Non-muscle-invasive bladder cancer may present as pTa, pT1, or carcinoma in situ (CIS) lesions with the majority of cases (70 %) being pTa disease [12]. Disease recurrence (50–80 % of pTa patients) and disease progression (10–30 % of pT1 and CIS patients) are the biggest threats for patients with non-muscle-invasive BC [12]. The most important clinico-pathological predictors for recurrence are multiplicity, tumor size, and rates of prior recurrences [12, 13]. The most useful predictors for progression are tumor grade, stage, and the presence of CIS, but these parameters also have some predictive value for disease recurrence as well [12, 13]. Sylvester and colleagues developed the European Organization for Research and Treatment of Cancer (EORTC) scoring system using six factors to estimate probabilities of recurrence and progression and defined patient risk into categories of low, intermediate and high. The European Association of Urology has subsequently incorporated this scoring system into its guidelines and the EORTC system has been shown to be a useful tool for identifying high-risk patients with non-muscle-invasive BC [12, 14].

Another important prognostic factor for determining recurrence and progression in patients with non-muscle-invasive BC is the quality of TUR [15, 16]. In up to 30 % of patients receiving a re-TUR for pT1 or high grade tumors, upstaging may occur [12, 16. Also, patients with high grade non-muscle-invasive BC have been shown to respond better to bacillus Calmette-Guerin (BCG) therapy following re-TUR [17]. In patients who develop residual tumors following initial resection, recurrence-free survival was significantly higher after 5 years follow-up in patients who received re-TUR (63 %) compared to those who underwent only one TUR (40 %) [18]. A complete TUR at the initial treatment or after disease recurrence is associated with a lower prevalence of residual tumors and higher rates of recurrence-free survival.

Randomized clinical trials have shown that perioperative intravesical therapy after TUR for patients with non-muscle invasive BC is associated with decreased rates of disease recurrence [19]. Reduction in recurrence may be as high as 39 % compared to patients who undergo TUR alone, and it was estimated that the number needed to treat in order to prevent one recurrence was 8.5 patients. Side effects associated with intravesical chemotherapeutic agents such as epirubicin or mitomycin C are generally mild; however, it should be noted that such treatments are contraindicated in cases in which bladder perforation is suspected.

Intravesical therapy with BCG has been shown to be an effective treatment option associated with a 32 % reduction in disease recurrence [20]. Furthermore, intravesical BCG treatments have been shown to be superior to intravesical chemotherapy in randomized trials [21, 22]. Ten-year progression-free rates and disease-free survival are improved in patients receiving BCG intravesical therapy [23]. Despite the beneficial effects of BCG therapy in these patients, it may still be an underutilized resource for high-risk patients with non-muscle invasive BC [24].

Early RC is the treatment of choice for patients with high-risk non-muscle-invasive BC who fail BCG therapy or for patients with high risk of cancer progression [12, 13, 25]. Adverse prognostic factors such as micropapillary histology, concomitant CIS, high grade, solid architecture, and lymphovascular invasion (LVI) are associated with high risk of progression [26–28]. For the vast majority of high-risk patients, treatment of TUR followed by adjuvant BCG may represent the most reasonable strategy with the option to perform RC early if progression is detected [12].

For patients undergoing RC for muscle-invasive BC, the most significant predictor of oncologic outcome is the extent of lymph node (LN) involvement [29]. Five-year survival rates are 20–35 % for patients with tumor metastasis to LNs [2–4. A more extensive list of LN-related prognostic factors reported to be predictors of outcomes includes the number of positive LNs, the extent of lymphadenectomy and number of nodes removed, and the LN density [29, 30–35]. While no well-defined guidelines for lymph node dissection (LND) during RC exist, numerous studies have suggested that extended LND is associated with better oncologic outcomes and lower risks of micrometastatic disease following RC [29, 30, 34, 36]. Furthermore, performing extended LND may provide more accurate staging. We are waiting for results of an important randomized trial that will tell us the optimal level of LND during RC in order to provide therapeutic benefit while minimizing unnecessary risks.

The second most important predictor of oncologic outcomes after RC is tumor stage [2–4]. The determination of tumor stage may take place prior to RC by evaluating TUR pathology or radiographic images; however downstaging may occur in nearly one quarter of cases [6], and this can have significant implications on how patients are selected for neoadjuvant therapies. Multi-institutional studies have shown that primary pT stage has significant prognostic value in muscle-invasive BC. The 5-year DFS of patients with pT0 or pT1 stage is 80–90 % but those numbers drop to 20–40 % in patients with pT4 stage [2–4]. Higher stages are associated with high risk of recurrence and mortality and may benefit from adjuvant or neoadjuvant chemotherapy.

While tumor grade has significant prognostic value in non-muscle invasive BC, it has not been shown to be a powerful predictor of oncologic outcomes in muscle-invasive bladder as nearly all patients undergoing RC will have high-grade disease [29]. Nevertheless, several grading systems have been developed to provide simple and reproducible tools for clinical use [37, 38].

The presence of lymphovascular invasion (LVI) in RC specimens has been shown to correlate with aggressiveness of BC and shown to be a prognostic predictor of oncologic outcomes independent of lymph node involvement [39–42]. In addition to transitional cell carcinoma (TCC), LVI is a prognostic factor after RC in patients with squamous cell carcinoma (SCC) of the bladder [43]. The presence of LVI may be a valuable prognostic tool when selecting patients undergoing RC for adjuvant or neoadjuvant chemotherapy.

The integration of several prognostic factors into nomograms has been shown to provide more accurate prognoses than grade and stage alone in patients with BC [7, 8]. The International Bladder Cancer Consortium (IBCC) Nomogram incorporates prognostic factors such as age, grade, stage, LN status, and histological cancer type into the nomogram in order to calculate the risk of disease recurrence after RC. It has been shown to have a predictive accuracy of 75 %. The Bladder Cancer Research Consortium (BCRC) Nomogram was similarly developed to predict oncologic outcomes after RC and incorporates grade, stage, LVI, presence of CIS, as well as use of adjuvant or neoadjuvant treatments [8]. Both of these nomograms have been externally validated and shown to be useful tools for patient counseling and selection for adjuvant therapies [44].

The integration of molecular biomarkers with existing nomograms improves the prognostic value and predictive accuracy of those nomograms [45–47]. Increased expression of several molecular biomarkers involved in cell cycle regulation, apoptosis and angiogenesis have been extensively studied and shown to be associated with advanced stage, grade, LVI, LN metastasis, DFS, and CSS in patients with BC [29, 48–52]. Furthermore, the assessment of multiple biomarkers or panels of biomarkers have been shown to be more accurate than assessments of individual biomarkers [45–47]. Evaluation of these biomarkers in patients being treated by RC has been shown to have significant prognostic value in terms of disease recurrence and progression and may be a useful predictor of upstaging in patients undergoing RC [53–55]. Importantly, panels of biomarkers may prove to be the most useful tool in identifying the most appropriate candidates for adjuvant or neoadjuvant chemotherapy.