Approximately 50 % of newly diagnosed cases of bladder cancer are low grade, noninvasive and papillary tumors [1]. The standard treatment for non-muscle invasive bladder cancer (NMIBC) is transurethral resection of the bladder tumor (TUR-BT) with or without adjuvant intravesical instillation (IVI) of chemotherapy or Bacillus Calmette-Guerin (BCG) therapy [2]. The most important problems associated with NMIBC are that they have high rates of recurrence and risk of progression. Approximately 50% to 70% of NMIBC have a recurrence within 5 years, and 5% to 20% progress to invasive tumors [3]. To the best of our knowledge, there have been no reports with longer follow-up data focusing on the effects of IVI on tumor recurrence and progression in patients with low grade Ta tumors based on the 2004 WHO classification [4, 5]. Also, there are few published studies in which a large number of patients were followed for more than 5 years after the initial diagnosis. Only two papers provide information on the risk of recurrence and progression after a long tumor-free period [6, 7]. Although some long follow-up studies also showed most tumors recurred or progressed within 5 years, recent data support the need for long term follow-up for more than 10–15 years in such patients even after an initial response to BCG therapy and a recurrence-free period for more than 5 years [6].

In this study, we analyzed the clinical outcome of initially diagnosed low grade Ta tumors after re-assessing all pathological specimens according to the 2004 WHO classification, with a special focus on tumor recurrence and worsening progression (WP) pattern and discuss the need for longer follow-up.

We reviewed the medical records of 242 patients (male: 199, female: 43) who underwent TUR-BT with complete tumor resection for the past 30 years at Keio University Hospital for an initially diagnosed TaG1-2 tumor. We excluded 44 patients who had already undergone TUR-BT at another hospital or who had a history of upper tract urothelial cancer. After re-evaluation of all pathological specimens by a dedicated uro-pathologist, with a special focus on the 2004 WHO grading system, 8 patients with G2 tumors were re-classified as high grade. The remaining 190 NMIBC patients (male: 156, female: 34) who were initially diagnosed with a tumor that was low grade Ta were included in the current analysis. All of the tumors were histologically confirmed as urothelial carcinoma.

These patients were followed by urine cytology and cystoscopy at 3-month intervals during the initial year, every 6 months for the next 5 years, and then yearly thereafter. Intravenous urography, ultrasonography, and/or CT scanning were used to evaluate distant metastasis and upper urinary tract recurrence (UTR) every 1 or 2 years for 5 years. Recurrence was defined as the occurrence of a new tumor in the bladder. Worsening progression (WP) was defined as confirmed (1) high grade Ta, all T1, or Tis/concomitant CIS of bladder recurrence, (2) UTR, or (3) progression to equal to or more than T2.

The use of adjuvant therapies including intravesical BCG or mitomycin (MMC) instillation depended primarily on the discretion of the attending physician. In the overall patient population, BCG and MMC therapies were performed in 71 patients (37.4%) and 12 patients (6.3%), respectively. This study was approved by Keio university hospitals ethical committee. We obtained the patient’s informed consent including their approval for potential use of their anonymized medical data from our data base for research and audit purposes.

The following were analyzed for each individual patient: age, gender, multiplicity and smoking status. Smoking status was classified as 1) nonsmokers; those who had never smoked during their lifetime, 2) ex-smokers; those who had quit smoking before the diagnosis and 3) current smokers; those who still smoked regularly at the initial TUR-BT. Recurrence-free survival rate curves were constructed using the Kaplan-Meier method, and were compared using the log-rank test. Differences among groups were regarded as significant when p < 0.05. Univariate and multivariate analyses of data were performed using the Cox proportional hazards model with stepwise forward selection. These analyses were performed with a statistical software package (SPSS, version 19.0).

We reviewed 190 patients with primary, low grade Ta NMIBC patients and evaluated whether patient-related factors (age, gender, multiplicity, smoking status and adjuvant treatment) were associated with tumor recurrence and WP. Multivariate analysis demonstrated that multiplicity was a risk factor for both tumor recurrence and WP, and that IVI did not affect the occurrence of WP. While none of the patients died of bladder cancer during follow-up, late recurrence and late WP occurred in 11 and 5 patients, respectively.

