Changing Paradigm of the Treatment of Philadelphia Chromosome–Positive Acute Lymphoblastic Leukemia
- First Online:
- Cite this article as:
- Ohno, R. Curr Hematol Malig Rep (2010) 5: 213. doi:10.1007/s11899-010-0061-y
- 96 Views
In the pre-imatinib era, the treatment outcome of patients with Philadelphia chromosome–positive acute lymphoblastic leukemia (Ph+ ALL) was dismal. Complete remission was generally achieved only in about 50% to 60% of patients, and allogeneic hematopoietic stem cell transplantation (allo-HSCT), when feasible in younger patients, was virtually the sole curative modality. Imatinib has changed the situation dramatically, however, in combination with conventional chemotherapy or with corticosteroid alone, producing about 95% complete remission and thus increasing the number of patients undergoing allo-HSCT. Currently, the overall survival of patients who have undergone allo-HSCT exceeds 50%, and a considerable proportion of patients for whom allo-HSCT is not feasible are predictably curable. The next question is how to prevent relapse, which is observed not only in more than half of patients for whom allo-HSCT is not feasible but also in a considerable number of patients after allo-HSCT. Thus, improvement of postremission therapy is crucial. Whether intensive chemotherapy with currently available cytotoxic drugs contributes to the prevention of relapse is questionable, because intensive chemotherapy alone in the pre-imatinib era nearly always failed to cure this disease. Promising partners to be combined with imatinib or with a second-generation tyrosine kinase inhibitor (TKI) will be corticosteroids and vincristine. New TKIs such as dasatinib should be incorporated into the early phase of postremission therapy. Recognizing the small number of patients with Ph+ ALL, intergroup or international studies are necessary to develop the best postremission therapy. In the near future, it is hoped that Ph+ ALL will become one of the leukemias for which allo-HSCT is offered only for relapsed or extremely high-risk patients.
KeywordsAcute lymphoblastic leukemiaPhiladelphia chromosomePh+ ALLBCR-ABLImatinibDasatinibNilotinibTyrosine kinase inhibitorTKIPrognostic factors
The Philadelphia chromosome (Ph) is the most frequent cytogenetic abnormality in adult acute lymphoblastic leukemia (ALL), occurring in 20% to 30% of adult patients  but in only 2% to 5% of children with ALL . Ph is derived from t(9;22)(q34;q11) reciprocal translocation, which generates a BCR-ABL fusion gene. Its transcript constitutively activates ABL tyrosine kinase and plays a critical role in leukemogenesis . The presence of Ph is the most adverse prognostic factor for ALL. Although complete remission (CR) is achieved in 50% to 80% of Ph+ ALL by intensive chemotherapy, long-term outcome is dismal, with overall survival (OS) of approximately 10% in adult patients and 20% in children. The most common cause of treatment failure is relapse, which occurs mostly within the first year after CR [1, 2, 4, 5].
Therapeutic Modalities for Ph+ ALL in the Pre-imatinib Era
In childhood ALL, with risk-directed intensive chemotherapy, the CR rate has exceeded 95% and 5-year event-free survival (EFS) has ranged from 70% to 83%, with a cure rate of about 80% . For Ph+ ALL in children (comprising 2% to 5% of cases), however, these rates did not apply. A meta-analysis of 326 children with Ph+ ALL (median age, 8.1 years) demonstrated 82% CR, 28% 5-year EFS, and 40% OS. The 5-year EFS was 49% for patients with the best prognostic factors, 30% with intermediate factors, and 20% with the worst factors. Patients who received allogeneic hematopoietic stem cell transplantation (allo-HSCT) showed 72% 5-year OS and 65% relapse-free survival (RFS); in patients receiving chemotherapy alone, the 5-year OS was 42% and RFS was 25% (P < 0.002) .
