Urologic Oncology

SLIDES & TRANSCRIPTS
Friday, December 13, 2002

Surgical Variables Impact Bladder Cancer Outcomes

Harry Herr, M.D.

Slide 1:

Good morning. When you think of non-muscle invasive bladder cancer as lifelong disease, in all likelihood, the majority at least with high-risk tumors, will eventually require a cystectomy. And my message here is very simple. That if a cystectomy is going to cure the patient, it must be done well.

I'm going to review the surgical variables that may impact post-cystectomy on survival. And I'm going to look at the SWOG-initiated intergroup trial in patients treated with chemotherapy and cystectomy versus surgery alone.

Slide 2:

This is that trial that was presented last year in ASCO in which patients who were randomized to receive three cycles of MVAC before cystectomy had an absolute reduction in mortality of 15 percent, and a prolongation in overall survival of about 2.4 years, over those patients who were randomized and received radical cystectomy alone.

Slide 3:

We looked at this trial retrospectively for the surgical variables that may impact on overall survival. And I looked at 309 evaluable patients, 270 of whom had a cystectomy, and 39 of whom who did not, largely because they refused, having had complete resections, with no evidence of disease after MVAC in most instances, or they were explored, and surgery was aborted because of positive lymph node or unresectability.

These patients were accrued from 1987 to 1998. It says a lot about our ability to conduct randomized trials in this country. The median follow-up is 7.1 years, and we are going to look at 5-year disease-specific and overall survival, and local recurrence.

Slide 4:

There are 109 institutions represented. And you can see the breakdown here -- the academic community, VA, military. There were 106 surgeons who participated, and they were categorized as either: a urologist or an oncologist, that is, someone who is a member of SUO and had special training in oncology; and by volume, those who did five or more cases on this trial, versus those that did fewer than five cases.

Slide 5:

The first piece of information is what happened to those patients who did not have cystectomy, those 39 patients. They virtually all relapsed, despite having received chemotherapy and/or radiation therapy, and eventually died.

So patients with locally advanced in this case bladder cancer, it is unrealistic to believe that chemotherapy and in many cases radiotherapy will salvage a significant proportion of those patients who do not, for one reason or another, have cystectomy.

Slide 6:

The next piece of information is the importance of a soft tissue margin. And again, we talk about this a lot in prostate. We don't say a lot in bladder. But a negative soft tissue margin -- we are not talking about prostatic urethra or ureter here -- but soft tissue margin is absolutely essential for survival.

Those patients who have a microscopic, or a gross positive surgical margin in the soft tissue all recur locally, and they virtually all die. You can see those patients, in some cases the pathologist did not mention the surgical margin, and the survival curve here suggests that in half those cases, it was indeed positive.

Slide 7:

If we look at the extent of the lymph node dissection -- this is my own diagram here, looking at a standard lymph node template, shown there on the left, which I think is what is described in the textbook -- versus a limited dissection, which many individuals do, which is basically only the obturator template.

Slide 8:

And then we look at the survival curves by extent of lymph node dissection. And again, those patients who had the standard dissection and more nodes did substantially better than those patients who had a limited, or in fact the 24 cases who had no lymph node dissection for a variety of reasons.

And in fact, there is really no statistical difference here in five year survival over patients who had a limited obturator dissection, versus no dissection at all.

Slide 9:

If one factors the effects of chemotherapy here, in those patients who had a radical cystectomy according to assigned randomized group, and a standard lymph node dissection, which was again up to the surgeon, their survival curves overlap those patients who had neoadjuvant MVAC and surgery with any kind of dissection, indicating that chemotherapy effects are eclipsed, unless good surgery is done.

Slide 10:

Number of nodes is seen here. You can see the median of 10 nodes in these 270 patients who had cystectomy; 6.3 percent had no nodes. Again, there were a number of patients who had no lymph node dissection, and the range goes from 0-54.

Slide 11:

And if one looks at patients in the top 2 curves that are node-negative, and the bottom two curves that are node-positive, and look at survival curves dichotomized around the median of less than or 10 or more nodes, in each case the curves significantly differ. And the patients who had more nodes retrieved, more nodes dissected and found by the pathologist survived better regardless of pathologic stage [nodes status] than those patients who had fewer in this case, than 10 nodes.

Slide 12:

If you looked at surgical volume, there were only 14 surgeons in this group of 106 that did 5 or more cases. Their survival is significantly improved over those surgeons who did fewer than five cases; 52 surgeons did only one case, 24 surgeons did 2 cases. Again, this is a geographic representation of cystectomy in America, all regions included.

Slide 13:

Significant, but not clear is those surgeons who had special training. Urologic oncologists tend to do a little better, but again, the p value here is quite soft. Volume is much more important than surgeon special training.

Slide 14:

Multivariate analysis, again, as expected, would show P category is highly significant; N status is somewhat significant. But look at the factor here of number of nodes. That is total number of nodes, both node-negative and node-positive. The extent of lymph node dissection, surgical volume, five or greater cases, and of course margins.
And when all of these surgical variables are taken into account, the group, that is whether they received MVAC versus surgery alone, tends to become less significant.

High quality surgery is important.

Slide 15:

To conclude, again this brief overview, there is a lot more data I did not show you. Surgical variables impact bladder cancer outcomes after cystectomy. This appears to be independent of those patients who received chemotherapy. This translates into an experienced surgeon, negative surgical margin, a standard pelvic lymph node dissection, retrieval of at least 10 or more nodes, and a density ratio of less than 20 percent. You will hear more about node density this afternoon.

This means the surgical standards need to be factored prospectively into clinical trials where multiple treatments, especially chemotherapy, are used in order to interpret their results favorably.

Thank you very much.