancer's early progression was examined. Regardless of the division used, the highest recurrence rate was found in the ciprofloxacin group. There were no statistical differences in the recurrence rate between patients who received beta-lactams and patients who did not receive any antibiotics. In addition, there were no differences due to the progression rate between the groups. Perioperative antibiotic administration does not influence the early recurrence rate in patients with nonmuscle invasive urothelial bladder cancer.


Introduction
Te characteristic feature of bladder cancer is the tendency to recur [1]. Te currently used intravesical adjuvant therapies do not sufciently limit the recurrence rate of urinary bladder cancer [2]. In addition, patients with urinary bladder cancer and co-existing pyuria had a signifcantly higher rate of highgrade tumors [3]. Antibiotic prophylaxis before surgical treatment of bladder cancer lowers the risk of infection complications. However, properly monitoring pathogen resistance and not escalating antibiotic use without indications is crucial to guide empirical antibiotic therapy [4].
On the other hand, it is postulated that antibiotics have other properties (including the efect on cancer cells), apart from antibacterial activity. In vitro studies on cancer cell lines proved the antitumor potential of several antibiotics [5]. In the study, we investigated ciprofoxacin as an antibiotic with the highest antitumor potential in vitro studies [6,7] and betalactams as the most often used antibiotics in perioperative prophylaxis for the treatment of urinary bladder cancer [8].

Aim
Te aim of this study was to evaluate the recurrence rate of nonmuscle invasive bladder cancer in patients undergoing transurethral resection of bladder tumors, depending on the usage of antibiotic prophylaxis. Patients who were given ciprofoxacin were assigned to interventional group number 1, while those obtaining beta-lactams were assigned to interventional group number 2 and patients without antibiotics were assigned to the control group.

Materials and Methods.
In the retrospect, the study patients came from four urological centers. After application of the inclusion and exclusion criteria, 199 patients treated in the years 2009-2016 were qualifed for the study (86 patients from Municipal Hospital in Torun, 61 patients from Oncological Center in Bydgoszcz, 36 patients from 10 th Military Hospital in Bydgoszcz, and 16 patients from Voivodeship Hospital in Zgierz). Patients were diagnosed with cystoscopy and primary nonmuscle invasive urothelial bladder cancer was confrmed by transurethral resection of bladder tumor (TURBT) (Figure 1).
Te study groups consisted of 40 patients who received ciprofoxacin and 83 patients who received beta-lactams as perioperative antimicrobial prophylaxis (Table 1). Te betalactams group included patients who received amoxicillin with clavulanic acid, cefuroxime, or cephazolin (Table 1). All antibiotics were given intravenously, frst dose 20-30 minutes before TURBT and the subsequent doses after 8-12 hours if there was a high risk of infection or UTI before TURBT. Te antibiotic choice was made upon local microbiological recommendations based on previous urine culture results. Te high risk of infection was defned as pyuria before TURBT and large tumor mass, which was related to the longer operation time. All patients received only one of the mentioned antibiotics. Patients with more than one antibiotic during the treatment and receiving antibiotics before TURBT were excluded from the study. Te control group included 76 patients who did not get perioperative antimicrobial prophylaxis. Patients who received intravesical instillations with chemotherapy in any form (single immediate instillation or maintained scheme) were excluded. Te lack of intravesical treatment in Polish urological departments relates to administrative problems such as the unavailability of drugs and the lack of adequate infrastructure for the administration of chemotherapy. Patients with recurrence in the frst control cystoscopy (3 months after TURBT) were excluded because it could indicate incomplete primary resection.
For a better investigation of the efects of antibiotics on urinary bladder cancer, additional divisions of the study group were introduced. According to EAU recommendations, groups were analyzed for risk stratifcation, which corresponded to the course of cancer [28] (Figure 2(a)). Other divisions referred to the infuence of antibiotics on the primary tumor. Patients were divided according to the degree of malignancy (Figure 2(b)) and size of the primary tumor (Figure 2(c)). Follow-up was applied according to current EAU recommendations [8]. Te basic follow-up tool was cystoscopy. Te average time of the follow-up was 24 months (Table 2). Every patient underwent minimum 2 control cystoscopies (on average 3 cystoscopies/patient) ( Table 2). Te frst cystoscopy was done three months after TURBT, the second 12 months after TURBT, and then every 6-12 months. Additionally, every patient during follow-up at least once underwent computer tomography of the abdomen and pelvis foor (uroCT protocol), chest X-ray at least twice, and urinary tract ultrasound during every ambulatory visit (on average, three urinary tract ultrasound examinations/patient). Urine cytology was done among patients with high-grade tumors. Detailed group characteristics are presented in Table 3. Patients who were administered ciprofoxacin had slightly more common pT1 high-grade disease, which corresponded with a more aggressive cancer path, while initially, there were no signifcant statical diferences in cancer advancement (primary tumor malignancy) ( Table 3).
Progression was defned as any recurrence with a deeper infltration and/or a higher grade of malignancy. Statistical analyses were carried out using the IBM SPSS Statistics 25 package. It was used to analyze basic descriptive statistics together with the Kolmogorov-Smirnov test, Mann-Whitney U tests, nonparametric analysis of variance Kruskal-Wallis, χ 2 tests, and Fisher's exact tests. Te recurrence rate was analyzed in Kruskal-Wallis variance with post-hoc analysis using the Dunn-Sidak test with the Bonferroni correction for multiple comparisons. Recurrence rates in particular divisions were analyzed in χ 2 and Fisher's exact tests. Te threshold α � 0.05 was considered a statistical signifcance level. Approval of Ethics Committee: Collegium Medicum Ethics Committee in Bydgoszcz Approval No. KB433/2018.