This meta-analysis included one very informative report, from the Berlin-Frankfurt-Münster–Associazione Italiana Ematologia Oncologia Pediatrica (BFM-AIEOP) group. Of 4760 children with ALL who received intensive BFM protocols, 61 (1.3%) had Ph+ ALL; their CR rate was 90%, 4-year EFS was 38%, and 4-year OS was 49%. Response to initial prednisone therapy was the only independent prognostic factor for survival. Other features, such as white blood cells (WBC), age, sex, and BCR breakpoint, were not independent factors. The 4-year EFS for 30 prednisone good responders was 55%, compared with 10% for 27 poor responders (P = 0 .0001). Interestingly, among good responders, 12 (63%) of 19 treated with chemotherapy alone and 14 (78%) of 18 who underwent allo-HSCT were alive, whereas among poor responders, 0 of 12 treated with chemotherapy alone and only 2 of 8 who underwent allo-HSCT were alive . These results imply that prednisone response was a critical prognostic factor. In other words, patients with Ph+ ALL were hardly curable unless their leukemia cells were sensitive to corticosteroid, even if they underwent very intensive chemotherapy or allo-HSCT.
For adults with Ph+ ALL, allo-HSCT was the only curative modality in the pre-imatinib era. The French GET-LALA Group (Groupe d’Etudes et de Traitement des Leucémies Aiguës Lymphoblastiques de l’Adulte) conducted a prospective study to evaluate the prognostic value of early response and HSCT in first CR in 154 adults with Ph+ ALL (median age, 42 years). The CR rate was 53% and 3-year OS was 19%. In 103 patients eligible for HSCT, donor availability and negative BCR-ABL status after intensive consolidation therapy were independent predictive factors for CR duration (P ≤ 0.001) and OS (P ≤ 0.02 ). The 3-year OS was 37% for the donor group (n = 60) and 12% for the no-donor group (n = 43) (P = 0.02) .
In patients with Ph+ ALL who were enrolled in the international UKALL XII/Eastern Cooperative Oncology Group (ECOG) 2993 trial, the 5-year relapse risk was 81% for those treated with either chemotherapy or autologous HSCT, and 32% for those treated with allo-HSCT. The 5-year OS was 19% for the former group and 42% for the allo-HSCT group .
The Japan Adult Leukemia Study Group (JALSG) conducted four trials for adult ALL (ALL87, ALL90, ALL93, and ALL97) in the pre-imatinib era [9–11; I. Jinnai, personal communication]. The CR rates for a total of 188 patients with Ph+ ALL ranged from 51% to 64%, compared with 76% to 91% for non-Ph+ ALL. The latest published ALL93 study showed that dose-intensified anthracycline did not improve the outcome of 51 patients with Ph+ ALL, who had a 51% CR rate and 5% 6-year OS, except for patients receiving allo-HSCT, who had 44% OS .
A group at M.D. Anderson Cancer Center reported 92% CR in 48 adults with Ph+ ALL who received eight cycles of the Hyper-CVAD regimen (hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone) , but figures in their latest publication suggest that long-term OS is only about 10% .
Treatment of Ph+ ALL in the Imatinib Era
Imatinib mesylate, a potent ABL TKI, has revolutionized the treatment of chronic myelogenous leukemia (CML) and become its first-choice treatment modality [14, 15]. Although its phase II study for relapsed or refractory Ph+ ALL resulted in 29% CR, the development of resistance and subsequent disease progression was very rapid , so the manufacturer was reluctant to conduct further study on Ph+ ALL. Gratifyingly, investigator-initiative studies for newly diagnosed Ph+ ALL demonstrated that imatinib-based therapy resulted in very high CR rates and far better outcomes than occurred in the pre-imatinib era [13, 17–19].
Treatment of Newly Diagnosed Adult Ph+ ALL
Imatinib Plus Standard Induction Chemotherapy
The JALSG Ph+ ALL202 Study first applied imatinib in newly diagnosed Ph+ ALL. After starting a standard chemotherapy with daunorubicin, cyclophosphamide, vincristine, and prednisolone, imatinib (600 mg daily) was begun in combination with daily prednisolone and weekly vincristine from day 8 until day 63. Consolidation therapy consisted of high-dose methotrexate, high-dose cytarabine, and alternately administered imatinib for 28 days. After four cycles of consolidation therapy, monthly maintenance therapy with imatinib, vincristine and prednisolone was given for 2 years. Allo-HSCT was recommended as soon as feasible . Of 80 patients (median age, 48 years), 77 (96%) achieved CR. Molecular CR was noted in 33 (50%) of 66 patients on day 63, and in 57 (71%) of all patients before allo-HSCT within 12 months. Two-year OS was 58% . The 4-year OS was 40%, far better than the 11% of 51 patients in the JALSG-ALL 93 study . Allo-HSCT was performed in 60 patients, and 2-year RFS was 63% for 44 patients who underwent allo-HSCT in first CR and 62% for those who did not . Updated results for a total of 100 patients enrolled into this study revealed 97% CR and 46% 3-year RFS .
