Acute Promyelocytic Leukemia: A Clinical Guide

© Springer International Publishing AG, part of Springer Nature 2018

Oussama Abla, Francesco Lo Coco and Miguel A. Sanz (eds.)Acute Promyelocytic Leukemia https://doi.org/10.1007/978-3-319-64257-4_1

1. History of Acute Promyelocytic Leukemia

Laurent Degos¹  

(1)

Paris Diderot University—Academy of Medicine, 16 rue Bonaparte, Paris, 75006, France

Laurent Degos

Email: [email protected]

Keywords

ATRACell modifiersDifferentiation therapyHistory of malignanciesLow-dose ARA-C

Introduction

Within a 60-year time span, the most severe acute leukemia—termed acute promyelocytic leukemia (APL) by the Norwegian author Leif Hillestad—was identified, characterized, and cured. The determination of a precise molecular definition of the genetic defect and the emergence of an antidogmatic paradigm have made the history of APL a model for the treatment of malignancies. However, the new concepts of in vivo malignant cell differentiation and cell death induced by cell modifiers were quite controversial, as illustrated by the anecdotes reported in this chapter. Furthermore, over the course of APL’s history, the dialogue between onco-hematologist physicians and scientists from several worldwide hospitals and laboratories led to an impressive achievement: for the first time, a malignancy (and the most severe malignant disease of the blood, no less) was cured in standard conditions using cell modifiers without any chemotherapy, cytotoxic agents, or bone marrow grafts. Today, the prognosis for APL depends more on the timing of treatment initiation than on the treatment itself, as we shall see in this chapter.

A Distinct Entity: A Special Type of Leukemia (1957–1987)

In the first 30 years after its identification, APL followed the conventional course of all types of leukemia: clinicians precisely described the disease and tried to manage its treatment. Long-term survival rates approached 25%, but clinicians were confronted with early bleeding diathesis. Late relapses (after 2 years) were rarely reported.

The Clinical Description

The first description of APL by Leif Hillestad, published in Acta Medica Scandinavia in 1957 [1], summarized the main clinical features of the disease and is still relevant today:

"Evidence is presented for the existence of a special type of acute myelogenous leukemia. Three cases are described, characterized by:

1.

A very rapid fatal course of only a few weeks’ duration.

2.

A white blood cell picture dominated by promyelocytes.

3.

A severe bleeding tendency due to fibrinolysis and thrombocytopenia.

4.

A normal ESR probably caused by reduced fibrinogen concentration in the plasma.

It is suggested that this type is named acute promyelocytic leukemia. It seems to be the most malignant form of leukemia."

In 1959, Jean Bernard [2] reported the first series of 20 patients with APL, disclosing more detailed criteria: numerous large granules in the cytoplasm of abnormal promyelocytes in the bone marrow, covering the nucleus and sometimes assembled in faggots of Auer rods, as well as a low count of blasts with a monocytoid appearance in the blood. Jacques Caen [3] more precisely defined the acquired hypofibrinogenemia. In fact, bleeding diathesis was the most impressive feature of the disease, accounting for 20–30% of deaths, which were mainly due to cerebral hemorrhages.

In 1976, the French–American–British (FAB) Nomenclature Committee assigned a specific classification of acute myelogenous leukemia (AML), the M3 type [4]. Later, the committee officially recognized a variant form of APL that combined bilobed nucleus blasts, nonvisible granules on light microscopy (microgranules), and a positive myeloperoxydase reaction, and which was often associated with high white blood cell counts and similar coagulation disorders [5]. A very rare variant form with basophilic microgranules instead of large azurophilic granules was also accepted as APL [6].

A Controversy About Treatment

The unpredictable onset of life-threatening bleeding diathesis was the major obstacle in the treatment of patients with APL. The disease seemed to be particularly sensitive to anthracycline treatment [7], showing high rates of complete remission. However, chemotherapy exacerbated bleeding diathesis and thus increased the risk of death. The central question of this era was how to manage coagulation disorders.

All hematologists agreed on platelet transfusions, so the controversies surrounded hypofibrinogenemia. Clinicians needed to determine if they were dealing with a primary or a secondary hypofibrinogenemia due to disseminated intravascular coagulopathy (DIC). These two conditions had different drug treatment options (antifibrinolytic drugs, heparin, or no coagulation drugs), platelet transfusion times, and chemotherapy initiation points (immediately or, in cases of DIC, after heparin treatment initiation). The presence of fibrinogen-fibrin degradation products in serum could not distinguish the two conditions. In France, a decrease of factors V and X led clinicians to treat patients with low-dose heparin. However, in Italy [8], normal levels of protein C and antithrombin III associated with an acquired reduction of alpha-2 plasmin inhibitor levels indicated a primary fibrinolysis.

Apart from age, the intensity of hypofibrinogenemia and high white blood cell counts were considered to be the two major prognostic factors: the first indicated an increased risk of early mortality and the second indicated a higher risk of relapse.

A Cytogenetic Signature

The abnormal cytogenetic feature that confirms the specific entity of APL was first described in 1976 as a partial deletion of chromosome 17 [9] and identified later by Janet Rowley from the same team as a balanced reciprocal translocation between the long arms of chromosome 15 and 17 [10]. The t (15;17) translocation was consistently found in APL bone marrow cells [11].

