In this issue of Cancer, Novara and colleagues have published a retrospective review of 269 patients from 3 institutions who underwent radical nephroureterectomy for clinically localized upper urinary tract transitional cell carcinoma (UUT-TCC) over a 16-year time frame.1 They also present a careful analysis of factors that predicted disease-specific survival in this cohort of patients. Their primary findings are an independent association of pathologic stage of the primary lesion, pathologic lymph node involvement, UUT-TCC tumor multifocality, synchronous muscle-invasive bladder TCC, and a history of previous bladder TCC with disease specific-survival. Of these, the factor that was associated with the worst disease-specific survival was synchronous muscle-invasive bladder TCC, in which the hazard ratio for 5-year disease-specific survival was 4.687 (95% confidence interval, 2.326–9.443; P < .001) using the Cox proportional-hazards model. The authors should be commended on the collaborative effort needed to undertake such a study that incorporates data across 3 different centers in 2 different countries. Their study highlights the strengths and weaknesses of these types of efforts and the challenges that remain in our management of this disease.
The strengths of multi-institutional, retrospective studies include the ability to increase the number of patients that can be analyzed in the study and the total number of patient years available for analysis. This is especially important for relatively rare malignancies like UUT-TCC. To date, many of our treatment decisions for this entity, especially in deciding on adjuvant or neoadjuvant approaches to augmenting the results of surgery, often are extrapolated from the literature on its more common cousin in the bladder. The results of multi-institutional studies also may be more applicable to the general urology community rather than reflecting the outcomes of a specific regimen at one (typically high-volume) center.
In studies like these, the issues, which are universal in all retrospective studies, lie in the inherent inability to account for all of the variables that may influence the results. Important variables often cannot be captured adequately and, thus, go unrecognized. A simple example is the small variation in surgical technique across institutions and surgeons that can have an impact on the ultimate outcome of these patients but that is virtually impossible to quantify. Another is the subtleties of patient selection for surgery, a process that involves the careful judgment and interaction between surgeon and patient that also is virtually impossible to quantify in any meaningful way. In addition, there is the ever-present problem of the patient who is not captured by such a study. For example, in their article, the Novara et al. acknowledge that they did not include patients who had isolated distal ureteral TCC who underwent distal ureterectomy and ureteroneocystotomy.1 There also are patients who may have undergone only endoscopic management of their UUT-TCC and patients who may never have made it to surgery at all. The former cohort is especially germane to modern practice, because there is growing recognition that small, low-grade, isolated UUT-TCC lesions may be amenable to endoscopic management without the need for extirpative nephroureterectomy.2–5 These issues are universal across all retrospective studies and not just the study by Novara et al.1 They account at least in part for the finding that, as illustrated by Novara et al. in Table 4 from their article, several studies have been published that examined UUT-TCC, all in a retrospective fashion, and, in many instances, came to different and even contradictory conclusions.
Despite these issues, studies like that presented by Novara et al. are important. Large-scale, Phase III, randomized trials in rare tumors like UUT-TCC often are difficult and impractical. Furthermore, any future trial, even if it has a smaller scale Phase II study design, will depend critically for its success on adequately defining patients in the appropriate risk category for the study in question. The ability to appropriately risk stratify patients is an important tool for appropriate patient selection, counseling on treatment, and selection of the highest risk patients for clinical trials. At least 1 recent report, which also was retrospective in design, has highlighted the need to rethink our approach to UUT-TCC.6 The most logical way to accomplish this would be multimodal therapy to enhance the results that can be achieved with surgery alone. However, if we are to add chemotherapy, for example, should this be done preoperatively or as adjuvant therapy? What is the appropriate choice of drug combination, and how many cycles should be administered? How do we identify the patient at the highest risk for recurrence who may benefit the most from such an approach? A better understanding of the biology and natural history of UUT-TCC and predictors of outcome after treatment will be important as we strive to answer these important questions.
The next challenge, then, is how to come to some consensus about the most important risk factors for UUT-TCC that should be incorporated into clinical practice and/or any future trial design. The relatively rare nature of UUT-TCC means that the studies published to date cannot answer the question adequately in isolation. The results often have been contradictory, as discussed above, leaving the clinician with a bewildering array of variables to consider. One solution, and perhaps the only one that may allow for consensus building, would be to have the centers that have published their results and/or that have an interest in UUT-TCC pool their data. This would allow for a large-scale analysis that, ideally, would incorporate the data and viewpoint of investigators from across a multitude of institutions from various parts of the world. The challenge, which is often the case, will be in the prodigious amount of work and coordination that such an effort would require. Nevertheless, the ability to come to a consensus risk-stratification scheme for UUT-TCC would be the first step in making significant progress toward moving the field forward. Globalization has become a fact of life in the modern age. For rare tumors like UUT-TCC, it is time to take advantage of our increasingly interconnected world to coordinate our efforts in studying this enigmatic and difficult to study disease.