Imatinib Plus Hyper-CVAD
The M.D. Anderson group gave imatinib at a dose of 400 mg (later increased to 600 mg) daily for the first 14 days of each of eight cycles of Hyper-CVAD chemotherapy, followed by imatinib-based maintenance, and reported 92% CR and 55% OS in 39 patients with Ph+ ALL . There was no difference in OS between 15 patients who received allo-HSCT in first CR and 28 who did not .
Concurrent Versus Alternating Schedules of Imatinib and Chemotherapy
The German group reported two sequential studies on newly diagnosed adult Ph+ ALL. Patients received the same two cycles of induction and one cycle of consolidation chemotherapy. Imatinib (400–600 mg daily) was administered concurrently with the second cycle of induction chemotherapy and thereafter until allo-HSCT (n = 45; median age, 41 years), or 600 mg was given daily alternately with consolidation therapy after the induction chemotherapy, until allo-HSCT (n = 47; median age, 46 years). Concurrent administration resulted in 95% CR and 52% negativity by polymerase chain reaction (PCR), whereas alternating therapy resulted in 100% CR but 19% PCR negativity (P = 0.01). In each cohort, 77% of patients underwent allo-HSCT in first CR; 2-year OS was 36% in the concurrent group and 43% in the alternating group (P = 0.97) .
Imatinib Plus Chemotherapy as Consolidation or Salvage Therapy
The French group treated 45 adults with newly diagnosed Ph+ ALL (median age, 45 years). For 14 early responders to prednisone and chemotherapy, imatinib (600 mg daily) was given concomitantly with high-dose cytarabine plus mitoxantrone as consolidation therapy. For 31 patients with prednisone-resistant and/or chemoresistant disease, imatinib (800 mg daily) was given in combination with dexamethasone and vincristine. Overall, the rate of CR was 96% and the rate of real-time quantitative PCR (RQ-PCR) negativity was 29%. All of 22 CR patients with a donor received allo-HSCT in their first CR, and their 18-month OS was 65%, significantly longer than reported in the pre-imatinib study (P = 0.05) .
Concurrent Intensive Chemotherapy and Imatinib Before and After HSCT
The Spanish group gave imatinib (400 mg daily) concurrently with intensive chemotherapy, HSCT, and imatinib maintenance therapy in newly diagnosed adult Ph+ ALL. Of 30 patients (median age, 44 years; 1 < 15 and 2 > 60), 27 (90%) achieved CR. Major molecular response (MR) was noted in 81% and complete MR was seen in 65% after consolidation. Of 27 patients who achieved CR, 16 underwent allo-HSCT and 5 underwent auto-HSCT. Imatinib could be administered after HSCT in only 12 patients, and it was interrupted in 10. Nine patients relapsed—4 before and 5 after HSCT—and 8 died of transplant-related causes. The 4-year OS was 30%. Post-HSCT imatinib administration was limited, mainly because of HSCT-derived complications .
Treatment of Newly Diagnosed Ph+ ALL in Elderly Patients
Imatinib Plus Prednisone Induction Therapy Without Additional Chemotherapy
The Gruppo Italiano Malattie Ematologiche dell’Adulto (GIMEMA) treated 30 elderly patients with Ph+ ALL (median age, 69 years; range, 61–83) with imatinib (800 mg daily) until day 48, plus prednisone for 45 days, without chemotherapy. All of the 29 evaluable patients obtained CR and most of the induction therapy required no hospitalization. Median OS was 20 months [27•].
