Chapter 1: Acute Lymphoblastic Leukemia

Acute lymphoblastic leukemia (ALL) is characterized by the proliferation and accumulation of lymphoid progenitor cells in the blood, bone marrow, and other tissues. It has a bimodal distribution. The overall age-adjusted incidence is 1.7 per 100,000 persons, but ALL affects 4 to 5 per 100,000 persons during age 4 to 5 years and half that number around the fifth decade of life. Approximately 60% of cases are diagnosed in patients ≤20 years old, with a median age at diagnosis of 14 years. In 2014, the American Cancer Society estimated that approximately 6,000 individuals would be diagnosed with ALL that year (^1,2). Acute lymphoblastic leukemia represents 20% of adult leukemias but is the most common childhood acute leukemia, representing approximately 80% of cases (^1,2).

The etiology of ALL is unknown in most cases (^3,4,5,6,7). Chromosomal translocations occurring in utero during fetal hematopoiesis have suggested genetic factors as the primary cause for pediatric ALL and postnatal genetic events as secondary contributors. Monozygotic and dizygotic twins of patients with ALL and individuals with genetic disorders, such as Klinefelter (XXY and variants) and Down (trisomy 21) syndromes, or inherited diseases with excessive chromosomal fragility, such as Bloom syndrome, Fanconi anemia, and ataxia telangiectasia, have all been found to have higher incidence of ALL, implicating a possible genetic predisposition. Additional studies have postulated infectious etiologies (⁴). Human T-cell lymphotropic virus type-1 is known to cause adult T-cell leukemia/lymphoma (⁵); Epstein-Barr virus has been associated with lymphoproliferative disorders, including Burkitt lymphoma and mature B-cell ALL (⁶); and varicella has been linked to childhood ALL (⁷).

Presenting symptoms can be nonspecific, particularly in children. They largely reflect bone marrow failure and include malaise, fatigue, bleeding or bruising, and secondary infections. The B symptoms, such as fever, night sweats, and weight loss, are frequent. White blood cell (WBC) count at presentation varies widely, and circulating blasts are generally noted. Symptoms related to hyperleukocytosis are rare in ALL, given the lymphoblast morphology, even when WBC counts are high.

Leukemic involvement of the central nervous system (CNS) ranging from cranial neuropathies to meningeal infiltration occurs in <10% of patients at presentation. It is more common in mature B-cell ALL (Burkitt leukemia) (⁸). A history or findings of abdominal masses, significant spontaneous tumor lysis syndrome, and chin numbness (mental nerve) indicating cranial nerve involvement are also more common in this subtype of ALL (⁹). Lymphadenopathy and hepatosplenomegaly, although rarely symptomatic, are noted in approximately 20% of patients (⁹).

Immunophenotyping

The diagnosis of ALL is largely based on flow cytometric immunophenotyping, although identification of cytogenetic-molecular abnormalities plays a significant role (Fig. 1-1). The World Health Organization (WHO) proposed new guidelines for the diagnosis of neoplastic diseases of hematopoietic and lymphoid tissues (¹⁰). The French-American-British (FAB) Cooperative Group diagnostic approach, which recognizes L1 to L3 morphologic subtypes, has been essentially abandoned. A blast count of ≥20% was established as sufficient for diagnosis.

Flow cytometric analysis successfully assigns lineage in more than 95% of cases. True mixed phenotype acute leukemia is rare (¹¹). Concomitant expression of markers from more than one lineage is seen in 15% to 50% of adult and 5% to 35% of pediatric ALL (^12,13,14), but this is not prognostically relevant. Targeted genomic profiling may further define ALL subtypes with different response profiles to therapy and prognoses, which are only partially discriminated by current diagnostic tools.

Immunophenotypically, ALL blasts are negative for myeloperoxidase (MPO), although low-level MPO positivity (3%-5%) may occur in rare cases that otherwise lack expression of myeloid markers by flow cytometry (¹⁵). Terminal deoxynucleotidyl transferase (TdT), although not a specific marker of ALL, helps separate malignant lymphocytosis from reactive processes and distinguish L3 ALL (TdT negative) from other ALL subtypes (¹⁶).

Both the prior FAB and current WHO classification systems rely heavily on morphologic assessment (¹⁷), which accounts for cell size, cytoplasm, nucleoli, basophilia, and vacuolation. The former FAB L3 morphology, characterized by a high rate of cell turnover, is associated with mature B-cell ALL (Burkitt leukemia) and gives rise to the “starry sky” pattern on marrow biopsies.

