Original article. Probability of long-term disease-free survival for acute myeloid leukemia patients after first relapse: A single-centre experience

Transcription

1 Annals of Oncology 7: 998, 99. O 99 Kluwer Academic Publishers. Printed in the Netherlands. Original article Probability of longterm diseasefree survival for acute myeloid leukemia patients after first relapse: A singlecentre experience M. Vignetti, E. Orsini, M. C. Petti, M. L. Moleti, C. Andrizzi, R. M. Pinto, S. Amadori & G. Meloni Hematology, Department of Human Biopathology, Universitd 'La Sapienza'; Hematology, Department of Internal Medicine, Universitd Tor Vergata, Rome, Italy Summary Background: Various polichemotherapy regimens, including either high or intermediatedose AraC, are generally utilized to reinduce remission in relapsed AML patients. After achieving second CR, bone marrow transplantation (either allogeneic or autologous) represents the treatment of choice for eligible patients, with the aim of prolonging remission duration and improving diseasefree survival. and methods: fifty AML patients in first hematological relapse were treated with MEC regimen, consisting of a day induction cycle [mitoxantrone mg/mvday, cytarabine (AraC) g/m /day and VP 80 mg/mvday] followed by a day cycle with the same drugs for patients achieving complete remission (CR); allogeneic or autologous bone marrow transplantation (BMT) were planned as postconsolidation treatment. Results: Thirtyfour patients (8%) achieved second CR, (%) died during induction and were refractory. CR rate was significantly higher in patients with a first CR lasting > months (8% vs. %, P < 0.00). Out of the patients in CR after the day cycle, 8 (5%) were not eligible to transplant and did not receive any further therapy and (7%) received autologous (5 cases) or allogeneic ( case) BMT at a median time of months from second CR. Twentytwo patients relapsed after a median time of months (range ), patient died from transplantrelated toxicity and are in continuous CR [7 out of (%) in the transplanted and out of (%) in the nontransplanted group]. Overall survival and eventfree survival for the 50 patients were 9% and 9% at 70 months, respectively. The diseasefree survival for the patients who obtained second CR is 9% projected at 9 months [% at 9 months for transplanted patients versus 8% at 9 months for the remaining 8 patients (P 0.007)]. Conclusions: These results show that MEC followed by highdose postconsolidation treatment is a promising approach in relapsed AML; however, alternative strategies are to be investigated for the relevant fraction of patients that, even achieving second CR, are not eligible for BMT. Key words: acute myelogenous leukemia, autologous bone marrow transplantation Introduction Currently, 0%80% of adult patients and 90% of children with newly diagnosed acute myeloid leukemia (AML) may achieve complete remission (CR) after induction chemotherapy [5]. Nevertheless, recurrence of the disease remains a major problem. A second CR may be achieved in more than 50% of relapsed patients by the use of one of several chemotherapy programs, most of which include highdose cytarabine (AraC) alone or combined with other drugs; the duration of these remissions is, however, quite short and the probability of prolonged survival with conventional chemotherapy is very poor [0]. Intensive postremission treatments including highdose chemoradiotherapy followed by allogeneic or autologous stem cell reinfusion have been widely utilized to increase the percentage of longterm survivors in second CR [, ]. A leukemiafree survival (LFS) of approximately 5% at 5 years has been reported by the International Bone Marrow Transplantation Registry (IBMTR) after allogeneic bone marrow transplantation (BMT) []. For patients without HLA identical siblings, autologous BMT could represent an alternative treatment; the last retrospective analysis from the European Bone Marrow Transplantation (EBMT) registry performed on patients autografted in second CR shows a LFS of % at 8 years []. Furthermore, our results in AML patients autografted in second CR using the BAVC conditioning regimen [5] seem very encouraging and have recently been confirmed on a larger series of patients with a longer followup []. In 988, our group started a therapeutic trial of intermediatedose AraC and mitoxantrone plus etoposide (MEC) in patients with advanced AML, which showed that this regimen could be an effective antileukemic treatment with acceptable toxicity. The CR rate was, in fact, 77% for patients treated in first relapse, and more than 90% in those with a first remission lasting more than months (late relapse) [7, 8]. Our two previous reports dealt mainly with the feasibility, toxicity and antileukemic activity of the MEC regimen, in particular with regard to CR rate and duration in different categories of 'highrisk' AML patients. The objective of the present analysis was to evaluate the complete history of the disease in a more homo Downloaded from on 09 April 08