Imatinib Compared with Chemotherapy
The German group randomly assigned 55 elderly patients with Ph+ ALL (median age, 68 years; range, 54–79) to either imatinib (600 mg daily) or multiagent chemotherapy. Consolidation with imatinib (600 mg daily) in combination with five cycles of chemotherapy was given regardless of the remission status. Overall, the CR rate was 96% in patients assigned to imatinib and 50% with chemotherapy (P = 0.0001). The 2-year OS was 42% for all patients. There was no significant difference between the two cohorts because the consolidation therapy with imatinib diminished the initial benefit .
Imatinib Plus Methylprednisolone Alternating with Chemotherapy
The French group treated elderly Ph+ ALL patients (median age, 66 years; range, 58–89) with induction chemotherapy followed by consolidation with imatinib (600 mg daily) and methylprednisolone from day 36 to day 95. Patients in CR were given 10 maintenance blocks of alternating chemotherapy, including two 2-month courses of imatinib. Of 29 assessable patients, 21 (72%) obtained CR after induction chemotherapy and 26 (90%) obtained CR after salvage with imatinib. The 1-year OS was 66%, higher than the historical control rate of 43% (P = 0.005) .
Treatment of Ph+ ALL in Children
Imatinib with Intensive Chemotherapy
After standard remission induction chemotherapy, the Children’s Oncology Group added imatinib (340 mg/m2 daily) to an intensive chemotherapy regimen in 92 patients with childhood Ph+ ALL (ages 1–21 years). Imatinib was increased progressively among five cohorts, which received it for periods ranging from 42 days (cohort 1) to 280 days (cohort 5) before maintenance therapy. Patients with sibling donors underwent allo-HSCT, with imatinib given for 6 months following HSCT. Continuous imatinib exposure improved outcome in the patients in cohort 5 (n = 50), who received imatinib for 280 days, and 3-year EFS was 80%, more than twice the rate in historical controls (35%). In cohort 5, EFS was similar for patients treated with chemotherapy plus imatinib (88%) or with allo-HSCT (57%). The outcome observed among patients treated with imatinib for 280 days in combination with intensive chemotherapy was dramatically better than the outcome of historical controls [30••].
Analysis of Factors Associated with Relapse in the Imatinib Era
Because almost all patients with Ph+ ALL are now induced into CR by imatinib-based therapy, analysis of factors associated with relapse is urgently needed. The JALSG analyzed prognostic factors in 80 adults with Ph+ ALL. Variables that showed a significantly adverse effect on RFS were additional chromosomal aberrations at diagnosis (P = 0.005), higher peripheral blood blasts (P = 0.024), and male sex (P = 0.03). By multivariate analysis, the presence of additional chromosomal aberrations was the only independent prognostic factor (P = 0.027), followed by peripheral blood blasts (P = 0.051). Results of RQ-PCR on day 63 of induction therapy were available for 66 patients, in 33 (50%) of whom BCR-ABL was negative. There was no difference in RFS between patients with or without PCR negativity on day 63 (P = 0.707). Neither BCR-ABL transcript types (23 major and 56 minor) nor copy numbers at diagnosis were associated with RFS .
In a subsequent analysis of 100 patients, monitoring of minimal residual disease (MRD) was prospectively performed using RQ-PCR in 367 follow-up bone marrow samples collected at predefined times. Of the 100 patients, 97 achieved CR; 3-year RFS was 46%. Complete MR at the end of induction therapy was associated with neither RFS nor relapse. In 29 patients who showed MRD elevation during CR, 16 underwent allo-HSCT in first CR and 10 of these were alive without relapse at a median of 2.9 years. Of the 13 who received no allo-HSCT, 12 relapsed. Thus, in Ph+ ALL treated with imatinib-based therapy, rapid negative MRD was not associated with a favorable prognosis, but elevation of MRD during CR was predictive of subsequent relapse, although allo-HSCT overrode its adverse effect .