Thus, by the mid-1980s, APL could be defined not only by its morphological features and typical bleeding diathesis, but also by a specific cytogenetic abnormality.

The disease needed an adapted treatment: an aggressive and urgent initiation of anthracycline treatment, with special attention to the coagulation disorders. With this, a complete remission rate of approximately 75% was achieved, despite a high early mortality rate. Relapses still occurred in the first months after complete remission, with approximately 25% of patients surviving for more than 2 years.

A New Paradigm: The Differentiation of Malignant Cells

During the same time period, studies demonstrating the ability to transform malignant cells into terminally differentiated normal cells were received with skepticism. The dogma of the irreversible status of malignant cells was deeply anchored in the spirit of physicians and scientists. The only accepted option to treat malignancies was to eliminate the cells by chemotherapy, radiotherapy, or surgical excision.

Abolishing the Dogma of the Irreversible Status of Malignant Cells

An Experimental Approach

In 1963, Leo Sachs (Rehovot, Israel; Fig. 1.1) developed a culture of cloned blood cells from mice [12]. From this, a number of growth factors were identified in 1970, , which he named in Hebrew; they were later renamed as granulocyte colony-stimulating factor, macrophage colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, and interleukin-3. Sachs also demonstrated that cell lines from leukemic mice could be differentiated and become nondividing mature granulocytes or macrophages in cell cultures as a result of stimulation by various differentiation factors [13]. His presentation in 1973 at the Congress of the International Society of Genetics in Berkeley, California, was received with respect and some skepticism due to the artificial conditions of the experiments (cell lines, in vitro studies) and the absence of formal genetics. Later, in 1982, Sachs found that myeloid leukemic cells injected into embryos participated in apparent hemopoietic differentiation in the cells of adult mice [14].

Fig. 1.1

Leo Sachs, 1924–2013

A Clinical Approach

At Hospital Saint Louis in Paris, France, in June 1980, Laurent Degos proposed that his 52-year-old female patient with a resistant relapse of acute myelogenous leukemia return home for the end of life. The patient lived far away, in French Britany. Because her family was abroad, low-dose cytosine arabinoside (ARA-C; 10 mg twice a day subcutaneously) was initiated to delay the patient’s death for few days. Four months later, the patient returned to the hospital in good condition with normal blood and bone marrow cell morphology. The general practitioner described a progressive improvement over several weeks. Two other patients experienced similar gradual improvements [15], with differentiation of leukemic cells in blood and bone marrow until complete remission was achieved. These clinical cases confirmed the experimental data of Leo Sachs on the reversibility of malignancies. Sachs often mentioned low-dose ARA-C as the human in vivo therapeutic achievement of his findings. The persistence of a cytogenetic signature in complete remission provided evidence in favor of the actual differentiation of malignant cells [16].

These results led to the new concept of differentiation therapy for human acute leukemia. However, in a trial using low-dose ARA-C, only 35% of patients obtained complete remission [17]. Later, alpha interferon treatment for hairy cell leukemia was also demonstrated to directly act as a cell modifier, interfering with the autocrine loop of proliferation [18].

All-Trans-Retinoic Acid as a Potential Candidate for Differentiation Therapy of APL

Although the treatment of malignancies was mainly focused on decreasing proliferation using antimitotic drugs, some scientists investigated the blockage of the differentiation of malignant cells. The arrest of maturation had been a major feature of malignancy since the first description by the pioneer Alfred Donné in 1844 [19].¹

Murine myeloid leukemia cell lines, such as M-1 or erythroleukemia induced by the Friend virus, as well as human HL-60, KG 1, or K 562 cell lines, could be in vitro differentiated into granulocytes or monocytes using several agents, including polar plana compounds (e.g., the use of dimethyl sulfoxide for murine erythroleukemia cells by Charlotte Friend in 1971), butyric acid, hypoxanthine, ligands to nuclear receptors (vitamin D, or retinoic acid) and some antimitotic drugs (mainly ARA-C and aclacinomycin). In 1981, Theodore Breitman demonstrated the sensitivity of HL-60 cell lines to retinoids, as well as of cells from two patients with APL treated in short-term cell cultures [20]. However, an HL-60 cell line only possessing one chromosome 17 and not carrying the specific t (15-17) is not a promyelocytic cell line.

Cellular biologists and clinicians working on differentiation therapy assembled bi-annually, led by Samuel Waxman, Giovanni Battista Rossi, and Fuminaro Takaku. The first international conference on differentiation therapy in cancer was held in 1986 in Sardinia. This conference included the first generation of scientists and physicians investigating this new and antidogmatic field.

The Dinner that Initiated All-Trans-Retinoic Acid Treatment for APL (1985)

Using a list of potential differentiation agents, Christine Chomienne tested more than 60 specimens of fresh bone marrow cells from various patients with leukemia (instead of cell lines) in short-term cultures. Chomienne demonstrated that the differentiating effect of retinoic acid was specific for APL. Among various vitamin A derivatives, all-trans-retinoic acid (ATRA) was potentially 10 times more effective than 13-cis or 4-oxo, whereas etretinate did not induce any differentiation [21]. At that time, etretinate (Tigazon) was the only derivative available in Europe and 13-cis (Roaccutane) was the only one available in the United States. Therefore, it was impossible to treat patients with an adequate derivative.