2 9 geneous and consecutive cohort of 50 AML patients, starting from time of first relapse. In particular, our aim was to determine: ) the real probability of becoming longterm survivors for the overall population assigned, at the time of relapse, to a theoretically ideal treatment intensive chemotherapy to induce second CR followed by a myeloablative postconsolidation procedure; ) the fraction of patients who actually will receive a transplant in second CR. Determination of the latter would be particularly useful for obtaining essential figures with respect to the controversial issue of riming of autograft in AML whether in first CR or after relapse. and methods From 988 until 99, all patients referred to our institution for AML in first hematological relapse entered the study. Eligibility criteria were: age less than 70 years, diagnosis of AML excluding M according to the FrenchAmericanBrinsh (FAB) classification, performance status < according to the World Health Organization (WHO) scale [9], bilirubin less than mg/dl, creatinine less than mg/dl, left ventricular ejection fraction greater than 50%, and no evidence of active infection at entry. Written informed consent was obtained from patients or from parents of patients under 8 years of age. The clinical characteristics of patients are detailed in Table. The postremission treatments administered in first CR are detailed in Table. Definition of response CR was defined as a cellular marrow with less than 5% of blasts, a normalization of peripheral blood (PB) counts and a disappearance of all clinical features related to leukemia persisting for more than month. A patient was defined as refractory to the MEC induction regimen if either progression of disease during induction or a persistent bone marrow (BM) blastosis on day +8 was shown. Relapse was defined by a BM blastosis greater than 5% observed at consecutive BM aspirates performed with a week interval. Treatment MEC regimen Consisted of a single day course of mitoxantrone ( mg/m ), etoposide (80 mg/m ) and intermediatedose AraC ( g/m ), administered daily on days through. achieving CR received a consolidation course with the same drugs at the same dosage for days. The drugs were administered according to modalities previously described [8]. Either autologous or allogeneic (if an HLAidentical sibling was available) transplant was planned for all patients still in second CR after recovery from the consolidation MEC cycle, if standard eligibility criteria for receiving myelosuppressive treatments were fulfilled. Autologous BMT The techniques of marrow collection, cryopreservan'on and reinfusion have been previously described [0]. Marrow was collected immediately prior to the start of pretransplant chemotherapy in all but one patient, who received marrow harvested in first CR. Thirteen of 5 patients received BAVC [5] in a preparatory regimen, followed after a day rest by unpurged bone marrow reinfusion. Table. Patient characteristics. Median age (years) Sex Male Female FAB subtype MO Ml M M M5 First CR induction treatments ' + 7' ICE Others First CR duration Median (months) Abbreviations: FAB FrenchAmericanBritish classification; CR complete remission; ABMT autologous bone marrow transplantation; BMT allogeneic bone marrow transplantation; ' + 7' daunorubicine ( days) plus cytosinearabinoside (7 days); ICE idarubicine (days 5) plus etoposide (days to 5) plus cytosinearabinoside (days to 0). Table. Post remission treatments received in first CR. Post remission chemotherapy Pts No ABMT BMT transplant AraC + Amsa/ AraC + DNR DAT ( cycles) AraC + Amsa ( cycle) DAT ( cycles) NOVIA Idarubicine + VP + AraC AraC + Amsa; DAT ( cycles) AraC + Antracyclines ( cycles) + 5; DAT ( cycles); AraC + VP ( cycles) AraC + DNR + VP ( cycles); AraC + Mito i.d. AraC; AraC + Adria + TG AraC + DNR + TG ( cycles); AraC + VP AraC + VP + DNR ( cycles); AraC + TG ( cycles) Total Abbreviations: Pts patients; ABMT autologous bone marrow transplantation; BMT allogeneic bone marrow transplantation; AraC cytosine arabinoside; Amsa amsacrine; DNR daunorubicine; DAT daunorubicine, AraC, thioguanine; NOVIA mitoxantrone, etoposide, AraC; VP etoposide; ' + 5' daunorubicine ( days) plus cytosinearabinoside (5 days); Mito mitoxantrone; i.d. intermediatedoses; Adria adriamycine; TG thioguanine. Two patients received the busulfancyclophosphamide regimen [] because of a previous meningosis in one, and autograft in first CR with BAVC regimen in the second. The patients occupied double rooms, and were treated with a central venous catheter placed for administration of chemotherapy, Downloaded from on 09 April 08

3 95 blood products, and fluids. All received prophylactic oral antibiotics (ciprofloxacine) when neutrophil counts fell below 0.5 x lo'/l. Acyclovir i.v. 5 mg/kg/die was started on day + to prevent herpes virus infection. Intravenous broad spectrum antibiotics were given for fever during aplasia, with the addition of amphotericin B if fever persisted longer than 5 days or if fungal infection was documented. All blood products administered were irradiated with 0 Gy before infusion to prevent graft wmshost reactions. Allogeneic BMT The patients occupied single rooms. The preparatory regimen consisted of busulfan administered orally for consecutive days for a total dose of mg/kg and cyclophosphamide (Cy) administered intravenously at a daily dose of 0 mg/kg for consecutive days []. The donor bone marrow was infused days after the final dose of Cy without manipulation. Reverse isolation and ciprofloxacine were started on the day of infusion and a sterile diet was instituted during neutropenia. To prevent graft vwsushostdisease (GVHD) cyclosporine plus four doses of methotrexate were given. Antimicrobial and transfusional treatments were similar to that previously described for autografted patients. Statistical analysis Overall survival (OS) and eventfree survival (EFS) were calculated from the start of induction therapy: events considered were induction failure and death for OS, plus relapse for EFS. Diseasefree survival (DFS) was evaluated from second CR, and the events considered were relapse and death. The significance of differences observed in proportion was tested by chisquare statistics. Survival curves were calculated according to the KaplanMeier method, with results expressed as percentages ± standard error (). The significance of differences between actuarial curves was estimated by the logrank test, and twosided lvalues were used throughout []. Toxicity All toxicities were graded according to the WHO grading system [9]. Results Outcome Of 50 patients enrolled in the study, (8%) achieved CR, (%) were refractory but survived induction, and (%) died during induction of infection with hypoplastic marrow. The treatmentrelated toxicities were similar to those previously reported by our group with the MEC regimen [7,8]. The remission rate was % (7 of 7 cases) for patients with a first CR lasting < months (early relapses) versus 8% (7 of cases) for the remaining (late relapses) (P < 0.0). In the subgroup of patients who were transplanted in first CR, the remission rate was 55% (5/9 cases) for late relapses and % (/ cases) for early relapses, with an overall remission rate of 50% (/) versus 7% (8/8) for the nontransplanted patients [P not significant (N.S.)]. Eighteen of the patients in CR after MEC therapy (5%) were not transplanted because of persistent toxic effects ( infections and liver toxicity), refusal ( patients), medical decision ( patients) or relapse prior to the time when logistical problems had to be finalized for transplant ( patients); (7%) received highdose treatment with either autologous (5) or allogeneic () BMT (Table ). Only / transplanted patients had received a previous ABMT in first CR Duration of response For the group as a whole the median duration of second CR was months, and it lasted up to 7 months in the transplanted group ( patients) versus months in the nontransplanted group (8 patients) (Table ). In particular, of the 8 nontransplanted patients relapsed after a median of.5 months from second CR (range, to ), and are in continuous CR with a followup of, 0, 7 and 9 months, respectively. None of them received any further