The German group reported that median disease-free survival in elderly patients with Ph+ ALL was significantly longer in the 43% of patients in whom BCR-ABL transcripts became undetectable (18.3 months vs 7.2 months, P = 0.002) . A Korean group also found that MRD at 4 weeks of imatinib therapy determined the outcome of allo-HSCT. Of 52 patients with newly diagnosed Ph+ ALL, 11 (21%) achieved molecular CR and 36 (69%) at least 3-log reduction in BCR-ABL transcripts. Among 48 patients who underwent allo-HSCT, the relapse rate was 21% and 4-year RFS was 70%; at least a 3-log reduction at 4 weeks was the most powerful predictor of a lower relapse rate (12% vs 45%, P = 0.011) and better RFS (82% vs 42%, P = 0.009) .
Resistance to Imatinib
The observation in the JALSG study that two thirds of the early relapses occurred during the consolidation courses consisting of imatinib alone implies a possible association with imatinib resistance [21, 22, 31]. The etiology of resistance to imatinib is multifactorial, and several mechanisms have been suggested, including mutations or amplification of the BCR-ABL gene, overexpression of BCR-ABL transcripts, molecules downstream of BCR-ABL signaling, and reduced intracellular imatinib concentration .
Of these mechanisms of resistance, mutations involving the ABL kinase domain are the most common. About 50 mutations have been identified, classified as occurring at four regions of the ABL kinase domain: the ATP-binding site (P-loop), contact site, SH2 binding site, and A-loop. P-loop mutations are the most frequent [34, 35]. T315I at the contact site is of particular importance because of its complete resistance to imatinib and the second-generation TKIs .
Relationships between the mutation status and clinical outcome have been studied intensively in patients with CML, but data for Ph+ ALL are limited. The German group demonstrated that, even before imatinib therapy, ABL kinase domain mutations were detected in 38% of Ph+ ALL. Importantly, the frequency of the mutant allele was quite low at diagnosis, with a median value of 0.5% and a maximum value of 2%. At the time of relapse, however, the dominant clone harbored the same mutation in most patients. Thus, elimination of mutant clones would be necessary during the early phase of treatment .
Accumulated evidence suggests that imatinib resistance also occurs through BCR-ABL–independent mechanisms. For example, SRC family kinases are considered to be involved in disease pathogenesis as well as in resistance to imatinib in Ph+ ALL . Therefore, simultaneous inhibition of these kinases expectedly yields a synergistic activity.
New BCR-ABL Tyrosine Kinase Inhibitors
Newly developed BCR-ABL TKIs for use in treating imatinib-resistant leukemia include dasatinib, nilotinib, bafetinib, and bosutinib. Among these, dasatinib and nilotinib are now on the market for CML with imatinib resistance or intolerance. All these second-generation TKIs are active in Ph+ ALL with imatinib resistance or intolerance, but so far, only dasatinib has been approved for Ph+ ALL.
Dasatinib, an oral TKI, belongs to the Src family. It has 325-fold greater potency than imatinib, is active against most BCR-ABL mutations conferring imatinib resistance (exceptions include T315I, V299L, and F359V), and potently inhibits TKIs implicated in other imatinib resistance. Unlike imatinib, dasatinib is not a substrate of the P-glycoprotein pump [36, 39].
In phase II study, 36 Ph+ ALL patients with imatinib resistance or intolerance were treated with dasatinib, 70 mg twice daily. Major hematologic responses were achieved in 42%. Complete cytogenetic response (CyR) was attained in 58%. The presence of BCR-ABL mutations did not preclude a response to dasatinib . Dasatinib crosses the blood-brain barrier (imatinib does not), so it is an efficient therapy for central nervous system leukemia .
In a phase III study, 84 Ph+ ALL patients with imatinib resistance or intolerance were randomly assigned to receive dasatinib at a dosage of either 140 mg once daily (n = 40) or 70 mg twice daily (n = 44). The rate of major hematologic response was 38% with once-daily dosing and 32% with twice-daily dosing, but major CyR with once-daily dosing (70%) was higher than for twice-daily dosing (52%). The once-daily schedule had longer progression-free survival (median, 3.0 months vs 4.0 months) and shorter OS (median, 9.1 months vs 6.5 months). Overall safety profiles were similar between the two groups. Pleural effusion was less frequent with once-daily dosing (18% vs 32%) .