Three broad immunophenotypic ALL groups can be distinguished: precursor B-cell, mature B-cell, and T-cell ALL (Table 1-1). Precursor B-cell ALL (B-ALL) stains positive for TdT, HLA-DR, CD19, and CD79a. According to the stages of maturation, further B-cell subgroups have been defined as pre-pre-B-ALL (pro–B-ALL), common ALL, and pre–B-ALL. Although they all stain positive for CD19, CD79a, or CD22, expression of CD10 (common ALL antigen [CALLA]) distinguishes common ALL (early pre–B-ALL), and cytoplasmic immunoglobulins with or without CD10 identify pre–B-ALL. Mature B-ALL (Burkitt leukemia) is TdT negative but expresses surface immunoglobulins (usually immunoglobulin M), as well as κ or λ light chains in a clonal fashion. It has almost ubiquitous expression of CD20, which has therapeutic implications (¹⁸).

T-cell ALL (T-ALL) further stratifies into subtypes based on different stages of thymic differentiation (¹⁹). As the most lineage-specific marker for T-cell differentiation, surface CD3 (sCD3) is typically positive in mature T-ALL, which is also positive for either CD4 or CD8, but not both. However, pre–T-ALL is negative for CD4, CD8, and sCD3 but may still express cytoplasmic CD3. A more simplified classification divides T-ALL into early T-ALL (sCD3^–, CD1a^–), thymic T-ALL (sCD3^+/–, CD1a⁺), and mature T-ALL (sCD3⁺, CD1a⁺). Only thymic T-ALL has excellent outcome with chemotherapy alone.

Cytogenetic-Molecular Profiling

Frequent cytogenetic and molecular abnormalities associated with adult ALL offer insight into the events leading to leukemic progression (Table 1-2) (²⁰). They are of both prognostic and predictive significance and have varying frequencies in children and adults, which explains some of the differences in outcomes in these two groups. This is particularly true in the case of ALL harboring Philadelphia chromosome [t(9;22)] (Ph) or other chromosomal changes with prognostic relevance such as Burkitt karyotypes [t(8;14), t(2;8), t(8;22)] or t(4;11). Next-generation sequencing, expression proteomics, and oligonucleotide microarrays have transformed our understanding of the genomic landscape of ALL and are yielding new molecular subgroups with actionable defects (^21,22,23).

Recently, a Ph-like signature in 10% of children with standard-risk ALL and as many as 25% to 30% of young adults with ALL has been defined using genome-wide gene expression arrays. This subgroup lacks the expression of BCR-ABL1 fusion protein but does have a gene expression profile similar to BCR-ABL1 ALL (^24,25,26). The vast majority of these patients have deletions in key transcription factors involved in B-cell signaling, such as IKZF1, TCF3, EBF1, PAX5, and VPREB1, as well as kinase-activating alterations involving ABL1, ABL2, CRLF2, CSF1R, EPOR, JAK2, NTRK3, PDGFRB, PTK2B, TSLP, or TYK2 and sequence mutations involving FLT3, IL7R, or SH2B3. The most common alterations (~50%) are rearrangements of CRLF2, which activate downstream signaling through JAK kinases, and approximately half of these cases have activating mutations in JAK1 or JAK2 (Fig. 1-2). Importantly, patients with ABL1, ABL2, CSF1R, and PDGFRB expression fusions were sensitive in in vitro and in vivo human xenograft models to ABL class tyrosine kinase inhibitors (TKIs; eg, dasatinib); rearrangements in EPOR, IL-7R, and JAK2 mutations and fusions were sensitive to JAK kinase inhibitors (eg, ruxolitinib); and patients with ETV6-NTRK3 fusion were sensitive to ALK kinase inhibitors (eg, crizotinib) (²⁵), further expanding therapeutic options in this subgroup with poor outcome.

Observations of epigenetic alterations regulating distinct molecular pathways that occur frequently at presentation and relapse have identified a “hypermethylator” phenotype of ALL (²⁷). These patients may respond favorably to treatment with hypomethylating agents (azacitidine or decitabine). Identification of these and other molecular and cytogenetic changes in adult ALL drives the development of risk-adapted and targeted therapies, particularly in high-risk groups (Table 1-3) (²⁸).

