Do oncologists treat patients as they would themselves?

ORLANDO-Do oncologists treat their patients with colon cancer the same as they would treat themselves? Usually. A survey presented at the 2008 Gastrointestinal Cancers Symposium probed the secret world of oncologists’ personal preferences (abstract 444).

“The choices regarding adjuvant chemotherapy are challenging for both the patient and physician,” said principal investigator Neil Love, MD. In colon cancer, he noted, there are a number of options, and sometimes the potential benefits may be modest while risks can be substantial.

“Patients in all areas of medicine frequently deal with this by asking their physician, ‘What would you do?’ We wanted to know if patients with colon cancer are doing the same,” said Dr. Love, president and CEO of the Miami-based medical education company Research To Practice. Dr. Love is a medical oncologist but is no longer in clinical practice.

Dr. Love and his colleagues from six major cancer centers were interested in learning whether oncologists’ personal treatment selections for colon cancer differ from standard recommendations, and whether they would treat patients differently than they would treat themselves.

To evaluate these questions, the researchers administered a 10-minute online survey to 150 medical oncologists. Participants were asked how they manage colon cancer for various types of patients, and what adjuvant therapy they would choose for themselves in the described situations.

Five scenarios

The survey presented five scenarios for a hypothetical 55-year-old colon cancer patient:

(1) Stage III disease with 2 of 18 nodes positive.

(2) Stage III disease with 15 of 18 nodes positive.

(3) Stage II disease with 8 of 8 nodes negative and no other high-risk features.

(4) Stage II disease with 18 of 18 nodes negative and no high-risk features.

(5) Scenario #4, but respondents were only provided the odds for 5-year relapse based on Adjuvant! Online data.

The treatment choices included the use of newer agents both on and off protocol for stage III disease.

How many ask?

“We found that on average 41% of patients ask their oncologists what they would do in their situation; 70% of oncologists said they regularly answer the question, and another 20% said they answer if pressed to do so,” Dr. Love reported.

For 73% of the responses, the oncologists’ recommendations matched their personal choices. This was true even in the very-high-risk situation represented by the stage III patient with 15 positive nodes (scenario 2): 31% said they would recommend FOLFOX to these patients and 23% said they would use it themselves in this setting.

“We wondered if the oncologist would want bevacizumab [Avastin]. Outside of a clinical trial, most would not,” he said.

For scenario 2 patients, 26% of the oncologists said they personally would try FOLFOX plus bevacizumab off-study, and only 10% said they would recommend that their patients do so

(P < .05). However, 42% said they would try this regimen on a clinical trial, and 50% would advise their patients to enroll in such a trial.

An even smaller percentage would consider FOLFOX/bevacizumab in stage III patients with two positive nodes (4% would recommend it to patients and 11% would use it themselves, P < .05), but about one-third would recommend it as part of a clinical trial, both for patients and themselves.

“Oncologists want to be convinced. They really believe we should have the evidence before we act,” Dr. Love said of the bevacizumab results.

Stage II more controversial

More controversial is the management of stage II patients. For scenario 3, the stage II patient with only 8 nodes examined (none positive), nearly half the respondents said they would recommend adjuvant chemotherapy for both themselves and their patients, in what is now accepted as a higher-risk situation.

The stage II patient with 18 nodes examined and none positive (scenario 4) is trickier, he said. In keeping with a conservative ASCO position statement, 70% of oncologists said they would not generally recommend adjuvant chemotherapy for this patient, though fewer (53%) said they would decline it for themselves (P < .05).

The only other significant difference in this setting was for use of capecitabine (Xeloda) monotherapy: 8% said they would recommend it to patients while 17% would try it themselves (P < .05).
 

Impact of Adjuvant! Online

Interestingly, their approach for the scenario 4 patient changed when the investigators described this case only in terms of the quantitative recurrence risks derived from Adjuvant! Online.

When presented with the scenario in which the 5-year risk of relapse for a patient is 13% without treatment, 10.5% with fluorouracil (5-FU) or capecitabine (Xeloda), and 8.1% with oxaliplatin (Eloxatin) and 5-FU, most said they would recommend adjuvant chemotherapy.

In particular, 83% said they would recommend FOLFOX to this lower-risk stage II patient, and 78% said they would use FOLFOX themselves.

“When you use Adjuvant! for this patient, a little note pops up to tell the user there is controversy about whether clinical trial evidence has documented a benefit to treatment in this situation,” Dr. Love noted. “However, most oncologists believe that, even in the lower-risk colon cancer patient, chemotherapy provides at least a small benefit and might be worth it. The data are clearer for breast cancer.”

A subtle disconnect

To sum up, Dr. Love said, 70% stated they would not recommend chemotherapy in a situation perceived qualitatively as “lower risk stage II” (scenario 4), yet when presented with a set of quantitative recurrence risks derived from Adjuvant! Online for the same scenario, almost all of the respondents said, yes, they would recommend chemotherapy in this setting.

The co-authors of the study are not sure how to explain this “subtle disconnect,” Dr. Love said. It likely reflects the ongoing uncertainty about the benefit of adjuvant chemotherapy for stage II disease, he said. “But it also suggests that medical oncologists would quickly adopt this treatment approach, even for a relatively small reduction in recurrence, if they were convinced that this modest benefit exists,” he said. “It is not that they don’t know the numbers, but that they don’t have confidence in them.”