Dasatinib Plus Hyper-CVAD for Adults with Newly Diagnosed Ph+ ALL
The M.D. Anderson group conducted a phase II study in newly diagnosed or lightly treated Ph+ ALL (median age, 51 years) by adding dasatinib (50 mg twice daily or 100 mg daily) for the first 14 days of each of eight cycles of alternating Hyper-CVAD and high-dose cytarabine and methotrexate. Patients in CR continued to receive maintenance dasatinib (50 mg twice daily or 100 mg daily) with vincristine and prednisone monthly for 2 years, followed by dasatinib indefinitely. Of 41 patients, 39 (95%) achieved CR after the first cycle or were in CR at the start. Of 39 evaluable patients, 31 (79%) achieved complete CyR after one cycle and 4 had major CyR; 22 (56%) achieved complete MR and 30 (77%) achieved major MR at a median of 14 weeks. Grade 3 and 4 adverse events included 18 instances of bleeding, 9 pleural effusions, 1 pericardial effusion, 10 reversible rises in creatinine, 6 deep vein thromboses, and 3 pulmonary emboli. With a median follow-up of 13 months, 29 patients (71%) were alive and 27 (66%) were in CR. Median OS was more than 52 weeks. Eight patients relapsed, and 5 had ABL mutations (three T315I, one F359V, and one V299L) [43•].
Although a combination of dasatinib with intensive chemotherapy is effective in achieving long-term remissions in patients with newly diagnosed Ph+ ALL, toxic events such as bleeding and pleural effusion are more frequent and more severe than those occurring with the combination of imatinib and chemotherapy. Dasatinib is associated with increased risk of bleeding in CML patients, even in the absence of thrombocytopenia , and with platelet aggregation abnormalities . Therefore, when myelosuppressive chemotherapy is combined, appropriate clinical monitoring for bleeding (especially gastrointestinal bleeding) and timely interruption of dasatinib therapy is essential.
Dasatinib Plus Prednisone Induction Therapy Without Additional Chemotherapy in Elderly Patients
GIMEMA treated 34 patients with newly diagnosed adult Ph+ ALL (median age, 54 years) with dasatinib (70 mg twice daily) for 12 weeks. A prednisone pre-phase was started 7 days before the dasatinib and was continued up to day 31. All 34 patients achieved CR, by day 22 in 32 (94%). In 13 patients, at least one severe adverse event was recorded. Nine patients relapsed at a median of 72 days, and the 10-month OS was 81%. Considering these results together with their previous experience with imatinib plus prednisone for elderly patients, these researchers questioned the use of chemotherapy to induce remission of Ph+ ALL [46•].
Nilotinib is an orally active, aminopyrimidine-derivative TKI. Like imatinib, nilotinib functions through competitive inhibition at the adenosine triphosphate (ATP) binding site of BCR-ABL. Nilotinib has a higher binding affinity and selectivity for the ABL kinase than imatinib. It has 20 to 50 times more inhibitory activity in imatinib-sensitive cell lines and 3 to 7 times more activity in imatinib-resistant cell lines; it is active in most imatinib-resistant ABL kinase mutations (exceptions include T315I and F359I) .
In a phase II study with 400 mg nilotinib daily, 26% of 35 Ph+ ALL patients with imatinib resistance or intolerance attained CR. Major CyR was observed in 51%, including complete CyR in 34% . So far, however, nilotinib has not been applied for Ph+ ALL. If nilotinib is used in newly diagnosed Ph+ ALL, a higher quality of CR might be obtained, and consequently relapse may become less frequent.
Allogeneic Hematopoietic Stem Cell Transplantation in the Imatinib Era
In the era before imatinib, all Ph+ ALL patients who achieved CR were recommended for allo-HSCT if any matched donor was available (including umbilical cord blood), because of the universally dismal prognosis. However, transplantation-related mortality and relapse after HSCT were inevitable.