Zieger et al. presented the natural history of 212 patients initially diagnosed with TaG1-2 tumors for up to 20 years. Only 14 patients received intravesical instillation in their study. Ten of the 212 (4.7%) developed into TaG3 or CIS, 18 (8.5%) developed into T1, and 23 (10.8%) showed muscle invasion or distant metastases [8]. According to our definition of WP, WP was seen in 24.1% in their study, which was relatively high com-pared to our study. Similarly, Prout et al. followed 178 patients with TaG1 bladder tumors for up to 10 years. They reported that a change in grade or stage progression occurred in 13 (7.3%) patients, while only 14 patients (7.9%) received intravesical chemotherapy [9]. Akagashi et al. reported no patients initially diagnosed with TaG1-2 tumors progressed to muscle invasive tumors, while 6 of 62 (9.7 %) patients developed into Tis or T1. One reason for this low percentage of progression was that most of the patients received intravesical chemotherapy for more than 2 years [10]. From these reports, the recurrence rate of initially diagnosed TaG1-2 bladder cancer was 50-60%, and the WP rate was highly variable (between 7% and 24%). In our population of initially diagnosed low grade Ta bladder tumors, the recurrence rate and WP rate were 43.2% and 11.1%, respectively.

We reviewed longer follow-up data for a maximum of 25 years and re-assessed all pathological specimens using the 2004 WHO classification. Only 8 of 198 (4.0%) of G1-2 tumors were re-classified as high grade in our study. Miyamoto et al. evaluated low grade papillary urothelial carcinoma after re-classifying all specimens and reported that 8 of 55 patients (14.5%) were re-classified as having high grade tumors [11]. Pellucchi et al. evaluated tumor recurrence and progression with both the 1973 and 2004 WHO grading systems in patients with primary low grade Ta NMIBC and concluded that the 1973 WHO grading system predicted the risk of recurrence more accurately than the 2004 system and the 2 classifications showed the same accuracy for predicting the risk of progression [12]. The 2013 EAU guideline states that both grading classifications should be used until the 2004 WHO system is validated by more prospective trials [13].

Holmang et al. reported the outcomes in patients treated with BCG intravesical therapy who were tumor-free for more than 5 years (N = 204). Of the 204 patients, 110 (53.9%) had a G1 or G2 tumor. They stated that patients with TaG1-2 tumors treated with BCG have a very good long-term prognosis, but late recurrences were observed. Furthermore, as all low grade recurrences were diagnosed at a follow-up cystoscopy and office cystoscopy generally is a simple procedure, they concluded that continuing to follow patients with TaG1-2 for more than 5 years is encouraging [6]. Our results support their findings. All patients in our study who experienced recurrence in years 5 and 10 were diagnosed at a follow-up cystoscopy. Meanwhile, recurrence in 2 patients who had been tumor-free beyond 10 years was found by gross hematuria. These results suggest that follow-up cystoscopy can be discontinued around 10 years from the initial diagnosis in patients with low grade Ta bladder cancer.

Smoking status is a well-known risk factor for poor outcome in bladder cancer and the strong association between smoking and primary NMIBC recurrence was observed in previous studies [14–16]. However, our results revealed that smoking status is not associated with bladder recurrence rate, WP rate, or late recurrence rate. One of the reasons for our negative result is that the relatively lower percentage of smokers and lower amount of smoking in Japanese NMIBC populations. Further studies with a larger population are warranted in order to evaluate the association between smoking status and tumor outcome in low grade Ta NMIBC.

The present study has several limitations. First, it was performed in a retrospective manner with a limited number of patients, thus unknown sources of bias may exist in the findings. However, since we re-reviewed all pathological specimens and reclassified them as absolute low grade tumors, our results represent more reliable data compared to data obtained before re-evaluation. Second, in our database, tumor size/volume was not included routinely because of the inaccuracy of measurements of tumor size by cystoscopic findings. Finally, we did not provide all patients with a single immediate postoperative instillation of chemotherapy within 24 h or any maintenance intravesical therapies, which may have improved the results.

The tumor recurrence rate and WP rate in patients with primary, low grade Ta bladder cancer were 43.2 % and 11.1 %, respectively. Multiple tumor was a risk factor for both tumor recurrence and WP, while IVI did not affect the occurrence of WP. The late recurrence rate and late WP rate were 14.5 % and 6.6 %, respectively. Our results suggest that routine follow-up of patients with low grade Ta bladder cancer is needed up to 10 years from the initial diagnosis.