What if High-Dose Chemotherapy for Adjuvant Breast Cancer Doesn't Work?
Special Feature
What if High-Dose Chemotherapy for Adjuvant Breast Cancer Doesn't Work?
By Thomas J. Smith, MD, FACP
I see a lot of 46-year-old women with 12 of 19 positive lymph nodes, ER-, high S-phase cancers. After all, I am one of two medical oncologists for our multidisciplinary Breast Health Program. They all want to know how to maximize their chance of disease-free survival.
Easy, right? Four cycles of standard CAF, then high-dose chemotherapy like at Duke, then adjuvant radiation therapy to the chest wall. After all, Dr. Peters had a 70+% disease-free survival at six years in his report on women with more than 10+ lymph nodes, compared to about 40% with conventional treatment. (Peters WP, et al. J Clin Oncol 1993;11:1132-1143.) And CALGB considered high-dose chemotherapy as one of two "standard" options in their maximum Adriamycin -> Paclitaxel -> Cyclophosphamide trial for women with 6-9 nodes; the cognoscenti did not support a trial of just four cycles of standard "AC" or "CAF" chemotherapy.
How about the first randomized, clinical trial that actually tested if high-dose chemo added anything? (Rodenhuis S, et al. Lancet 1998;352:515-521.) What, you didn't hear the news? Among big 1998 medical news events, this one seemed to get the least news coverage of any.
The Background
Since the publication of the Peters article in 1993, I have offered high-dose chemo as one alternative for those not fitting our clinical trials (and less than 10% do). I have tried to be scrupulously honest, and tell people of the preliminary nature of the results, the possibility of selection bias, the real selection bias of aggressive pre-chemo testing, etc. And I have offered standard chemotherapy plus radiation therapy as one option; usually people have come in with pre-determined notions, and, it has been rare that I have changed their minds about treatment. And the low toxicity of high-dose chemo, apparently confirmatory findings from the North American Blood and Marrow Transplant Registry, etc., have all contributed to my feelings of security.
The Trial
I was scanning PubMed for breast cancer genetics/recurrence rate articles late at night when I saw this Lancet trial. There had not been anything on National Public Radio, local news, the Washington Post, etc. All my medical friends were as unaware as I.
Investigators at the Netherlands Cancer Institute enrolled 97 women younger than 60 years old who had operable breast cancer with level III lymph node involvement (documented with infra clavicular node biopsy.) All had extensive staging with blood tests, chest radiograph, liver ultrasound, and bone scan. All underwent chemotherapy with 5-FU 500 mg/m2 + epirubicin 120 mg/m2, and cyclophosphamide 500 mg/m2 IV every three weeks (equivalent to FAC or CAF.) All patients had definitive surgery that included axillary node dissection after cycle three.
The standard group got a fourth cycle of FEC, then radiation therapy to the involved field, then tamoxifen 40 mg a day for two years.
The high-dose chemo group got a fourth cycle of FEC, then filgrastim 300 mcg for stem cell mobilization. At least 3 ´ 106 CD3 cells were collected. Then, the group received high-dose chemo with cyclophosphamide 6 g/m2, thiotepa 480 mg/m2, and carboplatin 1600 mg/m2 (a regimen in widespread use). Radiation therapy and tamoxifen followed.
Sixteen of the 97 patients were not randomized; 11 because they did not want to undergo high-dose chemo.
The trial had decent power; sufficient patients were treated in order to have an 80% chance to detect a 30% increase in disease-free survival. For this level, 35 patients in each arm and 38 events were required; this report was made on the 38th event.
The Results
There were no toxic deaths. Ninety percent of patients were discharged on or before day +18. There was no difference between the arms in disease-free or overall survival. (See Table.)
Could this be a fluke? Yes, of course. But it is the first randomized clinical trial from a group that knows how to do this well. It might have missed a much smaller effect, say three of 100 women or even 10 of 100 women. But it set the target at the original 43% to 72% improvement reported by Peters et al with historical controls. And, as this group has pointed out, there are adverse neurologic and psychologic consequences to high-dose chemotherapy. (Van Dam FSAM, et al. J Natl Cancer Inst 1998;90:210-218.)
Table
Results
Result |
|
|
Number treated |
|
|
Relapses |
|
|
Deaths |
|
|
Event-free survival |
|
|
Overall survival |
|
Ouch!
That was my first response. I hate breast cancer, and hope that I have been doing some good with high-dose chemotherapy, and now four cycles of paclitaxel after "AC." It hurts when our assumptions are trashed. And I get no pleasure from a negative trial that does not improve results, even if it is true.
What will it take to change practice?
I have one breast cancer patient in our BMT unit now undergoing high-dose chemo, and just discharged another. Both were offered Phase II trials to improve stem cell collection procedures, but that is incidental to the results. No doubt I will see similar patients this week.
Should I stop offering high-dose chemotherapy outside of a clinical trial? Should I have never started? We have one randomized trial open, the CALGB A->C->T vs. AC, then a high-dose chemo trial for women with 6-9 positive nodes that seems to suggest the standard care of four cycles of AC is not a gold standard. We have the NSABP trials open, but getting that group to allow chest wall radiation therapy for premenopausal women has not been easy. The last patient I remember trying to enroll on the CALGB 9082 of CAF vs. CAF + high dose chemo got disqualified for microscopic dermal lymphatic involvement on biopsy.
This trial should give all of us pause about recommending high-dose chemotherapy outside of a clinical trial. To be honest, we should show prospective patients these data and point out no difference in this small randomized trial. To ignore it as "too small," or "not long enough follow-up," or "non-standard regimen" will suggest a strong bias in interpretation. (Or, don't confuse me with facts unless they agree with my ideas.)
At a minimum, this trial should help us explain what the expected difference between standard and high-dose chemo is likely to be-not 30 of 100 women. For some women, three of 100, even with neurologic (mainly cognitive) toxicity might be acceptable to choose high-dose chemo. For others, this is not acceptable.
I have also faxed the study to the Virginia Breast Cancer Foundation for their education programs. We all want a cure for breast cancer. None of us want excess toxicity unless it is justified. And we need to be honest about the data as it appears, even when it hurts.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.