treatment. The median age in this group was 7 years (range, 0 to 9), and the median duration of the first CR was 0 months (range, 5 to 8), lasting > months in 5/8 cases. Of the transplanted patients, 7 are in continuous Table. MECregimen:clinical results. Induction failures Refractory disease Early deaths Complete remissions Postremission treatment Autologous BMT Allogeneic BMT 50 (%) (8%) (%) 5 Abbreviation: BMT bone marrow transplantation. Table. Followup of patients in second CR after MEC regimen. Treatmentrelated death No. relapses Median time to relapse Months No. CCR Followup (months) Median Median second CR duration (months) Inversions Diseasefree survival probability Total % (at 9 months) ABMT % (at 9 months) BMT No post consolidation therapy % (at 9 months) Abbreviations: ABMT autologous bone marrow transplantation; BMT allogeneic bone marrow transplantation; CCR continuous complete remission. Downloaded from on 09 April 08

4 9 CR ( autologous, allogeneic), with a median followup of 5 months (range, to 9). Eight patients relapsed in the bone marrow at a median of 5 months (range ) after achievement of second CR. The median age of this group was years, ranging from 9 to 5; 9 of the patients were female. The median duration of first CR was 5 months (range to 50), lasting more than months in / patients. Autograft was performed in 5/ patients after a median time of months (range ) from reaching second CR. The conditioning regimens were well tolerated and slight nausea and vomiting occurred during administration of chemotherapy. A median of. x 08 (range, 0. to.) nucleated marrow cells per kilogram of body weight were cryopreserved and reinfused. During aplasia, three patients developed severe oral mucositis (maximum WHO degree ) which resolved at the time of bone marrow engraftment. In 5 cases a parenteral nutrition was required. All patients had fever and started antibiotic treatment: in 8 cases positive blood cultures for bacteria were documented, in particular, staphylococcus ( cases), streptococcus, enterococcus, bacteroides, serratia and fusobacterium (one case each). Antifungine therapy was administered in 8 cases because of persistent fever: in one patient a pulmonary fungal infection was documented which resolved after amphotericin B treatment. One patient showed no evidence of engraftment and on day +97 received a second infusion of bone marrow harvested in first CR. One patient, previously autografted in first CR and conditioned with Bu + Cy, died on day 0 of hemorrhage. Recovery is evaluable in /5 patients. The median time required to attain an absolute neutrophil count in excess of 0.5 x 09/l was 9 days (range, to ). A sustained platelet count exceeding 0 x 07 was reached after a median of days (range, to 5). The allografted patient, transplanted 7 months after achieving second CR, received. x 08 cells per kilogram b.w. and obtained engraftment on day + for neutrophils >0.5 x 09/l and on day +7 for platelets >50xl0 9 /l. Discussion In this study /50 (8%) AML patients who underwent the MEC regimen in first relapse achieved a second CR. Treatmentrelated toxicity was fairly good, resulting in an induction death rate of only %. These data confirm results previously reported by our group Overall survival = 0.9 Eventfree survival» 0.9 Tlm«In months Figure I. Overall (upper curve) and eventfree (lower curve) survival of 50 AML patients treated with MEC regimen in first relapse. P = 0.0 CR lasted > months (n=): 0. CR lasted S months (n = 7): 0. Tim«In months Figure. Eventfree survival of 50 AML patients treated with MEC regimen in first relapse: influence of the first CR duration. Survival OS and EFS projected at 70 months for the entire group of 50 patients were 9% (±7%) and 9% (± %), respectively (Figure ). The univariate evaluation of prognostic factors showed a clear influence of first CR duration on the outcome: EFS projected at 5 months from the start of MEC therapy was % (± 8%) for early relapses versus % (± 8%) for late relapses (P 0.0) (Figure ). The overall probability of DFS for the patients achieving second CR was 9% (±8%) projected at 9 months (Figure ), in particular, % (± %) Tim«in months at 9 months for the transplanted group and 8% (±9%) at 9 months for the remaining ( P 0.007) Figure. Diseasefree survival of AML patients in second CR (Figure ). after MEC regimen. Downloaded from on 09 April 08