In 1985, a travel grant offered by Air France to Chinese professors of medicine provided an opportunity for the hematologist Wang Zhen Yi, Dean of the University of Medicine of Shanghai and French speaker by education, to visit Hospital Saint Louis. During a dinner at the home of Marie Thérèse Daniel, an FAB committee member (also attended by Chen Zhu, Wang’s young student who was working in Daniel’s morphology laboratory), a discussion between Laurent Degos and Wang Zhen Yi concerned a treatment designed to induce differentiation using low-dose ARA-C in patients and the specific in vitro activity of ATRA in APL (a drug not available in Western countries). The discussions resulted in the initiation of a close collaboration between the two researchers and in a decision to manufacture ATRA for APL patients in China. Several exchanges between the two institutes of hematology in Paris and Shanghai allowed for rapid progress against the disease. The two hematology institutes joined forces (Fig. 1.2) and established a formal agreement on April 25, 1987.

Fig. 1.2

Wang Zhen Yi with Laurent Degos and Christine Chomienne

ATRA was first used to treat patients with APL in 1987 at Rui-Jin Hospital in Shanghai, where Laurent Degos visited to observe the remarkable effects of this treatment. Degos and Wang delivered a joint presentation on differentiation therapies using low-dose ARA-C for AML and ATRA for APL at the Second Conference on Differentiation Therapy of Cancer in September 1987 in Bermuda [22]. Huang Meng Er [23] of Shanghai reported on the in vivo maturation of malignant cells until complete remission in 22 of 23 Chinese patients who were newly diagnosed with APL and treated with 45 mg/m² of ATRA. Sylvie Castaigne also reported on the remission of 19 of 20 patients who were treated in France after experiencing a relapse [24]. Complete remission was obtained without aplasia, alopecia, or primary resistance to the drug. Few infections were observed and coagulopathy rapidly improved within a few hours. The progressive terminal differentiation of leukemic cells in bone marrow (sometimes with the presence of Auer rods in mature granulocytes) completed the picture of unusual features for the treatment of myelogenous leukemia [24]. In response to a question from an Impact Médecin journalist on February 1, 1991, Wang Zhen Yi said in French: Without the Laurent Degos team researchers and without our meeting face to face in 1985, nothing could have occurred. The Paris Saint Louis institute studies opened my eyes about the possibilities of all-trans-retinoic acid.²

Hoops and Hurdles

Although the availability of ATRA was no longer an obstacle, the road to a cure was not clear. Western companies still refused to manufacture the drug. ATRA was kindly provided by Shanghai producers and was transported by Chinese students when they travelled to Paris, the first of whom was Huang Meng Er.

Two anecdotes illustrate the skepticism that existed in those times on the use of ATRA as an antileukemic agent. First, when Christine Chomienne called Werner Bolag (a scientist from Roche Headquarters in Basel, Switzerland, who specialized in ATRA) for some advice during the first ATRA treatment in France, he was horrified and asked her to immediately stop the treatment. Against his advice, Chomienne continued the treatment. In the second anecdote, a hematologist from New York asked the Paris team to obtain the drug for a young, relapsed patient. The package was ready to be sent with the approval of the Chinese collaborators, but the administration of the U.S. hospital refused to treat the patient with such an experimental drug made in China and provided by French physicians. The patient subsequently died.

No More Product

Events in China in June 1989 made it difficult for Chinese students to travel. After the Tiananmen square demonstration, the French Government required all French institutions to end their collaborations with China. A shortage of ATRA occurred while several French patients were being treated. Confronted with this difficulty, Laurent Degos contacted Roche France. Victorine Carré agreed to make the drug and excluded all women from the factory during production (to prevent any teratogenous effects from retinoid contamination in the air). The board of Roche France asked Laurent Degos to restrict the administration of ATRA to French patients and to take total responsibility for any adverse events. According to the policy of Roche headquarters in Basel, vitamin A derivatives were exclusively used for patients with skin disorders.

During a 1989 spring meeting in Paris on antibodies appearing during interferon treatment for hairy cell leukemia chaired by Loretta Itri (Vice President of Roche; Nutley, NJ, USA), Laurent Degos asked her to obtain an ATRA source from Roche USA. Loretta Itri’s answer was to consult her husband, Raymond Warrell, who was a hematologist at Memorial Sloan Kettering Cancer Center (MSKCC) in New York. Raymond Warrell was surprised by the differentiation effects of ATRA considering the irreversible status of malignancies. He suggested that Degos present his results in August 1989 in New York. In this presentation, Degos showed a series of bone marrow samples from several patients at various times after treatment, demonstrating the progressive terminal differentiation of malignant promyelocytes. The audience, which included the heads of MSKCC, was convinced. Raymond Warrell asked Loretta Itri to manufacture 2 million ATRA tablets—not only for patients with APL but also for extensive clinical trials involving other cancers and leukemias under the auspices of the National Cancer Institute. This happy ending opened the door to ATRA treatment for patients with APL all over the world.

Is APL a Pseudo-Leukemia?