Therapy for ALL consists of complex and comprehensive regimens consisting of several phases: induction, intensified consolidation, maintenance, and CNS prophylaxis (^9,29). Each involves the use of a core group of agents considered the backbone of therapy in a time- and dose-dependent manner, with a goal of restoring normal hematopoiesis, eradicating resistant subclones, providing adequate prophylaxis of sanctuary sites (eg, CNS, testicles), and eliminating minimal residual disease (MRD) during the consolidation and maintenance phases (^9,30). Combining anthracyclines (eg, daunorubicin or doxorubicin), vincristine, and dexamethasone (for better CNS penetration), often coupled with cyclophosphamide or asparaginase with growth factor support, represents the cornerstone of ALL induction regimens. This results in complete remission (CR) rates of 70% to 90% and median remission durations of 18 months (^30,31). Patients who achieve CR subsequently transition to the consolidation phase, which, depending on the risk-oriented subtype, may consist of consolidation chemotherapy (cytarabine, methotrexate, cyclophosphamide, and 6-mercaptopurine) or allogeneic hematopoietic stem-cell transplantation (AHSCT). Consolidation is followed by prolonged maintenance therapy with daily 6-mercaptopurine, weekly methotrexate, and monthly pulses of vincristine and prednisone or dexamethasone, given over 2 to 3 years (POMP or DOMP, depending on corticosteroid used) (^30,31,32). Maintenance, which is omitted in mature B-ALL due to high cure rates, may also involve the use of TKIs for patients with Ph-positive ALL. Primary CNS involvement at diagnosis is rare (<10%) but is as high as 50% to 75% at 1 year without prophylactic administration of intrathecal chemotherapy (IT) (³¹). Although high-dose cytarabine (1-7.5 mg/m²) and methotrexate (5-8 g/m²) successfully penetrate the blood-brain barrier, they are too toxic to serve as the sole CNS prophylaxis. The inclusion of IT prophylaxis (methotrexate, cytarabine, liposomal cytarabine, hydrocortisone, or thiotepa) reduces the incidence of CNS relapse to 4% by allowing sustained therapeutic concentration of the agents in the cerebrospinal fluid. The number of ITs varies according to protocol (usually 8 for standard risk, 12 for Ph positive, and 16 for Burkitt), and in rare cases of extramedullary disease spread (eg, masses or chloromas), IT may even be supplemented by radiation therapy.

One extensively studied regimen used in treatment of adult ALL is the hyper-CVAD (HCVAD) regimen, where patients receive hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with high-dose methotrexate and cytarabine for a total of eight alternating cycles approximately every 3 to 4 weeks (Table 1-4) (^30,31). This is followed by 2 years of POMP maintenance therapy, interspersed with intensification courses during months 6, 7, 18, and 19. The number of IT injections (two per course) depends on the risk of CNS relapse, which has been identified as high for patients with mature B-ALL. Our current approach is giving 8 ITs for nonmature B-ALL and 16 ITs for mature B-ALL, resulting in a 5-year overall survival (OS) between 38% and 50% (³⁰). Due the improved cure rates of Ph-positive ALL patients, an increase in the CNS relapse rate was observed, which is the reason the protocol was modified to include 12 ITs for Ph-positive ALL.

Mature B-Cell and Burkitt Acute Lymphoblastic Leukemia

The addition of rituximab to short intensive chemotherapy has also improved outcome in adult Burkitt and Burkitt-type lymphoma or ALL (^29,33,34). Hoelzer and colleagues have recently reported the benefit of adding rituximab to short intensive chemotherapy in 363 patients with Burkitt lymphoma/leukemia; the addition of rituximab resulted in CR and 5-year survival rates of 88% and 80%, respectively (³³). Higher rates of survival were reported in adolescents compared to adults and elderly patients (90% vs 84% vs 62%, respectively) (^33). Low-intensity chemotherapy with infused etoposide, doxorubicin, and cyclophosphamide with vincristine, prednisone, and rituximab (EPOCH-R) was recently tested in 30 adult patients with Burkitt lymphoma (³⁵). The progression-free survival (PFS) and OS rates were 90% and 100%, respectively. Of note, marrow involvement was present in only 13% of patients, and CNS involvement was present in only 3% of patients (³⁵).

CD20-Positive Pre–B-Cell Acute Lymphoblastic Leukemia

There have been several alterations to traditional protocols with further refining of the disease. Expression of cell surface marker CD20 in adult ALL ranges from 35% to ubiquitous depending on the subtype and has been associated with an inferior prognosis (¹⁸). The addition of two doses of monoclonal CD20 antibody (rituximab) administered with the first four cycles of chemotherapy and during maintenance intensification at months 6 and 18 resulted in improved OS in younger patients compared with similar chemotherapy historical controls (75% vs 47% at 3 years; P = .003) (³⁶). Improvement in the 5-year remission duration and survival rates was also reported in patients <55 years old by the German Multicenter Study Group for ALL (GMALL) when rituximab was added to standard induction and consolidation therapy (³⁷).

Ofatumumab is a more potent second-generation anti-CD20 monoclonal antibody that binds to a membrane proximal small-loop epitope on the CD20 protein. A phase II study in CD20-positive pre–B-ALL combined ofatumumab with HCVAD during induction, resulting in a 96% rate of both CR and MRD negativity. At a median follow-up of 14 months, the 1-year PFS and OS rates were 94% and 92%, respectively (³⁸).

Philadelphia-Positive Acute Lymphoblastic Leukemia

Philadelphia-positive ALL used to have a very poor outcome in general. The incorporation of TKIs into treatment regimens has significantly improved patient outcomes, as supported by several reports (^39,40,41,42). Incorporation of early, daily, and concurrent TKI with chemotherapy has proven more effective than intermittent pulses (^41,42).