5 97 CR AML, mainly because of the automatic assignment of early failures to the chemotherapy group. However, this experience underscores the good o.«results already reported by our group with the BAVC regimen in second CR AML, confirming the antileukemic efficacy of a treatment combining MEC with trans! p Transplanted (n): 0. plant procedures, as well as the possibility of obtaining I L = 0007 a significant number of longterm survivors in AML also after first relapse []. [ n. Not transplanted 8): 0.8 \ o.a Results reported by other groups in second CR vary widely: DFS ranges from % up to a maximum of o 50% []. Obviously, these results reflect the heterotlma In months geneity of patients treated in individual groups, with different chemotherapy schedules and different clinical Figure. Diseasefree survival of AML patients in second CR characteristics. For this reason an overall comparison after MEC regimen: Influence of the postremissional treatment. of these data is, in our opinion, very difficult. A larger analysis for autograft in AML in second CR has been with the MEC regimen in relapsed AML on a larger reported by Gorin et al. on patients from the series of patients and with a longer followup [7,8]. EBMT registry, showing a global LFS of % at 8 years First CR duration is strictly correlated with the pos [,]. Results from our singlecentre experience with sibility of attaining a new complete remission. The 5 a transplanted group of patients may be considered patients with first CR duration > months had a statis similar. tically significant higher CR rate: 8% as compared to It is worth noting that, in our study, more than 50% % for patients relapsing within months. Therefore, of complete responders to MEC were not eligible for both OS and EFS are significantly poorer in the group BMT. The outcome of this group is very poor, and of patients with a shorter first CR duration (% and alternative strategies should be investigated for patients % at 70 months, respectively) versus patients with a who are ineligible for intensive chemotherapy prolonger first CR duration (8% and % at 5 months, grams after achievement of a second CR [5]. respectively) (Table 5). Only onehalf of remitters were eligible for aggressive postconsolidation treatments. This means that less Acknowledgement than onethird of the initial group of patients actually received the planned myeloablative treatment in sec Supported by C.N.R., Special Project ACRO, Roma, ond CR. This point should be taken into account when Italy. evaluating the different impact of ABMT in the treatment strategy of AML if utilized in first CR or deferred after first relapse. References receiving highdose treatment show a higher probability of becoming longterm diseasefree sur. Preisler HD, Davis RB, Kirshner J et al. Comparison of three vivors: 7/ (%) transplanted patients are in conremission induction regimens and two postinduction strategies for the treatment of acute nonlymphocytic leukemia: A Cancer tinuous CR (CCR) after a median followup of 5 and Leukemia Group B study. Blood 987; 9: 7. months, with a DFS of % projected at 9 months. Phillips GL, Reece DE, Shepherd JD et al. Highdose cytaraversus 8% at 9 months in nontransplanted patients bine and daunorubicin induction and postremission chemo(p = 0.007). Obviously, this is a nonrandomized study therapy for the treatment of acute myelogenous leukemia in and a retrospective comparison cannot confirm a adull Blood 99; 77:95.. Burnett AK, Goldstone AH, Rees J et al. for the MRC Adult superiority for ABMT over chemotherapy in second Pro Table 5. Influence of first CR duration on outcome of AML patients after first relapse. CR rate after MEC Overall survival at 5 months Eventfree survival at 5 months First CR < months First CR > months P % % 8% 8% <0.0 < 0.00 % % <0.05 Abbreviations: CR complete remission; AML acute myelogenous leukemia; MEC mitoxantrone, etoposide, cytarabine; OS overall survival; EFS eventfree survival. Downloaded from on 09 April 08 Leukemia Working Party. The practicability of the comparison of intensive chemotherapy alone or with additional allo or auto BMT for adult AML: Progress in MRC tenth trial. Haematologica 99; 7 (Suppl ): 9 (Abstr).. Stevens RF, Harm IM, Wheatley K et al. Intensive chemotherapy with or without additional bone marrow transplantation in paediatric AML: Progress report on the MRC AML0 trial. Leukemia 99; (Suppl f Mandelli F, Awisati G, Petti MC. Therapy of acute myeloid leukemia: Where from now? In Lechner/Gardner (ed): Haematology Trends '9, Proceedings of the Educational Programme of the Xllth Meeting of the International Society of Haematology (European and African Division). Vienna 99; 8.. Larson RA, Day RS, Azamia N et al. The selective use of AMSA following highdose cytarabine in patients with acute

