This article summarizes the adventures and explorations in the 1970s and 1980s in the treatment of children with leukemia and cancer that paved the way for the current success in childhood cancers. Indeed, these were adventures and bold steps into unchartered waters. Because childhood leukemia the most common of the childhood cancers, success in childhood leukemia was pivotal in the push toward cure of all childhood cancers. The success in childhood leukemia illustrates how treatment programs were designed using clinical- and biology-based risk factors seen in the patients.
Key points
- •
The success in childhood leukemia illustrates how treatment programs were designed using clinical- and biology-based risk factors seen in the patients.
- •
In the mid-1960s a principal focus in curing childhood leukemia entailed control of the central nervous system part of the disease.
- •
New frontiers were explored and new supportive disciplines were established that paved the way for the current molecular era, which promises to discover new targets for therapy so that we can achieve high cure rates with low morbidity in children with cancer.
The search for understanding is an adventure or more commonly is a series of adventures…Now that geographical boundaries in our own and in other civilized lands have been determined, the pioneering spirits found in scientific research find enticing vistas for adventure.
The Way of an Investigator
When I received the invitation from Pediatric Clinics of North America to guest edit an issue on pediatric oncology, I gladly accepted this challenge, as it gave me an opportunity to present the advances from a perspective of one who started a career in pediatric hematology oncology when the cure rates were abysmally low in contrast to the optimism for curing all children with cancer now—current estimates project nearly 80% long-term survival rate for all children with leukemia and cancer. A quote from Dr Sanford Leiken’s section in the 1962 issue of Pediatric Clinics of North America on Leukemia: Current Concepts In Therapy illustrates status of childhood cancer therapy in the early to mid-1960s, “At present acute leukemia of childhood is not curable, but it is treatable; although fatal, it can be controlled in varying periods of time so that the patient’s life can be prolonged in a relatively comfortable and functional state.” It was in this milieu that I started my own career. The task of caring for children was grim indeed. Our standard opening dialogue with parents of a child with leukemia (most cancers) newly diagnosed started with the sentence “Your child has leukemia/cancer and there is no cure for it.”
One of my first experiences was coming across several children with leukemia who were long survivors, many of whom were included in the original publication of long survivors written by Joseph Burchenal and M. Lois Murphy from Sloan-Kettering Institute of Cancer Research and Cornell University Medical College in New York. In this publication, Drs Burchenal and Murphy attempted to list all of the known survivors of acute leukemia from direct correspondence with hematologists in the United States. There were 71 patients with acute leukemia living 5 years or more from diagnosis at that time. Of those, only 36 were living with no evidence of leukemia. A large cohort of these patients had come from Children’s Hospital of Michigan and was treated under the direction of Wolf W. Zuelzer, my mentor, and mentor to noted pediatric oncologists – Sanford Lieken, William Newton, and Theresa Vietti. One of the patients (GB) in the Burchenal cohort was a child with acute lymphoblastic leukemia (ALL) treated in 1952 who I tried to contact in preparation for a report on long survivors in childhood leukemia. To my disappointment, the parents did not permit me to contact the young man, as at that time the practice was to not tell the children of their diagnosis. He received aminopterin for a total of 10 months, the first generation of antifolates synthesized by Yellapragada Subbarow of Lederle Laboratories. This is one of the antifolates first used by Sidney Farber; although, in his famous article, Dr Farber failed to include Dr Subbarow as coauthor, an astonishing omission considering that Subbarow synthesized the antifolates specifically for treatment of childhood leukemia at Dr Farber’s request. As things turned out, I discovered patient GB at a fundraising golf outing in 1998 and fortuitously earlier this year, through a colleague at another golf event, he reestablished contact with me. Now 67, one of the longest survivors of childhood leukemia, he is doing well and has a thriving engineering business.