Second-generation TKIs, such as the dual src and abl inhibitor dasatinib, which is more potent than imatinib and crosses the blood-brain barrier (⁴³), have also been investigated in combination with chemotherapy. In an attempt to improve on the outcomes with imatinib, dasatinib was administered at 100 mg daily for 14 days with induction chemotherapy, followed by 70 mg continuous dosing with the consolidation cycles, and at 100 mg daily continuously during the maintenance phase (⁴⁴). Overall, 94% of patients achieved CR, 96% achieved complete cytogenetic response (CCyR), and 65% achieved complete molecular response (CMR). Allogeneic hematopoietic stem-cell transplantation was performed in 22 patients (12 in first CR and 10 in second CR), with 3-year disease-free survival (DFS) and OS rates of 49% and 61%, respectively.

Attempting to reduce exposure to cytotoxic chemotherapy by intensifying chemotherapy with TKIs can be very effective but toxic (^45,46). Patients in the GRAAPH-2005 study were randomized to imatinib 800 mg daily for 4 weeks combined with weekly vincristine and dexamethasone versus imatinib 800 mg daily for 2 weeks combined with HCVAD chemotherapy (⁴⁵). The CR rate was higher in the low-intensity group due to induction-related mortality in the HCVAD group (7% vs <1%; P = .01). An equal number of patients in each group proceeded to autologous stem cell transplantation and allogeneic stem cell transplantation, and at 3 years, OS was similar between the two arms (53% for low intensity vs 49% for HCVAD; P = .61).

Studies have also evaluated the use of dasatinib and nilotinib with low-intensity chemotherapy (^46,47,48). In the EWALL-Ph-01 study, dasatinib with low-intensity chemotherapy was administered to 71 patients with newly diagnosed Ph-positive ALL age ≥55 years (⁴⁶). Dasatinib was dosed at 140 mg once daily during induction and at 100 mg daily during consolidation, yielding a CR rate of 94%. The estimated 3-year OS was 45%.

Many Ph-positive ALL patients can relapse with threonine-to-isoleucine mutation at position 315 (T315I), which is refractory to imatinib and second-generation TKIs. A third-generation TKI, ponatinib, which has activity against T315I, was evaluated in phase I and II trials in patients with Ph-positive leukemias and was shown to have significant antileukemic activity (^49,50). More recently, 39 patients with newly diagnosed Ph-positive ALL were treated with HCVAD and ponatinib 45 mg daily for 14 days during induction and then continuously thereafter until CCyR and CMR were obtained, when decreases to 30 mg and 15 mg daily could be instituted, respectively. The CR, CCyR, and CMR rates were 100%, 100%, and 74%, respectively. After a median follow-up of 20 months, 1-year PFS and OS were 97% and 87%, respectively (⁵¹).

Although current standard of care still advocates AHSCT consolidation in first CR (³⁹), new information regarding the status of MRD in Ph-positive ALL has raised a question as to who should be referred for it. The predictive value of MRD assessment by quantitative reverse transcriptase polymerase chain reaction (RT-PCR) and multiparameter flow cytometry (FCM) was recently assessed in patients with Ph-positive ALL treated with combination chemotherapy and TKIs who did not undergo AHSCT. Achieving major molecular response at 3, 6, 9, and 12 months (P = .02, .04, .05, and .01, respectively) and having negative FCM at 3 and 12 months were associated with improved survival (P = .04 and .001, respectively) (⁵²). This information suggests that patients with early and sustained molecular response may not need consolidation with AHSCT.

T-Cell Acute Lymphoblastic Leukemia

Treatment of adult T-ALL and T-cell lymphoblastic lymphoma (T-LL) results in a long-term survival rate of 40% to 60%, and the outcome is strongly associated with T-cell phenotype (^53,54). Adding nelarabine, a selective anti–T-ALL agent may further improve the outcome. In a single-arm, phase II study, 48 patients with newly diagnosed T-ALL or T-LL were treated with HCVAD and neralabine (⁵⁵). The CR rate was 93%; the 5-year survival rate was 66% after a median follow-up of 41 months. These rates were 38% and 70% for patients with early T-cell precursor (ETP) and mature T-ALL, respectively. Indeed, ETP-ALL is a distinct T-cell entity characterized by the absence of CD1a, sCD3, and CD8 expression; weak CD5 expression; and expression of one or more myeloid or stem cell–associated markers (⁵⁴). It confers poor prognosis with the use of standard intensive chemotherapy, which results in high rates of remission failure and relapse compared to patients with typical T-ALL (72% at 10 years vs 10% at 10 years). This phenotype is in part a reflection of the higher degree of genomic instability (number and size of genetic defects) that ETP-ALL harbors, with over 60% of adult patients carrying mutations in DNMT3A, FLT3, or NOTCH1, which may allow for tailored induction regimens with targeted therapies (⁵⁶). Following induction, AHSCT should be considered in first remission for all ETP-ALL patients.