One year later, French clinical results obtained using ATRA from China and Roche France [24] and cellular investigations [25] were published in the same issue of the journal Blood, which also printed images of the cells on the front cover. The journal included an editorial by Peter Wiernick, which asked whether APL was another pseudo-leukemia.³ Is APL treated with a natural derivative of vitamin A similar to pernicious anemia treated by vitamin B12? A malignant cell could not remain malignant if its status was reversible. The concept that Leo Sachs formulated was not yet fully accepted in 1990. At that time, most thought leaders believed that if effective cell modifiers could treat a cancer, then the so-called cancer was not a malignancy.

Early Relapses with ATRA

Differentiation of malignant cells and rapid improvement of bleeding diathesis were the two breakthroughs in the treatment of APL [23, 24] confirmed in 1991 by a Chinese group [26] and by Raymond Warrell [27]. They provided evidence of the differentiation process by fluorescence in situ hybridization (FISH) and of the clonality using X chromosome-linked polymorphisms.

However, even though almost all patients experienced complete remission with ATRA (no primary resistance), all of them relapsed within 3–12 months (median: 5 months). The relapses were resistant to ATRA (secondary resistance) [24]. Considering the previous evidence that patients who were successfully treated with high-dose daunorubicin chemotherapy [28] had few relapses after 2 years, Laurent Degos decided to initiate a combination of ATRA and chemotherapy in 1990. Patients first received ATRA until they achieved complete remission. ATRA was then followed by intensive chemotherapy (induction and two courses of consolidation) to combine the positive effects of the two treatments—that is, complete remission with rapid disappearance of bleeding diathesis by ATRA and a relapse-free survival by intensive chemotherapy.

Hope for a Curable Disease

The first nonrandomized trial treated 26 patients using ATRA until complete remission, followed by three courses of daunorubicin and ARA-C. These results were compared to historical control groups of patients who received the same chemotherapy without ATRA [29]. The addition of ATRA greatly reduced early mortality and the number of early relapses. These favorable results prompted French investigators to launch the first randomized trial in 1991 using a similar study design but comparing the results to a control group instead of a historical group of patients. The trial ended prematurely after 18 months at the end of 1992 because event-free survival was significantly higher with ATRA [30]. This was the starting point for a big jump in long-term survival for patients with APL, from 25 to 75%.

Worldwide Enthusiasm

The second pivotal randomized trial was launched in 1993 by the same European Cooperative Group. The trial compared patients who were receiving ATRA followed by chemotherapy (the reference treatment) with patients who were receiving ATRA and concomitant chemotherapy followed by two courses of consolidation [31]. This trial demonstrated the superiority of the simultaneous regimen and the advantages of a maintenance therapy.

At that time, nonrandomized trials conducted by Wang Zhen Yi in China, Raymond Warrell in New York, and Akihisa Kamaru in Japan confirmed the beneficial effects of ATRA. Meanwhile, in 1993, an Italian group (Gruppo Italiano Malattie Ematologiche Maligne dell’ Adulto, or GIMEMA) launched by Giuseppe Avvisati investigated a combinatorial approach of ATRA plus idarubicin (AIDA) for induction followed by three courses of consolidation [32]. This approach was further refined by a Spanish group (Programa Español de Tratamientos en Hematología, or PETHEMA) led by Miguel Sanz, which explored the benefits of treatment deintensification (especially in the consolidation phase). Japanese investigators led by Luizo Ohno made efforts to organize a controlled multicenter trial (the Japan Adult Leukemia Study Group, or JALSG). Furthermore, a U.S. intergroup also met several times to find a consensus for a study including ATRA. They started a randomized trial comparing ATRA plus two courses of chemotherapy to three courses of chemotherapy. In 1997, Martin Tallman reported better results when ATRA was administered, but an unexplained lower complete remission rate than the other cooperative groups (68% vs. ~90%) [33].

By the end of 1993, a high rate of complete remission, a reduced risk of early mortality, an absence of primary resistance, and a low relapse rate generated hope for curing at least 75% of patients with APL [34]. The first meeting on APL entitled APL, A Curable Disease was organized in Rome and chaired by Franco Mandelli, the founder of the GIMEMA group (Fig. 1.3). This positive outlook on a new original treatment for malignancies encouraged national and international collaborations, which led to larger studies: the European Cooperative group, which including French, Belgian, and Swiss researchers; the U.S. intergroup; a Japanese cooperative group; Chinese cooperative trials; PETHEMA in Spain enrolling patients from South America and Hovon in the Netherlands; a France-India cooperation; several meta-analyses jointly conducted by GIMEMA-PETHEMA and the French-Belgian-Swiss group in collaboration with PETHEMA; and finally, as we shall see later, the Italian-German GINEMAZ-SAL-AMSLG group.

Fig. 1.3

First meeting on APL in Rome

New Clinical Horizons, Unknown Conditions

Hematopathologists were captivated by the features of progressive differentiation of malignant cells with distinctive Auer rods in mature polymorphonuclear cells. The nuclear bodies disrupted into multiple spots in leukemic cells but were reconstructed within 5 days of ATRA treatment [35]. The cell abnormalities of APL seemed to be restored by ATRA treatment—a treatment approach that was previously totally unknown for other malignancies. However, ATRA treatment led to three major unpredicted adverse effects: leukocyte activation, secondary resistance to ATRA, and thrombosis.