Results
In the general analysis. a statistically signifcant diference was observed in the ciprofoxacin group, patients who received ciprofoxacin experienced more recurrences than patients who did not receive antibiotics (p � 0.02) or who received beta-lactam antibiotics (p � 0.017) (Figure 3(a)). When comparing the group of patients not treated with antibiotics and those treated with beta-lactam antibiotics, no statistically signifcant diferences were found in relation to the frequency of cancer recurrence (Figure 3(a)). Tere were no statistically signifcant diferences in the progression rate.
Assuming that antibiotics like anthracyclines should not infuence progression rate extended analysis in relation to progression was abandoned (Figure 3(b)).
According to the EAU risk stratifcation, the recurrence rate was related to the grouping of the applied breakdown. Tus, an apparent statistical diference was demonstrated, in which the frequency of recurrences increased, being the highest in the high-risk group (Figure 4(a)). Furthermore, statistical analysis of the studied groups showed a statistical tendency (p � 0.083) of more frequent cancer recurrences in the intermediate-risk group among patients who received ciprofoxacin. However, it was not statistically signifcant ( Figure 4(a)).

Division of Study Groups According to Primary Tumor
Malignancy. By assessing the study groups concerning the degree of malignancy of the primary tumor, a statistical trend showed a higher frequency of recurrences in patients with the high-risk primary disease (Figure 4(b)). Patients from the low-risk group who received ciprofoxacin experienced a statistically signifcantly more frequent recurrence than patients who did not receive any antibiotics (p � 0.009). Tere was no stastical diference between patients who received beta-lactams and did not receive any antibiotics (Figure 4(b)).

Division of Study Groups According to Primary Tumor
Size. Te statistical trend of more frequent recurrences with initially large tumor mass was confrmed in this division (Figure 4(c)). It has not been confrmed that the use of any of the antibiotics reduces the frequency of recurrences. In the division by tumor mass, there was again a statistically more frequent recurrence rate among patients treated with   Journal of Oncology ciprofoxacin in comparison to patients who did not receive any antibiotics (p � 0.010) and there was no statistical diference among patients who received beta-lactams as those who did not receive any antibiotics (Figure 4(c)).

Analysis of Patients Included in the Study.
Inclusion and exclusion criteria were used to select patients who could beneft from the use of antibiotics in anticancer therapy. Assuming that antibiotics have similar efcacy to anthracyclines (including doxorubicin), such a group of patients should be patients with low-risk nonmuscle invasive urothelial bladder carcinoma. Patients who received ciprofoxacin and beta-lactams in perioperative prophylaxis were selected for the study. Patients who received ciprofoxacin were selected due to the high antitumor potential of ciprofoxacin proved in vitro studies [9][10][11][12]. Patients who received beta-lactams were selected due to the widespread use of these antibiotics in perioperative prophylaxis and antitumor potential proved in vitro studies [13][14][15][16]. Te main aim of the study was to examine the early recurrence rate, which could be the best reference to efectiveness in relation to anthracyclines as their most signifcant beneft is the limitation of time to frst recurrence [17].

