If I told you there was a new therapy for the treatment of acute lymphoblastic leukemia that improved survival from about 40% to 70%, you would justifiably be impressed. If I told you the treatment costs no more than conventional treatment and was no more toxic, you would be pleased. If I told you the therapy was not new but was developed years ago and is widely available, you would be puzzled. And if I said that its efficacy had been reported in the literature for years, but simply was not offered to many patients likely to profit from it, you would be disappointed or even outraged; as a pediatric oncologist, I am both.
This is not a description of some poor, third-world country that has no infrastructure for providing modern leukemia therapy or money for chemotherapeutic agents. This shameful situation exists today in the US; the patients are adolescents and young adults with acute lymphoblastic leukemia. This is an example of patients who by chance may be treated with a pediatric oncology approach or with an adult oncology approach resulting in very different outcomes.
The latest description of this problem is reported by James Nachman et al (Young adults with acute lymphoblastic leukemia have an excellent outcome with chemotherapy alone and benefit from intensive postinduction treatment: a report of the Children's Oncology Group. J Clin Oncol 2009; 27: 5189–5194).
The authors reviewed the experience of the COG in the treatment of patients with acute lymphoblastic leukemia in a study that enrolled patients from 1996 to 2002. The study was designed for patients who historically had relatively poor outcomes, in this case those 1–21 years of age with an initial leukocyte count of 50,000/µL or more and/or the age of 10–21; 262 of these patients were 16–21 years of age and make up the basis of this report.
The five-year event-free survival of these patients was 71.5%.
These results are outstanding. Though not as good as for patients with more favorable prognostic features, it is remarkably close. What is more important about these results is that results reported for the same group of patients who are treated by medical oncologists have been much worse, about 40% or less. Consequently, medical oncologists often recommend bone marrow transplantation as the initial therapy for patients in this category, a course of action that would probably not be taken if the results with chemotherapy alone were as good as those reported in this paper.
This substantial difference in outcome between treatment given by medical oncologists and pediatric oncologists has been reported earlier by the COG as well as by investigators in France (J Clin Oncol 2003; 21:774–780) and the Netherlands (Leukemia 2004;18: 2032–2035).
What could cause these differences? Nachman points out the variety of explanations that have been proposed, including differences in physician compliance with protocols, patient non-compliance with the regimens, or substantially different protocols, but these are speculations without convincing comparative evidence to back them up.
Nachman and colleagues provide a clue to another possible explanation in their paper. They demonstrated that the frequency of death during the induction of remission, death during remission, and complications during therapy such as stroke, encephalopathy, avascular necrosis, and corticosteroid-induced hyperglycemia, were all more common in the 10–21 year-old age group compared with the 1–9 year-old group. This may have resulted in substantial reductions in chemotherapy that were enough to compromise its effectiveness. This may help explain the less satisfactory outcome in the older age group, but does not explain the differences in outcome depending on who treats the patients of the same age group.
Although there can be differences in leukemia genotype between children and adults, that factor appears to be accounted for in this study. It is also unlikely that there are biological differences in the leukemias of patients who go to a pediatric oncologist compared with those who go to a medical oncologist.
Pediatric Hem-Onc Has Evolved Differently from Medical Hem-Onc
One is left with the possibility that the medical care and perhaps the system of care differ significantly. Pediatric hematology-oncology has evolved differently from medical hematology-oncology.
The first cooperative group was organized to treat childhood leukemia, so systematic clinical trials by cooperative groups and individual institutions have been a hallmark of cancer therapy in pediatric settings from the beginning of the modern era of cancer management.
Furthermore, the vast majority of all children with leukemia in the US in pediatric settings, regardless of geographic location, have participated in clinical trials. Virtually all the children have been treated in academic centers or children's hospitals.
This has had several positive effects. First, by centralizing therapy in these centers, the volume of patients has been sufficient to do comparative trials. Second, and even more important, the techniques and requirements of diagnosis and therapy were standardized in all settings, meaning that the infrastructure and processes of care have been quite similar for all pediatric patients.
But this still does not explain the differences in outcome of adolescents and young adults depending on the pediatric or adult oncology setting. Most adolescent and young adult patients with acute leukemia treated by medical oncologists are managed in academic centers with a smaller fraction treated in community practices. One is tempted to say the pediatric oncologists are better doctors, but there is no objective evidence for that.
So we are left with a dilemma. We don't know exactly why 16–21 year-old patients with acute lymphoblastic leukemia do much better in a pediatric setting. A prudent person would conclude that unless a medical oncology program got the same results, all of these patients should be treated in a pediatric oncology setting. Not to do so would deprive one-third of them of a long term survival. Is that feasible?
Although there are logistic challenges, let's look at it this way: If I told you I had a new miracle treatment for adolescents and young adults that would increase long-term survival from 40% to 70%, don't you think the parents and the patients would beat a path to my door, regardless of the hardships that might be encountered? Don't you think we would feel a strong responsibility to give those patients the best possible chance for survival?