During the early days of treatment with ATRA, physicians were surprised by a potentially lethal syndrome, particularly in hyperleukocytic and microgranular forms of the disease. These patients experienced fever, weight gain, dyspnea due to pleural and pericardial effusions, pulmonary infiltrates, and sometimes renal failure [24]. The ATRA syndrome affected a third of patients in Western countries and Japan (but not in China). It was often preceded by an increase of white blood cells, but it was treated by early chemotherapy (European group) and high doses of corticosteroids (New York group). The syndrome was considered to be leukocyte activation related to cytokine release by the differentiating cells [36].

Using ATRA alone, very rare primary resistance to ATRA (in part attributed to rare mutations of the retinoic receptor pocket) contrasted with a short-lived complete remission (3–12 months) and a permanent secondary resistance when relapses occurred. Additional genetic defects of the retinoic receptor could not be involved because all patients acquired resistance to ATRA; the resistance was sometimes reversible after 12 months. Consequently, catabolic mechanisms were suspected and confirmed by a decrease in ATRA plasma concentrations; this was due to an increase of cytochrome P450 induced by ATRA itself, together with upregulation of the expression of cytoplasmic protein binding protein (CRABP II) depriving the nucleus of ATRA. Optimal intranuclear ATRA concentrations correlated with the range of differentiation of APL cells.

Bleeding diathesis rapidly disappeared after few days of ATRA treatment, with a concomitant decrease of primary fibrinolysis (and of extended proteolysis due to endocellular cathepsin G, elastase, and myeloblastin) [37]. The prothrombotic trend (DIC) was not counteracted by ATRA treatment, which explained why severe thrombosis occurred in some patients (particularly those receiving ATRA alone) and indicated the need for low-dose heparin in such cases.

Nonstandard Regimens

Patients with high white blood cell counts (>10 G/L) are often associated with a microgranular variant, which has a higher risk of early death due to hemorrhages or ATRA syndrome and a higher risk of relapse. Massive platelet transfusion, early chemotherapy, and high-dose corticosteroids can greatly improve the prognosis of these patients. The questions in the 1995s were the role of ARA-C in induction and consolidation (PETHEMA-GIMEMA), risk stratification, and the advantages and disadvantages of maintenance therapy. After long discussions among cooperative groups, the results of a French-Spanish collaborative meta-analysis and other studies suggested that ARA-C did not appear to be mandatory in induction therapy (mainly when idarubicin was proposed) but could be beneficial for high-risk patients, who were also most likely to need consolidation. The modalities of maintenance therapy (intermittent treatment with 6-mercaptopurine, methotrexate, and ATRA) were also a matter of controversy. Positive findings were observed by European (France, Belgium, Switzerland) and American groups but not by Italian and Japanese groups (see Chap. 8, p. 106). This could be a posteriori explained by the intensity of front-line treatment: two courses of chemotherapy by the American group, three courses of chemotherapy by the European group, and four courses of chemotherapy by the Italian group.

APL is rare in children (10% of APL cases), with a higher incidence before the age of 4 years. Pediatricians were surprised by children’s particular sensitivity to ATRA at the standard adult dose, with the occurrence of pseudotumor cerebri syndrome in 3–4% of cases. In response, ATRA was administered to pediatric patients at a half dose (25 mg/m²), which was demonstrated to be as effective as the standard adult dose (45 mg/m²).

Furthermore, ATRA seemed to abrogate any unfavorable conditions. For instance, compared to other secondary acute leukemias, patients with therapy-related APL experienced similar outcomes as patients with de novo APL when ATRA was included in the treatment scheme. Approximately 10–15% of APL cases occurred after chemotherapy or radiotherapy, mainly when they included anthracycline and/or mitoxantrone for breast cancer or multiple sclerosis. Another striking effect of ATRA was the similar survival of patients with or without other cytogenetic abnormalities that are usually considered to be poor prognostic factors for leukemia.

Molecular Defects

The specific sensitivity of APL to ATRA treatment and the genetic signature t (15;17) were intriguing. Clinicians contacted molecular biologists to answer the question.

First Attempts at Molecular Approaches

In 1988, Marie Geneviève Mattei in Montpelier [38] located the retinoic acid receptor alpha (RARA) on band q21 of the long arm of chromosome 17. One year prior, Martin Petkovitch (Pierre Chambon laboratory, Strasbourg) had cloned the RARA gene [39]. These two findings, in conjunction with the demonstration of ATRA sensitivity, prompted Laurent Degos and Christine Chomienne to go to Strasbourg and obtain the RARA probe to investigate the possible location of the RARA site translocation.

After initial attempts with Huang Meng Er (who recently moved from Shanghai) to explore RARA mRNA products in the blasts of patients with APL, the researchers contacted Hughes de Thé from Anne Dejean’s laboratory at Pasteur Institute, who had previously cloned the RAR beta, and proposed a collaboration. Together, they found abnormalities of mRNA in the RARA of patients with APL but not of patients with other types of leukemia.

Early in 1989, they submitted an article describing these mRNA results to the New England Journal of Medicine. The manuscript was rejected. The reviewers believed that the observations were due to artefacts and polymorphisms at restriction sites, and they quoted a previous article from an American team that showed no particular expression pattern of RARA mRNA in APL. In fact, the blots from the quoted article depicted the RARA abnormalities, but the figures were printed upside down.