Postulated Mechanism of the Action.
Te antibacterial activity of ciprofoxacin is based on the inhibition of type II topoisomerase [9]. Ciprofoxacin, combined with topoisomerase II and DNA strands, impairs DNA replication [5]. Te inhibition of mitochondrial DNA synthesis leads to the depletion of intracellular ATP resources and consequently to the arrest of the cell cycle in the G2 phase, which initiates p53dependent apoptosis [7]. In eucaryotic cells, the administration of ciprofoxacin causes an increase in ATM kinase serum concentration, which is responsible for DNA repair, which suggests that ciprofoxacin can infuence eucaryotic   [2][3][4][5][6] cells, including cancer cells [15]. Anticancer activity of ciprofoxacin is believed to relate to the inhibition of topoisomerase II, which leads to S/G2 -M cell cycle arrest and apoptosis of cancer cells [7]. Such an activity is shown by doxorubicin, a recognized drug in the treatment of urinary bladder cancer [9]. In vitro studies have shown that depending on the dose used and the incubation time, ciprofoxacin may have a cytostatic or cytotoxic efect on neoplastic cells of colorectal cancer [10,11], lung cancer [10], and urinary bladder cancer [10][11][12]. It has been shown that the concentration of the drug in the range of 50-100 μg/mL inhibits the division of neoplastic cells and the concentration above 400 μg/mL causes apoptosis of neoplastic cells. Te optimal incubation time was 24 hours. Te antitumor activity catalyst was the medium's pH below 5.5 [11]. Te concentration of ciprofoxacin in urine 1-2 hours after oral administration of the drug in a dose of 1 pill of 500 mg averages 400-500 μg/mL, gradually decreasing to the value of about 100 μg/mL 6 hours after administration and 25 μg/mL 12 hours after administration [12,18]. Intravenous administration of ciprofoxacin achieves a similar drug concentration in urine faster and with a lower drug dose [19]. All beta-lactams have the same antibacterial mechanism based on inhibiting the bacterial cell wall synthesis by inactivating transpeptidases, which results in disturbances in muramine production [13]. It has been suggested that new generations of beta-lactams can afect cancer cells without damaging normal eukaryotic cells [14]. In in vitro studies of n-thiol beta-lactams, tumor cells underwent apoptosis by DNA damage induced by caspase 3 and 7 induction. Tis resulted in cell protein degradation and cell cycle arrest in the G1-S phase [15]. Te cytotoxic and cytostatic efect of beta-lactams has been proven for breast cancer cells, prostate cancer, head and neck, lungs, and hematopoietic system [13,15]. Betalactams in the concentration of 50 μg/mL caused apoptosis of the abovementioned neoplastic cells within 48 hours of the optimal incubation time [16]. Potentially, beta-lactams administered 15-30 minutes before transurethral electroresection (TURBT) should reach a concentration having antitumor properties for about 5-6 hours [20].

Analysis of the Results.
Regardless of the division used, the highest recurrence rate was found in the ciprofoxacin group. Ciprofoxacin was used in patients who were allergic to beta-lactams. Te signifcantly worse treatment results in the group with ciprofoxacin could be explained by the fact that the follow-up was the longest in this group. On average. the follow-up was four months longer in the ciprofoxacin group than in the other two groups. Despite diferent followup times in the ciprofoxacin group, it should be noted that antibiotics with the highest antitumor potential in vitro studies showed the worst results.
No statistical diferences existed among patients with highgrade bladder tumors and large tumor mass. Tis may suggest the lack of efect of treatment with alternative chemotherapy methods when malignancy stage and tumor size are increased. Similar conclusions were made by Elsawy and co-authors, who studied the efect of intravesical Epirubicin postoperative instillation among patients (236 patients) with moderate and high-risk nonmuscle invasive urinary bladder cancer [21]. Moreover, patients from both antibiotic groups did not experience any beneft due to early recurrence, which does not correlate with the reduction of 35% recurrence rate among patients with pTa-pT1 bladder cancer treated by intravesical instillation, which was proved in the randomized trial concerning 2278 patients by Sylvester and co-authors [17].
It should be remembered that in the in vitro studies conducted so far, the antitumor efect of antibiotics on cancer cells has been confrmed with long-term incubation of cancer cells (minimum 24 hours) and with a constantly high concentration of the antibiotic [10][11][12][13][14] and that is why the route of administration of the antibiotic is worth considering. No efect of antibiotics given intravenous on bladder cancer cells can be caused by too low concentration    Journal of Oncology chemotherapeutic agents. Just urinary bladder irrigation after TURBT has a benefcial efect on reducing the number of recurrences, which was proved by Grivas and co-authors [23]. Terefore, new more accurate methods of tumor resection (en-block) combined with widely available adjuvant treatment can signifcantly improve treatment outcomes. However, we should remember about growing antibiotic resistance and the specifc phenomenon of antibiotics, which can lower the efectiveness of BCG therapy when used chronic what proved Pak and co-authors [24], that is why new studies must be carefully designed.

Limitations of the Study.
Retrospective, multicenter, and nonrandomized character of the study caused diferences in crucial points such as periods of antibiotics delivery which could make diferences in drug concentration in urine. In the study, it was impossible to obtain a proper comparison in terms of cancer progression and late recurrences due to the insufcient follow-up time (average follow-up was 24 months), as well the time of follow-up difered among the groups, which could infuence results. Te important limitation of the study is the way of drug delivery. Intravenous way of antibiotic dosage may be insufcient to obtain proper urine concentration in adequate time; hence, it would be necessary to project a prospective randomized study with intravesical antibiotics delivery.

Conclusions
Te perioperative antimicrobial prophylaxis given intravenously was of no beneft in relation to the number of early recurrences among patients treated because of primary nonmuscle invasive urothelial bladder cancer.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.