Allogeneic stem cell transplantation (SCT) is currently considered the unique potentially curative therapeutic option for Philadelphia chromosome-positive acute lymphoid leukemia (Ph+ ALL).1, 2 However, SCT can be pursued only in a minority of cases. We previously described the ability of interferon-α (IFN-α) plus chemotherapy or autologous SCT (ASCT) to induce molecular complete remission (CR) in Ph+ ALL.3 We now show that such an approach can lead to sustained molecular remissions, by updating the follow-up of our original report.
From 1992 to 1998, 14 adult Ph+ ALL patients were referred to our institute and were candidates to receive either ASCT or chemotherapy (if age was over 55 years) followed by IFN-α within a pilot study. Allogeneic SCT was allowed/performed in patients with a suitable human leukocyte antigen-matched donor. Informed consent had been obtained from all the enrolled patients and the protocol was approved by the local ethical committee. All patients received standard induction chemotherapy (modified L-20)3, 4 consisting of vincristine 2.0 mg m−2 weekly (weeks 1–5); 6-methylprednisolone 60 mg m−2 day−1 (days 1–32); cyclophosphamide 1000 mg m−2 (day 4) and 800 mg m−2 (day 32); and adriamycin 30 mg m−2 (days 15–18 and day 32). Prophylactic central nervous system treatment was performed weekly, introducing the following drugs: dexamethasone 4 mg; cytosine–arabinoside 40 mg; and methotrexate 10 mg. Patients achieving morphological and cytogenetic CR were then submitted to a rotational consolidation regimen lasting 6 months,3 including cytarabine, cyclophosphamide, VM26, methotrexate, 6-mercaptopurine, vincristine and prednisone with monthly intrathecal therapy.4 Patients aged <55 years underwent ASCT or SCT; patients aged 55 years received standard maintenance treatment for 6 months (alternating vincristine/prednisone with methotrexate/6-mercaptopurine). Subsequently, maintenance treatment (all ages) was based on cycles of IFN-α (3 MU three times a week for 6 weeks) alternated with methotrexate/6-mercaptopurine for up to 2 years. Thereafter, patients maintaining CR had the same schedule of IFN-α (6 weeks on, 6 weeks off). BCR-ABL1 transcript levels were detected at diagnosis and during the follow-up by real-time quantitative PCR, initially as reported by Amabile et al.5 and subsequently by using a standardized method that was established within the framework of the EU Concerted Action.6 The method independently measures, in each sample by real-time quantitative PCR, the copy number of mRNA encoding the p210 BCR-ABL protein or the p190 BCR-ABL protein and a control gene (ABL1) to verify sample-to-sample RNA quality variations. Real-time quantitative PCR was performed on an ABI PRISM 7700 Sequence Detector (Perkin Elmer, Foster City, CA, USA). The quantification principles and procedure using the TaqMan probe have been previously described. All real-time RT-PCR experiments were performed in duplicate. The copy number of BCR-ABL1 and ABL1 transcripts was derived by the interpolation of threshold cycle (ct) values to the appropriate standard curve, and the result, for each sample, was expressed as a ratio of BCR-ABL1 mRNA copies to ABL1 mRNA copies percent. The threshold was systematically set at 0.1 to avoid any particular problems of baseline creeping. The lowest level of detectability of the method is 0.001 (corresponding to a sensitivity of 10−4 in the clinical setting). Molecular CR was defined as a level of BCR-ABL1/ABL1 × 100 lower than 0.001 followed by nested-PCR negativity in three consecutive determinations.
After induction and consolidation, 9/14 patients obtained morphological and cytogenetic CR but residual disease was still evident by quantitative RT-PCR analysis in all but one case. Three patients received ASCT, three chemotherapy only, two allogeneic SCT and one abandoned the study due to early relapse. Salvage chemotherapy was not successful and he died within a few months. The five refractory patients received either chemotherapy (N=3) or chemotherapy and allogeneic SCT (N=2); all died of leukemia within few months. Five out of 10 patients then received IFN-α as scheduled (as one out of the six remaining patients refused any further therapy). They received IFN-α for a median period of 92 months (range, 48–142). Three out of five patients (nos. 1, 2 and 3) achieved a molecular CR after starting IFN-α treatment after a median period of 14 months (range, 2–36 months). One patient maintained the minimal residual disease negativity obtained after consolidation (no. 5). One patient (who has been the object of a separate report7) did not achieve molecular CR, relapsed and then received imatinib. After a new relapse, she received imatinib plus IFN-α obtaining a molecular CR still persistent after a follow-up of 85 months.7 Treatment with IFN was overall well tolerated in all but one (no. 2, who experienced moderate flu-like symptoms) who discontinued the treatment after 2 years of IFN-α administration. For the other cases, however, a dose reduction to 1 MU three times a week was scheduled to control cytopenia (2/4 cases) or flu-like symptoms (2/4 cases). Interestingly, the mean CR duration was longer (98 months; range, 80–148) than in the previous reports of Ph+ ALL patients treated with chemotherapy regimens (excluding SCT). It should be underlined that time to minimal residual disease disappearance was highly variable after IFN-α therapy, with example of late response (36 and 14 months for patients 1 and 2, respectively). In addition, noteworthy, an attempt to suspend IFN-α administration (upon the request of patient no. 1) was followed by minimal residual disease positivity, suggesting the possibility of long-term requirement for immunological control of residual leukemic clones even in an acute disease, similar to that observed in chronic myeloid leukemia. On the other hand, patient no. 2 discontinued therapy after 2 years without evidence of subsequent relapse. Thus, whether IFN-α treatment should be maintained or not in all patients is still uncertain.
Patients' characteristics and treatment outcome are summarized in Table 1.
Though based on a small series, this is a notable example of potentially curative therapy other than SCT in Ph+ ALL. Recently, imatinib and other tyrosine-kinase inhibitors have been shown to be highly effective in Ph+ ALL and will probably significantly change the natural history of the disease.2, 8, 9 However, they do not appear, at the moment, to be curative as single agents due to the occurrence of specific resistance.2, 9 As there is evidence of possible synergism of IFN-α and imatinib,7, 10 it is indeed warranted to test the combination of IFN-α and tyrosine-kinase inhibitors in future clinical trials, as well as their association with ASCT/chemotherapy.
This work was supported by European LeukemiaNet, COFIN 2002-2003 (Professor M Baccarani and SA Pileri), FIRB/RFO (Professor M Baccarani and SA Pileri), AIRC, Progetto Strategico di Ateneo 2006 (Dr Piccaluga), Fondazione CARISBO Bologna, Fondazione del Monte di Bologna e Ravenna. AIL Pesaro Onlus and BolognAIL grants.