Meanwhile, the French group cloned and sequenced the breakpoint on the RARA gene [40] using the NB4 cell line established by Michel Lanotte [41] from a patient followed by Sylvie Castaigne. The partner gene of RARA was first named myl. Two other teams conducted concomitant investigations following different pathways and reached the same conclusions: one headed by Helen Solomon [42] on chromosome 17, and the second headed by Pier Giuseppe Pelicci [43] on several potential partner genes. One year later, the partner gene of RARA in the t (15;17) was sequenced by the French group [44] and Ron Evans’s team [45]. Both groups published their results in the same issue of the journal Cell. The gene was renamed PML for promyelocytic leukemia, avoiding any confusion of myl with the myosin light chain gene. The PML-RARA fusion gene was definitively considered to be the precise signature of APL.

Clinical Consequences: Diagnostic and Follow-Up Tools

Reverse-transcription polymerase chain reaction (RT-PCR) was rapidly established in France and Italy. A European Biomed Program led by Christine Chomienne organized a series of technical workshops to standardize the tests. The GIMEMA studies demonstrated that the RT-PCR reaction after the end of consolidation therapy was negative in 95% of the patients. A conversion from a negative reaction to a positive reaction predicted a relapse, which led to an anticipated early salvage and thus improved the outcome [46]. This strategy of RT-PCR follow-up was further developed and adopted by several other investigators, including David Grimwade in the United Kingdom.

Hughes de Thé in France, Jacqueline Dick in the United States, and Brunangelo Falini [47] in Italy produced antibodies against the PML protein, allowing for a better description of nuclear bodies [35], the location of the PML protein on the outer shell of the bodies in any cell of the body, and the association of PML with several other molecules. A PML molecule involved in the structure of a nuclear body could be responsible for the oncogenesis by itself (truncated PML) and/or for the effect of disruption of the body on companion products in the clone of leukemic cells. Brunangelo Falini proposed that the disruption and rapid reconstruction of nuclear bodies be used as a diagnostic tool and as a determinant of ATRA sensitivity, respectively. The novel PML gene and protein were scrutinized. The two major breaking points leading to long and short PMLs and seven groups of protein isoforms were compared to clinical status. Some correlations but no clear stratification with the prognosis were reported (see Chap. 2, p. 18 and Chap. 11, p. 153).

Several other very rare leukemias (sometimes represented by just one single case report) were found to harbor a translocation involving the RARA gene and partner genes other than PML. Some of them were sensitive to ATRA (NPM, NuMA partners) but others were not (PLZF, Stat 5 partners). These findings helped biologists to better understand the oncogenesis and specific effects of ATRA.

Toward a Biological Explanation of Oncogenesis and Restoration

Among retinoic acid receptors, RARA was recognized by Hughes de Thé [48] as the most involved in myelopoiesis. PML-RARA transfections impaired the normal function of the RARA receptor of Cos cells, halting the differentiation of HL-60 cells usually obtained by ATRA, whereas RARA-deficient mice had normal (and increased) hematopoiesis. The paradox between the leukemia induced by PML-RARA and the normal granulopoiesis observed in RARA-deficient mice was intriguing. RARA treatment seemed to alleviate the blockage of differentiation, allowing for the effects of coactivators.

PML-RARA suppression was extensively studied by Suk Hyun Hong, Hugues de Thé, Pier Giuseppe Pelicci, Anne Dejean, Ron Evans, and other teams. RARA, a nuclear receptor (transcription factor) acts as a dimer with the retinoic X receptor (RXR) and binds co-repressors (SMRT, N-COR, and histone deacetylase) in the absence of a ligand (retinoic acid). In the presence of retinoic acid, it becomes an activator linked to histone acetylase. It was suggested that the PML-RARA product blocked the nuclear receptor at a suppressor status, but pharmacological doses of ATRA overcame the suppression (see Chap. 2, p. 19 and Chap. 4, p. 44 and p. 48). APL became a model for the study of the link between gene expressions and chromatine reshaping through deacetylation of histones and methylation of promotors. Histone deacetylation inhibitors such as trichostatin A, valproate, and demethylating agents were investigated as tools for experiments and potential anticancer drugs.

Later, Chen Zhu (who had returned to Shanghai) found more than 150 retinoic acid–induced genes (RIG) using differential display techniques on the NB4 cell line before and after ATRA treatment [49]. Pier Giuseppe Pelicci investigated interferences on other genes, such as the p53 product. Several avenues with few clear explanations were presented at the second APL meeting in November 1997. Only simple conclusions could be reached: RARA is involved in granulopoiesis, PML-RARA blocks the differentiation, and ATRA restores the normal differentiation. Furthermore, PML is involved in leukemogenesis through a structural disruption of nuclear bodies and ATRA restores the normal structure.

Arsenic, a Manchurian Traditional Medicine (1995)

During the congress of the Chinese Society of Hematology at Da Lian (1995), to which Laurent Degos and Luizo Ohno were invited, two separate groups from Harbin in Manchuria reported that complete remission of APL was obtained with 10 mg/day of purified arsenic trioxide (ATO) given intravenously. The first trial was started in 1971 and included 60 patients—30 with de novo and 30 with relapsed disease; of these, 73% and 53% experienced complete remission, respectively. The second trial enrolled 72 patients; in this trial, 73% of the 30 patients with de novo and 52% of the 32 patients with relapsed disease obtained complete remission. A previous article from one of the groups, published in a relatively unknown Chinese journal [50], reported that 31 of the 42 patients treated with the AL-1 Harbin drug since 1971 achieved complete remission.