In the next few paragraphs I summarize the adventures and explorations in the 1970s and 1980s in the treatment of children with leukemia and cancer that paved the way for the current success in childhood cancers. Indeed, these were adventures and bold steps into unchartered waters. Because childhood leukemia is the most commonly known childhood cancer, success in childhood leukemia was pivotal in the push toward cure of all childhood cancers. The success in childhood leukemia illustrates how treatment programs were designed using clinical- and biology-based risk factors seen in the patients. Thus, in the mid-1960s, although the remission/induction rate was quite respectable by even current standards, relapse occurred frequently, and the overall cure rate still remained less than 10%. A major problem in child ALL was relapse at extramedullary sites, most often in the central nervous system (CNS). A collage of the types of extramedullary disease we were seeing is shown in Fig. 1 . Hence, a principal focus in curing childhood leukemia entailed control of the CNS part of the disease. Concurrently with these attempts, new developments were occurring rapidly in the immunophenotyping and karyotyping of the childhood leukemias and thus a better definition of the molecular biology and risk factors for the type of relapses seen. Investigators pursued varied strategies, and these initial forays remind me of an explorer of yesteryear who discovered new sea routes to India, discovered the New World, and changed the face of the earth.
The single most important event that changed the disease pattern undoubtedly is work by Donald Pinkel and his colleagues at St. Jude Children’s Research Hospital. I recall distinctly what was considered sensational at that time. The world came to know of the impact of craniospinal radiation as a means for preventing CNS leukemia one night on June 26, 1972 when Danny Thomas, founder of St. Jude Hospital, appeared on The Tonight Show with Johnny Carson and famously declared that the physicians at St. Jude’s found a cure for childhood leukemia. The impact of this one television appearance is unparalleled by any other single event in the modern therapy of childhood cancer. Overnight there was a nationwide communication of this news and then no doubt around the world. Pinkel and his group used 2400 cGy radiation to the whole brain and 1200 cGy to the spinal axis immediately after remission induction and noted that nearly 50% of the children treated in that manner had initial remission lasting more than 3 to 5 years. This news, of course, met with some expected skepticism at many levels and investigators at other leading centers attempted to duplicate the results or developed alternative methods for prevention of CNS disease. A debate ensued as to how the craniospinal radiation may be preventing CNS leukemia and achieving the cures noted. Those who accepted the sanctuary theory of cause of CNS disease attempted to duplicate the St Jude results. Skeptics of this notion, including Zuelzer felt that seeding of the CNS could occur throughout the duration of the disease. Additionally, half of the children still experienced relapses both in the CNS and extramedullary locations such as testes and marrow relapse.
Thus, 2 groups embarked on alternative approaches. Protocols were designed rapidly within a few days, and new methods of CNS prophylaxis were explored. In Detroit, we used the so-called intermittent intrathecal methotrexate and fractional radiation (IMFRA) regimen of intermittent methotrexate and fractional radiation in which intrathecal methotrexate was given at the end of induction and every 10 weeks thereafter, along with radiation of 200 cGy to the skull by 2 opposing ports and 100 cGy to the spinal axis. Ten such courses were given, and the estimate was that the total dose of radiation approximated the 2400 cGy of cranial radiation and 1200 cGy to the spinal axis used by Pinkel and colleagues. Simplistic as this notion may be, this approach also resulted in nearly 50% cure with long-term remissions at 3 years. At the same time, using the high-dose methotrexate therapy for treating children with leukemia devised by Isaac Djerassi, investigators at Roswell Park devised a CNS prophylaxis regimen that combined intrathecal methotrexate with 3 courses of intermediate-dose methotrexate at 500 mg/m 2 times a total of 3 courses with leucovorin rescue. In a seminal publication in 1983, Freeman and colleagues from CALGB (Cancer and Acute Leukemia Group B) showed that this approach yielded results comparable to that achieved with the cranial radiotherapy for childhood ALL. These 2 studies found that a strategy of intermittent CNS prophylaxis would be equally effective as the initial cranial radiation approach. An advantage of the regimen devised by Freeman and colleagues was that the methotrexate dose could be pushed to higher levels (up to 33 g/M 2 in one study), whereas the initial 2400 cranial radiation approach was limiting subsequent therapy especially because use of high-dose methotrexate after prior cranial radiation resulted in high incidence of leukoencephalopathy. A potential benefit was the avoidance of long-term effects of cranial radiation during brain growth in small children and late complications of secondary malignancy. Thus, today, the preferred approach to prevention of CNS disease in childhood ALL is based on intrathecal methotrexate and high-dose intravenous methotrexate.