Adolescent and Young Adult Acute Lymphoblastic Leukemia

Retrospective studies have shown that pediatric regimens resulted in better outcomes than adult regimens (which had deviated significantly from the established principles of ALL therapy in pediatric regimens). Pediatric-inspired regimens, such as the Berlin-Frankfurt-Münster (BFM) regimen (Table 1-5), deliver more intensive nonmyelosuppressive agents like vincristine, asparaginase, corticosteroids, and CNS therapy (^54,55).

The Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL) evaluated a pediatric-inspired regimen in patients up to age 60 years and compared the results to a historical control group treated with an adult regimen. In patients treated with the pediatric-inspired regimen, the CR rate was 93%, and at 42 months, event-free survival (EFS) and OS rates were 55% (95% CI, 48%-52%) and 60% (95% CI, 53%-66%), respectively (⁵⁷). Although the pediatric-inspired regimen resulted in improved survival compared with the control (66% vs 44%; P < .001), the cumulative incidence of treatment-related death in patients age 40 to 60 years old was 23%, erasing the margin of benefit gained with enhanced activity of pediatric regimens. Thus, the toxicity threshold can be reached and crossed in the adult population in attempts to reach higher cure rates, limiting the usefulness of intensifying chemotherapy to the pediatric-inspired strength.

The UKALL14 study of 91 adults with a median age of 47 years (range, 25-65 years) used PEG-asparaginase at a dose of 1,000 units/m² on days 4 and 18 during induction, resulting in a CR rate of 66%, with induction-related mortality rate of 20% and hepatotoxicity rate of 56%, prompting the omission of PEG-asparaginase in patients ≥40 years old (⁵⁸).

A recent US Intergroup study of 318 adolescent and young adult (AYA) patients (median age, 24 years) treated with a pediatric-inspired regimen was reported. With a median follow-up of 28 months, the estimated 5-year EFS and OS rates were 45% and 55%, respectively (⁵⁹). Presence of MRD at day 28 following initiation of induction therapy and presence of a Ph-like gene expression signature were significantly associated with worse EFS and OS. The Ph-like signature, which was detected in 28% of patients, resulted in 2-year EFS of only 52%, compared to 81% for those without Ph-like disease.

Our internal review of 85 AYA patients up to age 40 (median age, 21 years) treated with pediatric-like augmented BFM showed CR and MRD negativity rates of 94% and 69%, respectively (⁶⁰). The 5-year CR rate was 58%, and the 5-year OS rate was 62%. Compared with a historical control group of similar patients who received HCVAD with or without rituximab, 3-year OS rates were 72% and 71%, respectively. However, in patients age 25 years and older, the pediatric-inspired regimen was inferior and caused more liver dysfunction, pancreatitis, osteonecrosis, and thrombosis compared to HCVAD with CD20-targeted therapies.

Hence, HCVAD-based regimens that use the backbone ALL agents but eliminate or reduce the exposure to asparaginase show similar CR and remission rates and survival outcomes compared with the pediatric-inspired regimens in similar patient populations.

Acute Lymphoblastic Leukemia in Elderly Patients

Conventional ALL chemotherapy is associated with high mortality rates (30%-35%) during consolidation/maintenance in elderly patients (>60 years) (⁶⁰). A low-intensity regimen may improve outcome. In a phase II study with inotuzumab ozogamicin and low-intensity hyper-CVD therapy, 26 patients with a median age of 67 years (range, 60-79 years) were treated for newly diagnosed ALL (⁶¹). Inotuzumab, which is a CD22-directed monoclonal antibody bound to calicheamicin (chemotoxin), was administered at a dose of 1.3 to 1.8 mg/m² once with each of the first four courses; doxorubicin was eliminated in induction; cyclophosphamide and steroids were 50% reduced; methotrexate was reduced to 250 mg/m² on day 1 and cytarabine to 0.5 mg/m² × 4 on days 2 and 3 of even courses. The overall response rate (ORR) was 96% (CR, 79%; CR with incomplete platelet recovery [CRp], 17%), with all patients with cytogenetic abnormalities achieving CCyR. All responders also achieved MRD-negative status, 75% of which occurred after cycle 1. The 1-year PFS and OS rates were 86% and 81%, respectively. The 1-year survival rate was superior to previous results obtained with HCVAD with or without rituximab in similar patient populations (1-year OS, 81% vs 60%, respectively).