A Second Era for Clinicians

Chen Zhu’s team in Shanghai used ATO with the so-called AILING I injection, manufactured by the First Clinical Medical College of Harbin Medical University (Fig. 1.4), to treat 10 patients with APL who relapsed after ATRA and chemotherapy. Of these patients, nine obtained complete remission. The only nonresponder lost the t(15;17) in malignant cells at relapse. They demonstrated differentiation and apoptotic effects at low and high concentrations, respectively [51].

Fig. 1.4

Original Chinese ATO drug

Arsenic (Fowler liquor) had been designated since 1890 as an antileukemic drug in medical textbooks in Europe, with some rare but good outcomes. The spectacular effect of ATO in APL was first confirmed at MSKCC and then by several groups in Europe and Japan. Similarly to ATRA therapy, bleeding diathesis disappeared rapidly with ATO, eliminating not only the primary fibrinolysis but also the DIC. Like ATRA, ATO induced an ATRA syndrome consisting of leukocyte activation. Also like ATRA but with faster kinetics, the PML-RARA transcript was cleared.

ATO was rapidly manufactured (at a high price) in the United States and was used by many groups for relapsed patients to induce a second complete remission. The question at that time was whether to introduce ATO in first-line regimens. Some teams combined ATO with ATRA and chemotherapy during consolidation treatment and later during induction treatment to reduce the amount of chemotherapy needed; their results were at least similar to those obtained with classical ATRA–chemotherapy regimens. In other countries (mainly China, India, and Iran, who were manufacturing their own drugs at low prices), ATO was used alone, yielding complete remission rates of approximately 90% and long-term survival rates of approximately 70%. Thus, a second effective drug was available for patients with APL. However, Chinese researchers, who had acquired great experience with ATO, found that it was highly toxic for the liver in de novo patients when given in association with ATRA.

A Better Biological Understanding

Chen Zhu, in collaboration with Arthur Zelent, identified the first translocation that included the RARA gene with a different partner, namely PLZF [52]. The resistance of this rare disease to ATRA treatment was investigated by several groups. It was determined to be a result of a second binding to histone deacetylase through PLZF. For an unknown reason, the disease was sensitive to the association of ATRA plus a granulocyte colony-stimulating factor.

Animal models were used to gain a better understanding of the effect of the two drugs, ATRA and ATO (see Chap. 4). RARA knockout mice were viable, with no obvious defect in myelopoiesis. This was explained by the natural gene repression of RARA in the absence of a ligand. The absence of the receptor not only allowed for but also accelerated myelopoiesis . Italian scientists also aimed to reproduce APL disease in mouse models. PML-RARA transgenic mice were generated by Pier Giuseppe Pelicci and other groups using regulatory elements of the gene in promyelocytes, mainly cathepsin G or hMRP8. The mice developed ATRA-sensitive leukemia after a a long preleukemic state. Pier Paolo Pandolfi produced several transgenic mice (including transgenic, double transgenic, and knockout mice) by inserting not only PML-RARA or RARA-PML but also PLZF-RARA and other X-RARA fusion genes. Whatever the partner of RARA was, the mice developed a form of leukemia.

Using an APL model by transplanting spleen cells from APL transgenic mice, Valérie Lallemand and Hughes de Thé [53] demonstrated that ATRA and ATO had beneficial effects on induced APL in mice, but also that the combination of ATRA and ATO eradicated the disease. These findings gave hope for a definitive cure of the disease, encouraging clinicians to overcome the threats of the toxicity of ATRA-ATO combination previously described by Chinese investigators. A meeting jointly organized in October 2001 by Francesco Lo Coco and Samuel Waxman for the third symposium on APL, a Curable Disease chaired by Franco Mandelli and the 9th international conference on differentiation therapy led by Samuel Waxman envisaged new avenues for treatments, not only for patients with APL but also for patients with other malignancies using cell modifier agents.

ATRA-ATO Combination in Patients: From Curable to Cured APL

The next achievements are described in each chapter of this book. To summarize, in the 1991–2001 decade, international cooperative groups improved the outcomes of patients with APL using a combination of ATRA and chemotherapy by refining therapy though the addition or omission of maintenance treatment, grading the disease in high- and low/intermediate-risk groups of patients, adapting treatment to these criteria, modulating the chemotherapy for elderly patients, and decreasing the ATRA dose for children. Trials for relapsed patients also prompted to the implementation of worldwide cooperative groups due to the rarity of this event.

The synergic action of ATRA and ATO was further demonstrated by collaborative research between Shanghai and Paris, which showed that arsenic targeted PML moiety, whereas ATRA acted on the RARA moiety of the PML-RARA fusion gene. Several complementary (or controversial) studies led to the conclusion that the PML-RARA oncogenic product is degraded (and eliminated) through an ubiquitination at the RARA end by ATRA and through a sumoylation at the PML end by ATO.