With the availability of imatinib, allo-HSCT still seems to be the first choice of treatment for younger patients, as evidence is scanty that imatinib-based therapy alone can cure this disease. Because about 95% of cases of Ph+ ALL are now induced into CR with an imatinib-based regimen, most patients have an opportunity to receive allo-HSCT in their first CR. Reduced-intensity conditioning regimens for older patients or those with comorbidities may allow more patients to undergo allo-HSCT.
Post-HSCT strategies should also be explored. The German group is conducting a randomized study comparing the initiation of imatinib as early as possible after allo-HSCT with an MRD-oriented approach in which imatinib is started only in patients with detectable MRD. Interim analysis of 49 patients seems encouraging, with only two relapses after a median follow-up of 16 months, although tolerability appears poorer in patients starting treatment early after allo-HSCT . In the Spanish Group study, however, post-HSCT imatinib was not quite successful, mainly because of HSCT-derived complications rather than drug-specific toxicity, and did not result in better outcome .
The results of JALSG Ph+ ALL202 [20, 21] and the M.D. Anderson study of Hyper-CVAD plus imatinib [13, 23], however, indicate that the prognosis of patients who could not receive allo-HSCT is not significantly worse than that of patients undergoing allo-HSCT, at least during short-term follow-up. If new TKIs are widely used to prevent relapse, RFS of these patients may improve more, despite a lack of evidence that second-generation TKIs prevent relapse in patients who achieve CR by imatinib-based therapy. Perhaps in the near future, allo-HSCT will be used only for patients with relapse, just as in childhood ALL , acute promyelocytic leukemia (APL) , and CML [14, 15].
The Children’s Oncology Group showed that treating childhood Ph+ ALL with imatinib plus intensive chemotherapy was in no way inferior to treatment with allo-HSCT (88% EFS vs 57% at 3 years) [30••]. Thus, it is already true that if pediatric Ph+ ALL is properly treated with TKI and chemotherapy, allo-HSCT may be applied only in relapsed patients, as in standard-risk childhood ALL.
What’s Next in the Treatment of Ph+ ALL?
Imatinib combined with chemotherapy, corticosteroid, or both easily yields more than 95% CR in newly diagnosed Ph+ ALL, even in multicenter study settings. Thus, there will be little room for improvement of induction therapy. Adding more intensive chemotherapy will prolong myelosuppression and increase treatment-related mortality. Therefore, the next step is better prevention of relapse.
Lessons learned from past studies of acute leukemia should be applied here. First, risk factors should be identified, and then risk-directed therapy should be explored. So far, additional chromosomal aberrations, higher initial WBC, MRD status, and presence of gene mutations have been identified as risk factors, but the number of studied patients was too small for satisfactory statistical analysis. Considering the number of patients with Ph+ ALL, intergroup studies at international levels will be necessary.
Incorporation of new TKIs into front-line therapy will be worth studying, but preliminary results from M.D. Anderson, where dasatinib was used in combination with intensive chemotherapy, showed worse toxic profiles than those for imatinib. Incorporation of second-generation TKIs into the postremission phase will be the next step to be studied, especially in patients with identified risk factors.
The recent finding of BCR-ABL–specific cytotoxic T cells during long-term imatinib therapy is interesting . Whether specific T cells expanded ex vivo can be used for immunotherapy remains to be tested. Another finding on clonal expansion of NK/T cells during dasatinib therapy is also interesting. Kim et al.  reported that during dasatinib therapy, 8 of 18 patients (17 with CML and one with Ph+ ALL) developed large granular lymphocytosis (LGL), and all patients who developed LGL (including the one with Ph+ ALL) achieved optimal MR (8/8 with LGL versus 3/10 without it, P = 0.002). Mustjoki et al.  also reported that 22 of 46 Ph+ ALL patients receiving dasatinib therapy developed LGL and had superior survival. They speculate that, by inhibiting immunoregulatory kinases, dasatinib may induce a reversible state of aberrant immune reactivity associated with good clinical responses. Thus, dasatinib therapy may be associated with better outcome in patients with Ph+ ALL through clonal expansion of NK/T cells.
Is Intensive Cytotoxic Chemotherapy Indispensable in the TKI Era?
