Radical cystectomy is commonly the treatment of choice for patients with muscularis propria–invasive urothelial carcinoma. Unfortunately, up to 40% of patients whose disease is confined to the bladder wall will develop tumor recurrence; virtually all of them will die as a result of the disease. Thus, it is imperative that we find additional markers (clinical, pathologic, or molecular) to predict adverse clinical events in these patients. The retrospective multi-institutional study reported in this issue of the Journal of Clinical Oncology suggests that lymphovascular invasion is such a marker, given that its presence was found to be associated independently with local and distant recurrence as well as causespecific and overall survival in lymph node–negative patients. This study was based on 750 patients who underwent cystectomy for “usual” urothelial carcinoma and in whom lymphovascular invasion status was determined through review of the pathology reports. The final cohort was selected from a series of 958 patients who underwent cystectomy for bladder cancer at one of three participating medical institutions between 1984 and 2003. Exclusion criteria included incomplete data sets (120 patients), non–transitional-cell histology (62 patients), and lack of information on lymphovascular status (26 patients). The results, if confirmed in additional large, welldesigned—and preferably prospective—studies, suggest that lymphovascular invasion should be included in the clinical staging of bladder tumors, particularly usual urothelial carcinomas, but possibly other histologic variants as well. A precedent for this approach was seen in recent TNM staging criteria used for testicular germ cell neoplasia, in which the presence of vascular invasion in an organ-confined tumor is classified as pT2 rather than pT1 disease. Similarly, patients with pT1-4, N0 bladder cancer could be upstaged accordingly. We have much to explore before considering this hypothesis. As the authors state, the importance of lymphovascular invasion in predicting recurrence and survival in bladder cancer is controversial, in part because of the difficulty in determining its presence at the morphologic level. This fact is well documented in the literature in virtually every organ system. For example, in an article published in this journal by the Testicular Cancer Intergroup Study, the authors found that the recognition of vascular invasion differed significantly between local pathologists and central pathology review. More importantly, vascular invasion as determined by central pathology review was a better predictor of relapse in both stage I and II disease. Similar problems have been reported in bladder cancer. Larsen et al published their experience using Ulex europaeus agglutinin to confirm the presence of vascular invasion in T1 urothelial carcinoma. Of the 36 patients reported to have vascular invasion, this marker was able to confirm its presence in only five patients. The authors correctly emphasized the fact that retraction artifact, which mimics vascular invasion, is commonly seen in invasive urothelial carcinoma. In an article reviewing pathologic prognostic factors in bladder biopsy, transurethral resection, and cystectomy specimens, Lapham et al echo this remark and wisely advise that the presence of lymphovascular invasion should be reported only in unequivocal cases, and that—in some instances—it may require ancillary studies such as immunohistochemistry. Additional validation of lymphovascular invasion as a truly independent marker on which therapeutic decisions will be made is warranted. In addition, pathologists should strive to standardize criteria for establishing the presence JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 27 SEPTEMBER 2