The simultaneous use of ATRA and ATO in the first-line treatment of patients with APL in nonrandomized trials was previously approached by Chen Zhu (China) and Elihu Estey (United States), who demonstrated a synergistic effect in inducing prolonged complete remissions. Elihu Estey first proposed this combination as an alternative to chemotherapy for newly diagnosed patients and published excellent outcomes using a protocol that completely omitted chemotherapy.

The GIMEMA group, led by Francesco Lo Coco, conducted a randomized trial in 2006 in collaboration with German cooperative groups AMLSG and SAL [54], which challenged the conventional use of ATRA and idarubicin (AIDA) against the ATRA-ATO scheme designed by Estey [55]. The results demonstrated the superiority of the latter approach and ultimately the extraordinary achievement of the ATO-ATRA venture: APL is a malignant disease cured by cell-modifying agents without any chemotherapy. The GIMEMA-SAL-AMLSG study published in July 2013 in the New England Journal of Medicine and another independent randomized study conducted by the Medical Research Council led by Alan Burnett and presented at the 6th APL Symposium in Rome (October 2013) clearly showed that, at least in patients with low- to intermediate-risk APL, ATRA-ATO was better than ATRA-anthracycline-based chemotherapy. Nearly 100% of patients were event free at 2 years, without notable toxicity [56].

However, despite these compelling data, ATO was still not available for first-line treatment in Western countries. Regulatory bodies (mainly the European Medicines Agency, or EMA) did not provide market approval because the trials were conducted by academic groups and not by the industrial companies manufacturing the drug. In February 2015, a common letter signed by all European thought leaders and chairs of large cooperative groups on APL treatment was sent to the EMA, requesting a solution to this unfavorable situation. Hematologists needed drug approval for first-line therapy of APL without having to conduct further randomized trials, which would be unethical given the compelling results published in randomized studies. The EMA reacted with an open-minded attitude and proposed a meeting to discuss the issue with all involved partners and stakeholders (regulatory experts, manufacturing companies, thought leaders, and patient advocacy representatives) in London in July 2015. Solutions were explored and the way was paved toward a solution.

And Beyond: Lessons and Questions

History has demonstrated that a cure for APL using cell modifiers was not obtained by chance. Rather, it resulted from a long-term aim to modify malignant cells. Collaboration between several institutes of hematology around the world counteracted the obstacles. Clinical and biological findings were achieved by academic collaborations, often against the strategies of companies. The APL story is a model for international academic collaboration, both in clinical and laboratory investigations.

ATRA and ATO, the two cell-modifying agents, were neither promoted nor defended by companies. They were not actually recognized as innovative drugs, or even considered as nonconventional drugs. They were often the subject of controversies and discussions that delayed their use in patients. Conversely, when Gleevec, another target therapy, appeared 10 years later in the market, the support of the company facilitated the approval.

Today, the outcomes of patients with APL around the world are partly attributable to treatment choices and the availability of clinical facilities, but mostly attributable to rapid diagnosis and the delivery of appropriate treatment. In fact, the risk of early mortality before the initiation of ATRA treatment still persists and is the main obstacle to an APL cure. Some countries are educating physicians (including general practitioners and residents) and laboratories to shorten the time between the first symptom and hemogram, between the hemogram and myelogram, and between the myelogram and treatment. APL outcomes are related to the healthcare facility, so the differences in developed countries (early mortality rates before treatment ranging from 5% to 20%) should be further investigated and improved.

The purpose of ATRA and ATO targeted treatment is unique compared with the other targeted drugs used for cancer therapy (e.g., the series of antikinases). ATRA and ATO modify and restore normal transcription activities, whereas antikinases are generally antienzyme agents that inhibit hypersignals of transduction. The pathway to transcription normalization is still unclear: Is it the degradation of the oncogenic PML-RARA product, histone reshaping, or a mix of both mechanisms? Concerning the PML protein, what are the respective roles of PML and its partners on the nuclear body? Which metabolic pathway is modified (P53, others) and restored? Do the effects of ATO mainly consist of differentiation, restoration of apoptosis, normal self-renewal, or normal senescence? What is the role of the proteasome? Would a proteasome inhibitor block the effects of ATRA and/or ATO? What is the pathway between sumoylation and degradation of the product, and is it a degradation or an autophagy? So many question remain without definitive answers.

It is also unclear why one series of ATRA plus ATO is enough to definitively cure patients, whereas Gleevec (and other target antikinase drugs) are often administered for the rest of a patient’s life. Does it eradicate the clone and the stem cell upstream to which the clone is derived? Why does the change from co-repressor to co-activator remain for life? How could degradation of the PML-RARA product erase a clone? Is it due to an immunological effect, like a vaccine (which was investigated by Rose Ann Padua)? What is the immunogenic product? Could a relapse occur in cases of immune deficiency?

More intriguing facts are as yet unexplained. When ATRA or ATO are administered, additional cytogenetic abnormalities do not affect prognosis, as commonly occurs in other leukemias. Is the differentiation of malignant cells the major defect to be treated, whatever the other events may be? If so, should we abandon the other avenues to cure patients that target, for instance, proliferation? Except for Gleevec in chronic myeloid leukemia , where cells are already well differentiated, any targeted antiproliferative treatment (i.e., antikinases) has delayed the progression of disease and patients were not cured. Although few studies have focused on the restoration of normal transcription and normal differentiation (probably due to the difficulty of manipulating transcription factors), the APL story should encourage researchers to move in