Treatment of Acute Lymphoblastic Leukemia: A Historical PerspectiveCHAPTER 1

A true pioneer in cancer chemotherapy, Yellapragada Subbarow, already an accomplished biochemist at Harvard University had moved to Lederle Laboratories (a division of American Cynamid Company) in Pearl River, New York after he was denied a permanent appointment at Harvard. At Lederle, Subbarow developed a method to synthesize folic acid following which he worked on various folic acid conjugates. He provided Drs Farber and Louis Diamond with adequate quantities of various folate antagonists including 4-aminopteroyl glutamic acid (aminopterin) for clinical testing.^2,3 During the period from November 1947 to April 1948, these investigators treated 16 children with acute leukemia with various folate antagonists including aminopterin. Ten of these patients showed “clinical, hematologic, and pathological evidence of improvement” to aminopterin, a remarkable result at a time when no effective therapy existed and even temporary remissions were unheard of. Farber, Diamond, and colleagues published the details of five of the responders in a seminal publication in the New England Journal of Medicine on June 3, 1948.⁴ Probably unaware of the magnitude of their contribution to ALL chemotherapy, they were cautious in interpreting their results and made it clear in the paper that these remissions were “temporary” and did not represent a “cure”. It is a testament to Dr Farber's conviction in his ideas that he, a pathologist turned clinician persevered at a time when there was nihilism among clinicians about the value of treating pediatric leukemia as well as skepticism from his colleagues of the value of his findings. His work laid the foundation for specialized care for pediatric leukemia and led to the establishment of Children's Cancer Research Foundation which would later become the world renowned Dana Farber Cancer Institute.²

Eventually, the less toxic folate antagonist methotrexate (amethopterin, MTX) replaced aminopterin as a mainstay of ALL therapy. Subsequently, intrathecal administration of MTX was found to be effective in treating central nervous system (CNS) involvement of ALL.^5,6

In a landmark paper published in 1953, Burchenal et al. reported results of treating various cancers including acute leukemia with 6-MP. The majority of patients in this study were children with acute leukemia. These children had the best responses to 6-MP with 15 of 45 achieving “hematologic remissions” including some who were resistant to folic acid antagonists. These investigators were particularly surprised by the lack of non-hematologic toxicity especially when 6-MP was used to treat childhood acute leukemia.⁸ The pronounced antileukemic activity of 6-MP in ALL would soon be confirmed by other investigators (see below under Early Clinical Trials). It was also noted later that the antileukemic activity of different nucleoside analogs against acute leukemia was markedly different with agents like 6-azauracil showing no benefit.⁹ Hitchings, Elion, and Black were awarded the Nobel Prize in 1988 for “discoveries of important principles for drug treatment”.

Soon other institutions would join this group which was named Acute Leukemia Cooperative Group B (ALGB), a precursor of the modern cooperative group Cancer and Leukemia Group B (CALGB).^17,18 In one of their earliest trials (Protocol 2), this group conducted a 318 patient trial from 13 institutions and conclusively proved that treatment with 6-MP and MTX together produced higher rate of complete remission than either agent alone.¹⁹ The era of large cooperative group studies in acute leukemia had begun.

These early trials were limited in their use of modern statistical techniques. For the next trial (Protocol 3), however, the group with the assistance of statistician Edmund Gehan applied novel statistical techniques to the trial design.¹⁷ This was a trial comparing 6-MP to placebo maintenance in patients who had achieved a complete remission with corticosteroids alone. Patients in complete remission were randomly assigned to 6-MP or placebo in a double-blind fashion and patients on placebo 5could cross over to 6-MP if they had a relapse. The duration of remission between the arms was compared using a restricted sequential procedure where patients at each institution were paired and randomly assigned one or the other therapy followed by assessment of preference for 6-MP or placebo based on remission duration. The trial had to be stopped early when a clear preference for 6-MP maintenance was noticed. After enrolment of 21 patient pairs, the median duration of remission was 33 weeks for 6-MP versus 9 weeks for placebo.²⁰ The statistical analytical techniques used in this study were truly novel at that time and in many ways this trial embodied much of the principles of a modern double-blind, randomized clinical trial.¹⁷

Drug screening in the 1950s was aided in large measure by the development of the carcinogen induced murine leukemia model L1210 by Lloyd Law at NCI which became the primary screening model for acute leukemia.²¹ Cyclophosphamide was found to be active in ALL by Donald Fernbach and colleagues.²² Subsequently, trials of multiagent combination chemotherapy would be conducted and were found to yield remissions of durations not achieved in ALL thus far. Notable in this regard are the NCI VAMP [vincristine, amethopterin (methotrexate), mercaptopurine, and prednisone) regimen and CALGB protocol 6313 (13th protocol of 1963) consisting of same agents as in VAMP plus cyclophosphamide and carmustine, both of which tested combination therapy with active agents available at that time.^23,24 These regimens were formulated based on the fundamental principles of combination chemotherapy, i.e., combining active agents with different mechanisms of action and non-overlapping toxicity. Development of combination regimens was also helped by the studies of Howard Skipper, a mathematical biologist at Southern Research Institute in Birmingham, Alabama who formulated the “cell kill” hypothesis based on kinetics of the L1210 mouse model.^23,25

