Toxicity of CNS Prophylaxis for Childhood Leukemia

Over the past 20 years, dramatic improvements have been made in the treatment of childhood malignancies. Today, most children who have cancer are expected to survive their disease and become healthy, productive members of society. Children with acute lymphoblastic leukemia (ALL) are a good example. Although pediatric ALL was an invariably fatal disorder in the past, children with this cancer now have a 5-year event-free survival rate of more than 70%.

We do not like to use the word "cure" when talking about children with malignancy, however, because to be "cured" these children have to live to their 70s or 80s and succumb to old age. Since successful, modern chemotherapeutic regimens began in the 1960s, we should not talk about cures, but we can take pride in the more than 10-, more than 20, or more than 30-year survivors. Considering the intensive multimodality therapies required for this achievement and all their potential side effects, it is gratifying that, when compared with the population at large, "most pediatric cancer survivors successfully achieve 'adult life goals' as measured by employment, acquisition of health and life insurance, marriage and reproduction."[1,2]

The Down Side

Unfortunately, every silver lining has a cloud. Heart disease and cancer, the two leading causes of death in the general population, may occur earlier and at a higher frequency in survivors of childhood malignancy.[3-6] Furthermore, comprehensive assessment of the health status of long-term cancer survivors has revealed overt and subclinical dysfunction in various organs of some.[7] Many of these problems occur in a only small fraction of the population and are of variable clinical significance, but some may have a major impact on patients' quality of life. Moreover, some dysfunction, such as anthracycline-induced subclinical cardiomyopathy, is expected to progress as the population ages and develops atherosclerotic heart disease.[8]

Survivors of childhood leukemia share many of the late effects seen after the treatment of solid tumors. These include: (1) endocrine complications, such as ovarian, testicular, pituitary, or thyroid dysfunction, abnormal pubertal timing, infertility, and growth disturbance; (2) neuropsychological complications, such as psychological, intellectual, neurologic, social, and vocational dysfunction; (3) abnormal organ function, such as cardiac, pulmonary, renal, hepatic, gastrointestinal, dental, and ophthalmologic disorders; and (4) carcinogenesis.

Neurotoxicity of CNS Prophylaxis for ALL

In the setting of multiagent, multimodality therapy, it is sometimes difficult to determine the cause of a specific dysfunction. The article by Waber and Tarbell concentrates on the toxicities of central nervous system (CNS) prophylaxis used in the treatment of children with ALL. Without CNS prophylaxis, more than 50% of these patients would develop a CNS relapse even though they have no evidence of CNS leukemia at diagnosis.[9] Central nervous system relapse is a very ominous event, not necessarily because these cells "reseed the marrow," but because CNS relapse is frequently the first sign of resistant disease at multiple sites and because patients are frequently left with severe neurologic dysfunction.

Although CNS leukemia is much more damaging to the nervous system than is CNS prophylaxis, the three modalities employed in CNS prophylaxis-- cranial irradiation, intrathecal drug administration, and IV chemotherapy that crosses the blood-brain barrier--all have some degree of neurotoxicity. When more than one modality is used, the toxicities can be more than additive.

The Waber/Tarbell article concentrates mainly on the impact of CNS prophylaxis on physical growth and mental development, two problems that differ in their clinical impact. The authors nicely summarize the reported problems and offer some useful interventions for those affected. But we still do not have a good understanding of the prevalence or severity of these problems. For example, how widespread is the problem of major learning disabilities after cranial irradiation?

The difficulty lies with the lack of good data on which to base such observations. Multiple studies of small patient groups have reported minor, but statistically significant, differences in group mean IQ score or another measure of neuropsychological function between patients who received 2,400 or 1,800 cGy of cranial irradiation for CNS prophylaxis and those who received only intrathecal therapy. Unfortunately, it is not the three-, four-, or five-point difference in mean IQ that is significant to an individual patient. What is significant is that a few patients may have a 30- to 40-point drop in IQ, and yet this may not even change the mean or median score of the group.

Avoidance of CNS Irradiation

Nevertheless, it is fairly clear from all of the studies that cranial irradiation is a major contributor to the neuropsychological, neuroendocrine and carcinogenesis problems seen in patients who receive such therapy. Fortunately, this was recognized a number of years ago, and more effective systemic and intrathecal therapy has been successfully substituted for cranial irradiation in more than 60% of children with ALL. Randomized studies have demonstrated that good- and average-risk children with ALL no longer need cranial irradiation for CNS prophylaxis.[9]

The long-term survivors whose problems we have to deal with today are still the product of the initial attempts at curative therapy. Hopefully, the next cohort will have fewer difficulties. Today's long-term survivors do need our help.[10] As Waber and Tarbell aptly state, "identifying and acknowledging the problem as legitimate can allow solutions to emerge," both for the problems at hand and for modifying therapies so that such problems can be prevented in the future.