To summarize the experience with the first-generation CNS prophylactic regimens, it was evident that 50% long-term remissions can be achieved. Clinical risk factors for prediction of outcome were devised combining the initial white count, the original prognostic marker, with age at diagnosis and the newer markers were applied for classification of ALL. The so-called lymphosarcoma variant of ALL with a thymic mass and poor-outcome thymic enlargement was identified as T-cell derived using E-rosettes, mature B phenotype was recognized by surface Ig staining, and the so-called null cell (neither T or B) leukemias could be identified as either B lineage (B precursor) or T lineage by the first-generation polyclonal antibodies. These developments ushered in the era of immunophenotyping of ALL, completed by the development of monoclonal antibodies for the common ALL antigen marker CD10 (CALLA), T cell markers (OKT3, 5, 7), B lineage markers (Ia, Dr) and the CD34 stem cell marker. The first wide-scale application of cytogenetics in childhood ALL led to the recognition that high hyperdiploidy was associated with good outcome. We now know that the clinical markers of yesteryear are surrogate markers for biology of the disease—mediastinal mass indicates T-ALL, FAB L3 is associated with mature B phenotype, infant ALL is invariably CD10 negative, E2A- PBX all is CD34 negative, and MLL gene rearranged leukemia frequently expresses CD15. The Children’s Cancer Study Group (CCSG) defined ALL good prognostic group of age 3 to 7 years and white count less than 10,000 is within the 2 to 7 year age peak observed in childhood ALL and identifies most cases of high hyperdiploidy and ETV6/RUNX1 cases, the two subgroups of ALL associated with best outcome. For a description of the current understanding of the biology of childhood ALL see the chapter by Bojwani and Pui in this issue.
The euphoria initially accompanying first-generation CNS prophylactic regimens in ALL was tempered with the recognition that half of precursor cases fail, results were lower in T-ALL, and the results in mature B-ALL were worse ( Fig. 2 ). Based on these facts, several investigators, including the investigators from Pediatric Oncology Group (POG) and Nordic Pediatric Hematology/Oncology Society, explored increasing the dose intensity of high-dose methotrexate by increasing the dose per course and the number of doses, ultimately resulting in the current strategy of including 4 to 6 courses of high-dose methotrexate as a part of the consolidation strategy in childhood ALL. This strategy improved the outcome over the 50% level achieved with the prior regimens, with a particular benefit to the group with E2A-PBX1–associated ALL and T-ALL. Over time, the intracellular mechanisms for methotrexate resistance were defined (high content of dihydrofolate reductase, the target enzyme for methotrexate, and low folylpolyglutamate synthase in T-ALL blasts) thus explaining the improved outcome in T-ALL with dose intensification of methotrexate. The concepts of age-based dosing of intrathecal methotrexate and triple intrathecal therapy added to improved CNS control. Addition of l -asparaginase to the standard 2-drug regimen by Children’s Cancer Study Group was another important development. Although the initial remission induction rates were not different, there was improvement in long-term outcome by reducing CNS relapse. Other groups like the Berlin, Frankfurt, Munster (BFM) group who continued to use cranial radiotherapy for CNS prophylaxis, introduced the concepts of reintensification courses, and the Dana Farber consortium explored high-dose asparaginase, both resulting in significant improvements in outcome. Mature B-ALL, once invariably fatal, suddenly became curable with the use of high-dose methotrexate and fractionated cyclophosphamide, concepts developed by investigators from France, St. Jude Research Hospital, and the Pediatric Oncology Group.