Role of Allogeneic Stem Cell Transplantation

Allogeneic stem cell transplantation has traditionally been reserved for patients with high-risk features including B-lineage with WBC ≥30 × 10⁹/L, T-lineage with WBC ≥100 × 10⁹/L, hypodiploid, Ph-positive, or mixed-lineage leukemia translocation ALL [eg, t(4;11)]. However, there has been some debate regarding who should be referred for AHSCT in first CR based on recent data that indicate that patients with standard-risk disease, and not high-risk disease, benefit the most (⁶²). As an alternative, many centers have incorporated MRD via FCM or RT-PCR after induction or consolidation to stratify patients based on their response to chemotherapy (⁶³). In fact, when controlled for other known risk factors, failure to achieve MRD has emerged as a powerful indicator of future relapse (⁵²) and therapeutic approach (ie, AHSCT vs more chemotherapy). In one study, MRD status at various time points after CR was used to guide treatment in adult patients with ALL (⁶⁴). Patients who remained MRD positive at the end of consolidation were deemed to be higher risk and underwent AHSCT instead of receiving prolonged maintenance therapy. Patients who achieved MRD-negative status had a significantly improved 5-year OS (75% vs 33%; P = .001). Furthermore, in a recent update of the GRAALL experience in 423 younger adults with Ph-negative ALL in first remission (265 B-cell precursor ALL and 158 T-ALL patients), postinduction MRD level ≥10⁻⁴ and unfavorable genetic characteristics (ie, MLL gene rearrangement or focal IKZF1 gene deletion in B-cell precursor ALL and no NOTCH1/FBXW7 mutation and/or N/K-RAS mutation and/or PTEN gene alteration in T-ALL) were independently associated with worse outcome (⁶⁵). Therefore, for patients with standard-risk ALL, MRD status should guide the postremission therapy, whereby patients who fail to achieve MRD negativity can be transplanted in first CR. In addition to the MRD status, new genomic and immunophenotyping technologies were essential in identifying patients with poor prognosis. Patient with Ph-like ALL and ETP-ALL should be considered for AHSCT in first CR (^24,54).

Minimal Residual Disease

Postinduction assessment for persistence or reemergence of MRD in patients with ALL is the most important adverse prognostic factor and identifies chemorefractory disease (^64,66,67). Virtually all adults with ALL and molecular failure exhibit poor prognosis despite continued chemotherapy and are candidates for stem-cell transplantation and targeted therapies (⁶⁸).

The Programa Español de Tratamientos en Hematología (PETHEMA) ALL-AR-03 trial in adolescent and adult patients with high-risk Ph-negative ALL showed poor MRD clearance by FCM after early consolidation and identified the pattern of MRD clearance as the only prognostic factor for DFS and OS (⁶⁹).

Blinatumomab is a bispecific T-cell engaging (BiTE) antibody and is the first agent in its class that engages host T cells to the target cell surface antigen–expressing cancer cells. It contains the variable domains of a CD19 and a CD3 antibody, joined via nonimmunogenic linker (⁷⁰). Cytotoxic T cells are activated upon binding to CD19, inducing cell death through the perforin system. Given the pharmacokinetics of the construct (short half-life and the mechanism of action), continuous infusion over several weeks resulted in significantly improved drug activity in ALL and minimization of side effects. Twenty-one patients in hematologic and morphologic CR with persistent or reappearing MRD during consolidation were first studied with a blinatumomab dose of 15 μg/m²/d as a continuous infusion for 28 days every 6 weeks for a total of 4 cycles or proceeded to AHSCT if a donor was available (⁷¹). Minimal residual disease conversion following one cycle of therapy was seen in 80% of patients. After a median follow-up of 33 months, 60% of patients remained in CR, with the same percentage of patients experiencing estimated 3-year refractory-free survival (⁷²). Nontransplanted patients had a similar favorable outcome compared to the nine patients who underwent AHSCT.

A recent confirmatory, open-label, multicenter, phase II trial of blinatumomab in 116 patients with MRD-positive B-cell precursor ALL in CR resulted in overall MRD eradication in 80% of patients, almost all (78%) after a single cycle of therapy (⁷³). As a result, the MD Anderson Cancer Center (MDACC) will open a trial evaluating the role of blinatumomab in patients with positive MRD in CR.

The prognosis of adult patients with relapsed ALL remains poor, with limited effective therapies available. Relapsed disease carries a median survival of only 24 weeks, and patients who have short duration of first CR or primary refractory disease do particularly poorly, with a median OS of less than 5 months (⁷⁴). The goal of salvage therapy is to reinduce CR and consolidate with an AHSCT. Given that we currently lack agents or regimens that can singularly achieve cure in the relapsed setting, patients should be enrolled on a clinical trial if possible. Choice of salvage is contingent upon the previous treatment history, remission duration, ongoing comorbidities, and the relapse-specific features that may be targetable.

