The results of the enclosed survey of 150 randomly selected US-based medical oncologists and 27 clinical investigators specializing in gastrointestinal cancer provide an intriguing snapshot of current oncologic practice. From a macro, human perspective, perhaps the most important aspect of these treatment patterns is that in spite of enormous efforts by patients and healthcare professionals and the commitment of substantial economic, physical and intellectual resources, 52,000 individuals — enough to fill up your average football stadium — will still succumb to this disease annually.

Could some of the morbidity and mortality of this disease be avoided with a more standardized implementation of the colorectal cancer clinical research base? Considering the results of this survey, we can make the argument that, in general, most patients are receiving state-of-the-art oncologic care. However, some of the findings suggest exceptions to this rule and provide fascinating insights into the variations in how oncologists make recommendations for patients. Here are a few of the most thought-provoking numbers...

1. 48%...

Fraction of practicing oncologists who would not recommend adjuvant chemotherapy or would recommend only a fluoropyrimidine to a 65-year-old patient with colon cancer and eight negative lymph nodes (Figure 2).

Eighty-two percent of investigators would offer FOLFOX in this situation because a number of data sets clearly demonstrate that patients with node-negative tumors but fewer than 12 identified nodes are at comparable relapse risk to some individuals with Stage III disease. The potential undertreatment of this patient subset has significant public health implications, especially considering data presented at the 2007 ASCO meeting suggesting that approximately 50 percent of patients with “Stage II” disease — perhaps as many as 20,000 patients in the United States alone — have fewer than 12 nodes identified after resection.

Part of the solution to this unfortunate situation is that pathologists simply need to be more vigilant in looking for and finding at least 12 nodes. I personally don’t buy into the claim that this is a multifaceted problem caused both by inadequate surgery and pathologic evaluation, as commented on by Dr Alberto Sobrero during his discussion at the recent ASCO meeting in Chicago.

Dr S acknowledged that he may be a bit biased on this issue as his wife is a pathologist, but I have heard time and time again from many oncologists in practice and at tertiary centers that when the pathology report describes fewer than 12 nodes, they simply “send the pathologist back to the basement,” and usually, a couple of days later, at least 12 nodes show up.

A joint education effort targeting pathologists, surgeons and oncologists may result in fewer relapses and maybe less need for chemotherapy in this important patient population.

2. 74%...

Fraction of investigators who generally would offer adjuvant FOLFOX to an 85-year-old patient with 15 of 25 positive nodes (Figure 8).

It is interesting that only 27 percent of practicing docs would take this approach, favoring instead either observation or a fluoropyrimidine alone. What’s even more fascinating is that it’s not at all obvious who’s right. The docs in practice are probably approaching this from an Adjuvant! Online perspective, taking into consideration that the calculated absolute benefit from treatment at age 85 will likely be modest due to competing causes of mortality.

On the other hand, the investigator perspective is also valid and is perhaps best expressed by Rich Goldberg, who at our recent clinical investigator Think Tank noted that he uses FOLFOX for the elderly patients with positive nodes unless there is severe unrelated comorbidity, based on the logic that if therapy isn’t administered (and even if it is), the patient will likely end up receiving chemotherapy anyway in the next two or three years for relapse, so why not treat for cure?

3. 36%...

Fraction of practicing oncologists who consider bevacizumab a reasonable nonprotocol adjuvant option for a 65-year-old patient with 15 positive nodes (Figure 9).

In stark contrast, none of the investigators would consider nonprotocol bev in this situation. The compelling question of whether to utilize an experimental arm of an ongoing randomized trial is a daily part of oncology practice, and in breast cancer, we can make a valid argument that in 2004, after the first publication of safety data with adjuvant trastuzumab, we moved too slowly in using it as off-study treatment for select patients.

At this point, it’s a bit difficult to defend off-study adjuvant bevacizumab for colon cancer, although one might argue that 15 positive nodes is tantamount to metastatic disease. Hopefully, the day will soon arrive when we will have an initial answer to this critical question in the form of NSABP-C-08, but at the moment, data from this historic clinical trial are quietly percolating in the recesses of a computer somewhere in Pittsburgh.

4. 74%...

Fraction of investigators who believe patients in the adjuvant setting should be informed that modest exercise (and diet) may significantly reduce the risk of recurrence (Figure 18).

Most docs in practice also support this strategy, and on this one it’s difficult to argue, particularly when the “downside” — as noted by WINS maven Rowan Chlebowski — is less heart disease. The most important data set on this issue comes from CALGB-C89803, Len Saltz’s “negative” trial of adjuvant IFL. While this highly anticipated study turned out to be a major disappointment, the silver lining is that it provided a stunning data gold mine of lifestyle correlations with cancer outcomes, as reported by Jeff Meyerhardt, Charlie Fuchs and colleagues.

In 2005, this team presented prospectively gathered findings demonstrating a remarkable 50 percent lower relapse rate among patients who exercised moderately (six hours of walking a week). Similarly impressive findings on diet and relapse rate were subsequently presented at the 2007 ASCO meeting, and interested parties can read all about it in the August 15, 2007 issue of JAMA. While dietary and exercise changes seem a bit low-tech compared to targeted molecular therapy, the data are impressive and cannot be ignored.

5. 68%...

Fraction of practicing docs who would be comfortable putting patients on the new SWOG randomized trial (iBET) evaluating continuation of bevacizumab upon disease progression in the metastatic setting (Figure 31).

Not surprisingly, 100 percent of investigators support this critical trial, and both groups tell us that part of their motivation to see this study succeed comes from our previous failures in breast cancer. To date, no randomized clinical trial has evaluated the benefit of continuing trastuzumab upon disease progression for patients with HER2-positive metastatic breast cancer, and this has put patients and doctors in a bind for many years.

Without data from a randomized trial, clinicians and patients are stuck with gut feelings and scarce data as the main means to make the decision to continue this relatively safe but also expensive therapy that may or may not potentiate cytotoxic agents even on disease progression. Based on these encouraging responses showing support for this trial, I’m betting on iBET.

6. 41%...

Fraction of clinical investigators who have used oxaliplatin as part of nonprotocol neoadjuvant chemoradiation therapy for rectal cancer (Figure 38).

Approximately one quarter of practicing docs replied similarly, and one wonders whether the post hoc addition of an oxaliplatin randomization to the current major US Phase III randomized study evaluating neoadjuvant therapy (NSABP-R-04), along with encouraging Phase II data, had led some docs to consider using this agent, particularly for younger patients with more locally advanced disease.

Most investigators believe that if neoadjuvant therapy is used, postop adjuvant treatment is required, even in the face of a pathologic complete response, and that adjuvant therapy, as in colon cancer, is likely to be FOLFOX. Incorporating oxaliplatin into neoadjuvant therapy is appealing not only for local control, but to get a head start on systemic protection.

7. 43%... (mean)

The likelihood that first-line systemic therapy would control progression of disease for two or more years in a 60-year-old symptomatic patient with newly diagnosed colon cancer, according to investigators (Figure 22).

These experts would also tell such a patient that there was a 68 percent likelihood that antitumor therapy would control cancer symptoms, at least in the short term.

In our recent Patterns of Care surveys, we have continuously experimented with new and subtle ways to tease out the thinking behind current practice patterns.

One powerful tool for accomplishing this goal is to ask questions that go to the heart of what docs actually say to their patients. I don’t know how I would feel if I had metastatic colon or rectal cancer and heard these numbers, which are probably a bit better than I would have imagined — and a lot, lot, lot worse than they need to be.

Neil Love, MD
NLove@ResearchToPractice.com

Select Publications