Based on outcomes of the various clinical trials done thus far, Donald Pinkel at St. Jude Children's Research Hospital in Memphis, Tennessee introduced the concept of “Total Therapy” consisting of phases of induction, consolidation, CNS directed therapy, and maintenance, of all which form the cornerstones of modern ALL regimens. By around 1965, this therapy was yielding survivals of 5 years or more in a significant number of treated children and “cure” in ALL had become a reality.^26,27 L-asparaginase was subsequently found to be active in ALL and added to pediatric ALL regimens (see chapter 8 for detailed history of its development). Over the next 4 decades or so, various cooperative groups including the Berlin-Frankfurt-Munster group formed in 1970 would obtain incremental improvements in survival by modifications of the concept of total therapy and achieve astounding success in the treatment of childhood ALL. It is important to note that the only major drug development during this period was the introduction of tyrosine kinase inhibitors which improved the outcome of Philadelphia chromosome positive ALL. The survival improvements in adult ALL have been very modest and painfully slow with 6long-term survival rates only around 50% at the present time. Various reasons have been attributed to this disparity between adults and children and are discussed in Chapter 7. Cooperative group trials also have made therapy better tolerated with less long-term side effects; the elimination of prophylactic CNS radiation is most notable in this regard.

E Donnall Thomas at University of Washington, Seattle who would later receive the Nobel prize for his pioneering work on bone marrow transplantation had done extensive studies in mice and dogs to define radiation doses for conditioning. In 1959, Thomas and colleagues performed syngeneic bone marrow transplants in two children with ALL who were conditioned with total body radiation. Although engraftment was achieved, both patients would succumb to disease relapse.²⁹ Mathé is credited with successfully performing the first sibling bone marrow transplant for acute leukemia in April 1963. This patient, a 26-year-old physician with refractory ALL was conditioned with total body irradiation and infused with bone marrow from six donors including four siblings and both parents. Hematopoietic engraftment from one of the brothers was demonstrated and so was tolerance to that donor. This patient remained in remission until December 1964 when he died of complications from herpetic encephalitis.³⁰ Although Mathé and colleagues had performed allogeneic bone marrow transplants in at least 21 ALL patients by 1965, there were very few long-term survivors with most patients succumbing to infections or graft-versus-host disease, the latter complication was well characterized in their reports and referred to as “secondary syndrome”.^30,31 With knowledge of the genetics of human leukocyte antigens (HLAs) as well as refinements in HLA typing, developments in immunosuppression, and supportive care, allogeneic hematopoietic stem cell transplantation (HSCT) would become a widely used and effective therapy for ALL (see chapter 12).7

Mathé and colleagues also tested other forms of immunotherapy in ALL including Bacillus Calmette-Guérin (BCG), irradiated allogeneic lymphoblasts, and even infusion of leukocytes collected by leukapheresis from patients with chronic myeloid leukemia.^28,32 In one controlled trial, they showed prolongation of remission duration by immunization with BCG or irradiated leukemic cells.²⁸ These therapies were clearly innovative at that time. However, interest in immunotherapy other than hematopoietic stem cell transplantation waned due to the high efficacy of chemotherapy in childhood ALL, lack of adequate technology, as well as the belief that ALL was not amenable to immune mediated therapy, probably based on the lack of efficacy of donor lymphocyte infusions in inducing remissions in patients who had relapsed after allogeneic HSCT. This was generally perceived as evidence for lack of potent graft versus leukemia effect in ALL.

Much progress needs to be made in adults to achieve results similar to childhood ALL. Paradoxically, in children, the current focus is to reduce or tailor therapy especially in good risk patients in order to minimize deleterious effects of therapy on a growing population of childhood ALL survivors. Assessment of minimal residual disease status and pharmacogenetic factors would be expected to allow such modifications of therapy (discussed in chapters 9 and 11, respectively).

The early history of ALL therapy is a remarkable story of successful collaboration between academia and industry as well as collaboration among clinical investigators to efficiently perform meaningful cooperative group clinical trials. Both of these aspects of clinical research have been the subject of much criticism recently. One can only hope that the current decade will see progress in adult ALL at a pace that is reminiscent of progress in childhood ALL therapy made in the 1950s.8