Asparaginase could be incorporated into salvage for patients without previous exposure to it. This can be achieved through the augmented HCVAD protocol designed at MDACC, which intensifies the standard vincristine and corticosteroid backbone and adds the pegylated asparaginase (⁷⁵). In an initial study, 80% of 88 evaluable patients initially received conventional HCVAD and 76% were receiving this regimen in first salvage. The ORR was 64%; 47% of patients achieved CR, whereas the remainder achieved partial response (PR) or CRp. Twenty-eight patients (32%) underwent AHSCT, 19 of whom were in CR at the time of transplantation. Despite the favorable response rate, median OS was 6 months, with some long-term responders. As such, regardless of whether they have received conventional HCVAD previously, this regimen represents a reasonable choice for relapsed patients with good performance status who can tolerate intensive chemotherapy.

Clofarabine is a new-generation purine nucleoside analog modeled after fludarabine and cladribine that is approved as a third-line therapy for pediatric ALL (⁷⁶). Attempts have been made in both pediatric and adult population to build on its modest activity as a single agent by combining it with other chemotherapeutics (^77,78,79). In a phase II multicenter study evaluating clofarabine added to cyclophosphamide and etoposide in 25 heavily relapsed refractory pediatric patients, ORR was 44% (7 patients with CR), and 10 patients eventually underwent AHSCT (⁷⁷). Patients with prior AHSCT were excluded due to a high rate of veno-occlusive disease (VOD). In the adult population, a French group evaluated clofarabine in patients with relapsed ALL by combining it with dexamethasone, mitoxantrone, etoposide, and asparaginase (VANDEVOL, n = 37) or with cyclophosphamide (ENDEVOL, n = 18) (⁷⁸). Complete remission was achieved in 41% and 50% of patients, respectively, and less than one-third of patients received subsequent AHSCT. The median OS was 6.5 months. The PETHEMA group reported on 31 heavily pretreated relapsed/refractory adult ALL patients, of whom 84% had received two or more previous treatment regimens (⁷⁹). Complete remission was achieved in 26% of patients, with a median OS of approximately 3 months.

Nelarabine is soluble prodrug of 9-β-D-arabinofuranosylguanine and is approved as a nucleoside analog. It has predominant activity in patients with relapsed T-ALL who have failed two prior regimens. By selectively accumulating in T cells, it lends itself as particularly useful in this subset of patients. The GMALL group analyzed 126 patients with relapsed T-ALL/lymphoma who received two cycles of nelarabine as a single agent, with a goal to consolidate patients in CR with AHSCT (partial responders appeared to have potential to achieve CR with continued therapy) (⁸⁰). After one to three cycles, the CR rate was 36%, with 80% receiving AHSCT. Median OS for the entire cohort was only 6 months, with a 12% probability of survival at 3 years; those who underwent AHSCT after CR had a far better outcome, with a 3-year OS probability of 36%. The major toxicity of nelarabine is neurotoxicity; however, this study noted a low rate of grade 3/4 neurotoxicity, even in patients with prior heavy exposure to vincristine. Nelarabine remains a viable salvage option, and its use continues to be optimized in the frontline combination setting or with alternative dosing schedules (⁸¹).

Vincristine is a standard component of almost all regimens in both adult and pediatric ALL and is frequently capped at 2 mg due to compromising dose-limiting neurotoxicity manifested as peripheral neuropathy or moderate to severe constipation. In an effort to optimize its pharmacodynamics and delivery, a liposomal formulation was synthesized. A phase II study evaluated 65 patients with relapsed and refractory ALL with vincristine sulfate liposome administered as a weekly intravenous infusion at 2.25 mg/m² with no capping parameter (⁸²). In this heavily pretreated population, the ORR was 35%, and 12 patients underwent AHSCT. Grade 3 peripheral neuropathy was observed in 15% of patients, and the median OS was 4.6 months (five patients were alive at 12 months).

Monoclonal Antibodies

Further intensification of conventional regimens has been limited by unfavorable toxicity/antileukemic profile. The development of monoclonal antibodies directed against cell surface antigens that are better tolerated has since led to significant improvement in outcomes for a number of malignancies, including ALL.

Blinatumomab, a bispecific antibody (anti-CD3 and anti-CD19), was studied in patients with relapsed/refractory ALL as a continuous infusion for 28 days every 6 weeks. The ORR (CR or CRp) with two cycles of therapy was 68%, with an estimated median OS of 9 months (⁸³). In an open-label, single-arm, multicenter phase II study in 189 patients with relapsed/refractory disease, the ORR was 43%, with 80% of the responses during the first cycle. The median response duration and OS were 9 and 6 months, respectively (⁸⁴). Blinatumomab causes constitutional symptoms (fever, chills) that coincide with a rapid rise in activated T cells, leading to secondary cytokine release syndrome (CRS) that occurs shortly after the start of therapy (⁸⁵). It can be mitigated with the short course of steroids. Blinatumomab is currently being evaluated in a phase II study in patients with relapsed Ph-positive ALL and in a phase III trial in patients with ALL in first or second relapse (blinatumomab vs investigator's choice of therapy). In a priority review designated by the US Food and Drug Administration, blinatumomab was granted accelerated approval in December 2014 for the treatment of patients with Ph–negative, relapsed or refractory B-ALL.

