Mixed low- and high-grade non-muscle-invasive bladder cancer: A histological subtype with favorable outcome, "Beyond the Abstract," by Matthew Danzig, BS; Tina Schubert, MD; and James McKiernan, MD

BERKELEY, CA (UroToday.com) - Our aim was to determine whether mixed-grade, non-muscle invasive urothelial carcinoma of the bladder (UCB) behaves differently from purely high grade UCB. Specifically, we wanted to know if it has a different response to intravesical bacillus Calmette-Guerin (BCG) treatment. Towards that end we analyzed patients from our database receiving primary BCG treatment for non-muscle invasive bladder cancer (NMIBC) over a 5-year period from 2008-2012. We found that 88.2% of patients with mixed low- and high-grade disease vs only 54.4% of patients with pure high-grade disease were successfully treated with BCG alone. The remaining patients died of recurrent disease or received additional intravesical agents or surgery. We termed this measure of success advanced intervention-free survival (AIFS), and analyzed it in a time-dependent fashion with Kaplan-Meier analysis, finding mixed-grade patients to have significantly greater AIFS duration (p=0.030 by log-rank test). This was confirmed in a multivariate Cox regression model, which found high grade vs mixed grade to have a hazard ratio of 4.4 (p=0.040) while controlling for covariates including age, race, gender, smoking history, tumor size, initial stage, multifocality, presence of concomitant CIS, and a history of prior low-grade disease. Of the patients requiring additional treatment who received intravesical BCG + Interferon as their next intervention, 2/2 (100%) of patients with mixed-grade disease but only 5/19 (26%) of patients with pure high-grade disease responded (p=0.035).

Our results are intriguing as they suggest that a substantial subset of patients presently being considered among the high-grade NMIBC cohort in fact has a different character of disease that may have very different clinical behavior.

We consider this study to be hypothesis generating, because while the results were striking, their generalizability are tempered by inherent limitations of our data. Our study was retrospective and observational, with all the typical limitations of such an approach. Follow-up duration was variable, and there was no comparison group of patients undergoing BCG for pure low-grade disease. We also lacked a comparison group of patients who were candidates for BCG induction but did not receive it. Most importantly, we determined mixed-grade vs pure high-grade disease based on the description in the pathology dictation and were unable to carry out re-review of slides to quantify the proportions of high- and low-grade disease and to confirm that the pure high-grade patients did not in fact carry low-grade foci which went unmentioned by the dictating pathologist.

Because of these limitations, additional studies are needed that can confirm our hypothesis. Since pathological data of sufficient specificity and detail is not available in large databases or registries, institutional series will be the domain where confirmatory analyses can be conducted.

A critical element to investigating and elucidating the clinical behavior of mixed-grade UCB may be changing the standard practice of pathological reporting. Our study methodology involved reviewing pathology reports for mention of a mixed-grade picture. As mentioned in our manuscript, we likely miscategorized many cases where mixed-grade histology was present but was not noted by the pathologist. Additionally we were unable to quantify the degree of mixed-grade histology – reports usually made only general mentions such as ‘predominantly low grade with several high-grade foci.’ Lacking specific percentages of low- and high-grade histology, we were unable to evaluate questions such as whether there is a linear relationship between high-grade proportion and outcome, versus a threshold effect with tumors above a certain percentage behaving like their pure high-grade counterparts and tumors below a certain percentage behaving differently.

It would be highly valuable for this change in reporting practice to occur before bladder cancer cases begin to be compiled in the AQUA registry. Going back and retrospectively obtaining histology data through re-review of slides would be far less efficient than including the data in a prospective manner, and perhaps unrealistic at all.

An unfortunate Catch-22 may arise – it may be difficult to justify the effort that would be required to begin collecting this data routinely without studies conclusively proving its value, while such studies may be difficult to conduct without the routine reporting of the data.

A key implication of our findings is that certain cases of mixed-grade NMIBC UCB may behave so differently from their pure high-grade counterparts that treatment with intravesical BCG may not be universally necessary, potentially sparing patients from the toxicities and side-effects of BCG treatment. Since patients with mixed-grade disease are currently treated as high-grade, the vast majority of such patients undergo BCG induction, and so our study was unable to evaluate this possibility. We can only emphasize that mixed-grade disease is a potentially distinct disease entity and wide-ranging investigations into the clinical manifestations of this fact are needed.

Written by:
Matthew Danzig, BS; Tina Schubert, MD; and James McKiernan, MD as part of Beyond the Abstract on UroToday.com. This initiative offers a method of publishing for the professional urology community. Authors are given an opportunity to expand on the circumstances, limitations etc... of their research by referencing the published abstract.

Columbia University College of Physicians and Surgeons/New York Presbyterian Hospital
Department of Urology, 161 Fort Washington Avenue 11th Floor
New York, NY, 10032
Correspondence to: Matthew Danzig at

Mixed low- and high-grade non-muscle-invasive bladder cancer: A histological subtype with favorable outcome - Abstract

More Information about Beyond the Abstract