Systematic searching and strict eligibility criteria ensured we could synthesize evidence from five studies, involving 2,057 people. Participants were predominantly men (≈ 71%) and were aged between 30 to 91 years. All included studies were retrospective with follow-up periods ranging from 1 to 300 months. This sample of studies can be considered a reasonably high level of evidence, despite the relatively small numbers. Meta-analysis suggested there is no significant difference between those who develop SBCa and those who do not. However, subgroup analysis of pT0-3N0M0 patients appears to confirm that people who develop SBCa are at greater risk of lower cancer specific survival and overall survival.
For localised or locally advanced UTUC, RNU with BCE provides lasting local control but CSS rates can be as hgh as 70% (18, 19). However, the recurrence rate and progression rate of this type of malignant tumor remains very high (1, 5). Like most other malignancies, pathological stage is the strongest predictor of survival in UTUC patients (3, 18, 20). However, the predictive value of this as an individual risk factor is limited. Given the relative rarity of this disease, analysis such as this are not likely to identify stronger predictive values. As such, there is an increasing awareness of data mining, specifically in oncology, which will enable us to identify further risk factors affecting survival, although at present we do not have the breadth of data required. So even though there are a few studies available in this field, little is currently known about subsequent bladder cancer rates following radical surgery, and how patients’ respond, prognosis and the importance of timing treatments such as chemotherapeutics.
In this study, meta-analytical techniques were adopted to synthesize evidence, not only to produce a more reliable recommendations but also in order to perhaps identify issues which require further consideration in primary research. We first analysed differences in CSS and OS between the SBCa and non-SBCa groups. Results were consistent with the individual studies of Lee et al., Elalouf et al., and Elawdy et al., included here (10, 16). We found that SBCa after radical surgery had no significant effect on CSS and OS; however, each of the individual studies had limitations which may have affected this analysis, For example, most of the included studies only provided single-factor analysis, and therefore we could not obtain the data after the correction of other factors, such as tumor stage, grade, age and gender. This is likely to have caused a degree of bias in the effect values obtained.
In order to reduce the amount of bias related to tumor staging in this meta-analysis, we performed subgroup analysis which excluded T4 or node positive patients. The findings then appear to suggest that patients with SBCa have significantly, poorer survival. Again, this finding is in keeping with previously reports in which researchers observed that SBCa is a predictor of both overall survival and cancer specific survival in the organ-confined UTUC population (11). With non-muscle-invasive UTUC also having been confirmed as a predictor of poorer survival in patients with SBCa (15). While this appears to be self-evident it also raises a number of questions around seeding and cancerization which ought to be addressed.
There are two main theories which may account for these findings i.e. the field-cancerization hypothesis and the tumor seeding theory (21–23). Field cancerization is an hypothesis which assumes that multifocal urothelial carcinoma arise secondarily, and perhaps independently, within the urothelial tract as a consequence of potentially external cancer-causing influences i.e. smoking and the use of specific pharmaceuticals. The tumor seeding theory on the other hand, adopts the rather prosaic analogy of cancer as a flower, budding and then seeding (24–26). While both the field cancerization hypothesis and tumor seeding theory have merit, the seeding theory is more likely in this instance due to the urological procedures involved. Although, further research is necessary to understand subsequent tumor locations and genetic instability.
According to the recent research by Du et al. and Li et al., urothelial carcinoma at which manifest at different sites can occur with independent clonal origin. In some instances, subsequent tumors can be both polyclonal and monoclonal which raises another set of questions about the mechanisms involved in subsequent bladder cancer (22, 27). Indeed, many academics and clinicians alike believe that SBCa which occurs during the early postoperative period can be attributed to the spread and planting of primary carcinoma cells although this may be related to soiling and perhaps to urothelial injury during surgery, which may encourage cellular adherence (15, 28, 29). This is certainly an issue which requires further attention; However at present, we do not have either the data, nor are we designing studies which incorporate a more complex set of indicators for analysis. This should not be left entirely to basic medical researchers. We can, with more rigorous clinical study designs, gain insight into the mechanisms involved.
The development of SBCa research has however revealed that tumor cells are commonly, more aggressive after RNU. As we are acutely aware, quicker cell proliferation and growth corresponds with worsening prognosis which means that postoperative surveillance must be enhanced. At present, guidelines and the included studies appear to have adopted standard follow-up procedures but seeding (and indeed cancerization) may also be related to tumor size and injury or surgical methods e.g. open or laproscopic approaches (1). At present, we have few indicators and fewer effective interventions. One may advocate indiscriminate adjuvant intravesical instillation with chemotherapy but this could have a greater impact on a patient’s physical and emotional well-being.
In this instance, more active postoperative surveillance and further accurate predictive model for SBCa after RNU should be proposed for UTUC patients. This may help improve the long-term prospects of these patients who need improved oncological outcomes. From a strictly surgical perspective, radical resection must comply with oncological principles, that is, surgery of this nature must also attempt to prevent tumor seeding by avoiding entry into the urinary tract during surgery (19). However, previous high-quality evidence has demonstrated that a single postoperative dose of intravesical chemotherapy (pirarubicin, mitomycin C) soon after surgery (< 72 h) reduces the risk of SBCa occurrence within the first year post-RNU (30–33). This is, at present, administered according to the duty physician’s disgression; however, this may need to become a standard procedure to reduce the likelihood of cancerization and seeding taking place.
To the best of our knowledge, we are the first to assess the effect of SBCa on survival in patients with UTUC using standard meta-analytical techniques. However, before providing recommendations, limitations should be also be considered. Firstly, there are inherent issues with retrospective studies such as selection bias which cannot be ignored. The retrospective nature of research also inhibits analysis of necessary factors such as the detailed bladder irrigation chemotherapy regimen which would be helpful as we move forward. Second, there may be a need to incorporate more surgical details and regarding surgeon specific experience and skills. There may be an increased risk of seeding through biopsy and surgical resection (34). Theoretically, circulating cancer cells may be encouraged to leave the upper urinary tract and seed in distant sites and potentially at sites where surgical damage is inadvertently caused. Therefore, future primary research ought to provide more sophisticated details around surgeons’ skill and perhaps around operation specifics.