Inotuzumab ozogamicin is a CD22 immunoconjugate linked to calicheamicin, which is a potent cytotoxic inducer of double-strand DNA breaks. A single-institution phase II study of inotuzumab in 49 patients with highly relapsed/refractory ALL (73% receiving two or more salvage therapies) treated every 3 to 4 weeks resulted in an ORR of 57%, with a median OS of 5.1 months. Survival was comparable between patients who underwent a subsequent AHSCT and those who did not. Serious toxicity in the transplant group included the development of VOD in five patients (23%), although four of these patients had multiple prior alkylating therapies. A modified weekly dosing schedule developed based on preclinical studies yielded a similar ORR to the single-dose schedule (59% vs 57%), with a median survival of 9.5 months. This mode of delivery was less toxic, and hepatotoxicity was also significantly reduced, including a 7% incidence of VOD after autologous stem cell transplantation. An additional multicenter phase I/II study of 37 patients with relapsed/refractory ALL, in which 54% of patients were receiving two or more salvage therapies, also evaluated weekly inotuzumab. The CR and CRp rates were 79% (19/24) and 46% (6/13) in the dose-expansion and dose-escalation cohorts, respectively. Among the responders, 22 patients achieved MRD negativity. Furthest along in the development process, inotuzumab was investigated and compared in a randomized trial to physician's choice of therapy in patients with ALL in first and second salvage. Early results have shown a significant improvement in the response rate (CR/CR with incomplete hematologic recovery) favoring inotuzumab versus standard-of-care intensive chemotherapy (80.7% vs 33.3%, P < .0001) (⁸⁶). Survival data are maturing.

Chimeric Antigen Receptor T Cells

Autologous T cells can be engineered to express a receptor directed at CD19. Response can be durable given the ability of T cells to expand and persist in vivo. As such, chimeric antigen receptor (CAR) T cells have become an effective approach for targeting lymphoid malignancies (^87,88). In a pilot study with 25 children and 5 adults who were treated for relapsed/refractory ALL with CTL019, 27 patients (90%) achieved CR, with 6-month EFS and OS rates of 67% and 78%, respectively (⁸⁹). In a recent phase I trial of 21 children and young adults with relapsed/refractory ALL treated with CAR T cells after lymphodepletion, 21 evaluable patients (70%) achieved CR, with 12 (60%) achieving MRD negativity. At a median follow-up of 10 months, the OS rate was 52% (⁹⁰). In another study of relapsed refractory ALL with antecedent detectable disease before T-cell infusion, 14 patients (88%) showed response, 10 (63%) with CR and 4 (25%) with CRp. Overall, 75% of treated patients achieved MRD negativity (⁹¹). Patients responding to this form of therapy invariably develop some degree of CRS, which is usually very manageable with steroids or tocilizumab. Ongoing research is trying to identify the most optimal use of this innovative therapeutic strategy.

A number of innovative therapies for various stages of disease, tailored to risk-adapted strategies, are transforming the treatment of adult ALL and are beginning to result in significant improvements in long-term outcomes. There are several ongoing trials available for patients with ALL at our institution and elsewhere in the frontline and salvage setting (Table 1-6). Our choice of therapy for particular ALL subtypes is outlined in Table 1-7.

Therapeutic capabilities in adult ALL have rapidly reached new heights over the past decade with the introduction of highly promising monoclonal antibodies, immune conjugates, CAR T cells, and new-generation TKIs. Rituximab has repeatedly been shown to improve OS, and blinatumomab and inotuzumab have demonstrated significant activity in a highly relapsed/refractory population. Genomic profiling has identified new prognostic markers (eg, IKZF1), as well as new therapeutic targets (eg, ABL, JAK, ETV6-NTRK3) that are amenable to targeted therapies that can improve the adverse prognosis of Ph-like ALL. As many of the newer agents advance through the final stages of development, we will be seeking to determine optimal combination and order of delivery and the role of cytotoxic chemotherapy in the safest achievement of durable cure rates. Although the role of these agents still continues to be defined, frontline introduction of most effective therapies can be expected to increase the rate of MRD negativity, optimizing responses and closing the outcome gap separating pediatric from adult ALL. Harnessing the full potential of the immune system with the durable presence of autologous T-cell constructs may ultimately lead to obviation of stem cell transplantation in search of better cure rates in adult ALL.