The dismal outcome of Ph+ ALL in the pre-imatinib era teaches us that intensive cytotoxic chemotherapy may play a minor role in the TKI era. GIMEMA used imatinib and prednisone without cytotoxic drugs in elderly patients with Ph+ ALL and attained 100% CR [27•]. They recently gave dasatinib and prednisone without cytotoxic drugs to adults with Ph+ ALL, and again reported 100% CR [46•].
Which Drugs are Candidates for Combining with TKIs?
Kano et al.  reported an in vitro combination study of imatinib and chemotherapeutic agents in Ph+ leukemia cell lines. They found a synergistic effect of imatinib with vincristine; an additive effect with cyclophosphamide, daunorubicin, cytarabine, and etoposide; but an antagonistic effect with methotrexate. Thiesing et al.  reported additive or synergistic effects for the combination of imatinib with interferon, daunorubicin, or cytarabine, but antagonistic effects with hydroxyurea, in a BCR-ABL–expressing human cell line. Regrettably, neither of these in vitro studies tested corticosteroid as a combination partner.
Hongo et al.  examined in vitro drug resistance against 14 agents using bone marrow samples from 274 children with newly diagnosed ALL, including 16 (6%) with Ph+ ALL. The Ph+ ALL cells were significantly more resistant to vinblastine (relative resistance compared with non-Ph+ cells, 2.0; P = 0.01), etoposide (1.7; P = 0.04), L-asparaginase (1.6; P = 0.049), vincristine (1.4; P = 0.06), 4-hydroperoxy-cyclophosphamide (1.5; P = 0.179), cytarabine (1.4; P = 0.217), daunorubicin (1.2; P = 0.291), dexamethasone (1.2; P = 0.309), and prednisolone (1.1; P = 0.706). Thus, although Ph+ ALL cells are generally resistant to chemotherapeutic agents, the difference seems less for corticosteroid. Ramakers-van Woerden et al.  also reported in vitro drug resistance of Ph+ ALL cells from 21 children, compared with non-Ph+ ALL cells from 254 children. They found no significant difference between Ph+ and Ph– samples for prednisolone and dexamethasone, as well as for L-asparaginase, vincristine, and anthracyclines, even after matching for prognostic features such as age, WBC, and immunophenotype. Adult Ph+ ALL (n = 12), however, was more resistant to prednisolone (>270-fold; P = 0.030) and displayed an overall tendency to resistance when compared with matched cases of non-Ph adult ALL (n = 15). Among all the Ph+ ALL samples, prednisolone resistance increased significantly with age (P = 0.006).
Though there is some discrepancy between these two in vitro studies, both studies show that corticosteroids seem to work on Ph+ ALL as effectively as on non-Ph+ ALL. The remarkable CR rates obtained by the two GIMEMA studies using imatinib or dasatinib plus prednisone suggest that corticosteroids are an ideal candidate to be the combination partner of TKIs. Vincristine will be another good candidate, based on the in vitro study by Kano et al. 
Judging from the dismal outcome of Ph+ ALL with the use of intensive chemotherapy alone in the pre-imatinib era, myeloablative chemotherapy may not work even with imatinib. Therefore, early incorporation of second-generation TKIs may be the best choice for postremission therapy. Corticosteroids and vincristine will be candidates to be used as partners if combination therapy is desired. MRD assessment will be very helpful in determining appropriate timing for the switch from imatinib to second-generation TKIs.
The history of leukemia therapy tells us that induction regimens achieving a CR rate greater than 95% (eg, childhood ALL treated with chemotherapy alone  or APL treated with all-trans retinoic acid-based therapy ) have resulted in high cure rates. Currently, a CR rate close to 100% is obtainable in newly diagnosed Ph+ ALL. Therefore, little room is left for the improvement of induction therapy. Efforts should be directed toward improving postremission therapy. Allo-HSCT will still be the preferred choice, but, as in childhood ALL and APL, the time will soon come when Ph+ ALL will be cured by TKI-based therapy, and allo-HSCT will be applied only for relapsed or very high-risk patients. Further clinical studies, ideally at international levels, will be necessary to develop the best treatment for this rare but potentially curable leukemia.
No potential conflicts of interest relevant to this article were reported.