6 98 myeloid leukaemia in relapse: A Leukemia Intergroup Study. Br J Haematol 99; 8: Miller LP, Pyesmany AF, Wolff LJ et al. Successful reinduction therapy with amsacrine and cyclocytidine in acute nonlymphoblastic leukemia in children; A report from the Childrens Cancer Study Group. Cancer 99; 7: Walters RS, Kantarjian HM, Keating MJ et al. Mitoxantrone and highdose cytosine arabinoside in refractory acute myelogenous leukemia Cancer 988; : Hiddemann W, Kreutzmann H, Straif K et al. Highdose cytosine arabinoside and mitoxantrone: A highly effective regimen in refractory acute myeloid leukemia. Blood 987; 9: Keating MJ, Kantarjian HH, Smith TL et al. Response to salvage therapy and survival after relapse in acute myelogenous leukemia. J Clin Oncol 989; 7: Gorin NC, Dicke K, Lowenberg B. Highdose therapy for acute myelocytic leukemia treatment strategy: What is the choice? Ann Oncol 99; (Suppl ): Wells RJ, Gold SH, Krill KE et al. Cytosine arabinoside and mitoxantrone induction chemotherapy followed by bone marrow transplantation or chemotherapy for relapsed or refractory pediatric acute myeloid leukemia. Leukemia 99; 8:0.. Sobocinski KA, Horowitz MM, Rowlings PA et al. Bone marrow transplantation99: A report from the International Bone Marrow Transplant Registry and the North American Autologous Bone Marrow Transplant Registry. J Hemother 99; : Gorin NC, Labopin M, Meloni G et al. Autologous bone marrow transplantation for acute myeloblastic leukemia in Europe: Further evidence of the role of bone marrow purging by mafosfamide. Leukemia 99; 5: Meloni G, De Fabritiis P, Petti MC et al. BAVC regimen and autologous bone marrow transplantation in patients with acute myelogenous leukemia in second remission. Blood 990; 75: 85.. Meloni G, Vignetti M, Awisati G et al. BAVC regimen for autograft in acute myelogenous leukemia in second complete remission: Updated experience on 0 cases. Seventh International Symposium on Autologous Bone Marrow Transplantation. Texas: Arlington 99; (Abstr 9). 7. Amadori S, Arcese W, Isacchi G et al. Mitoxantrone, etoposide, and intermediatedose cytarabine: An effective and tolerable regimen for the treatment of refractory acute myeloid leukemia. J Qin Oncol 99; 9: Spadea A, Petti MC, Fazi P et al. Mitoxantrone, etoposide and intermediatedose AraC (MEC): An effective regimen for poor risk acute myeloid leukemia. Leukemia 99; 7: World Health Organization: WHO Handbook for Reporting Results of Cancer Treatment. World Health Organization Offset Publication No. 8. Geneva: WHO 979: Meloni G, De Fabritiis P, Papa G et al. Cryopreserved autologous bone marrow infusion following high dose chemotherapy in patients with acute myeloblastic leukemia in first relapse. Leuk Res 985; 9:07.. Tutschka PJ, Copelan EA, Klein JP. Bone marrow transplantation for leukemia following a new busulfan and cyclophosphamide regimen. Blood 987; 70:88.. Kaplan EL, Meier P. Non parametric estimation from incomplete observations. J Am Stat Assoc 958; 5:578.. Cox DR. Regression models and life tables. J R Stat Soc Ser B 97; :870.. Meloni G, Vignetti M, Awisati G et al. BAVC regimen for autograft in acute myelogenous leukemia in second complete remission: Updated results on 0 cases. BMT 99; 8 (in press). 5. Foa R, Fierro MT, Tosti S et al. Induction and persistence of complete remission in a resistant acute myeloid leukemia patient after treatment with recombinant interleukin. Leuk Lymph 990; :7.. Maraninchi D, Blaise D, Viens P et al. High dose recombinant interleukin and acute myeloid leukemias in relapse. Blood 99; 78: Foa R, Meloni G, Tosti S et al. Treatment of acute myeloid leukaemia patients with recombinant interleukin : A pilot study. Br J Haematol 99; 77: Meloni G, Foa R, Capria S et al. IL for the treatment of acute leukemias. Proceedings of 5th International Symposium on Therapy of Acute Leukemias. Leukemia 99; (Suppl ): Mandelli F, Vignetti M, Tosti S et al. Interleukin in acute myelogenous leukemia. Stem Cells 99; : Meloni G, Foa R, Vignetti M et al. Interleukin may induce prolonged remissions in advanced acute myelogenous leukemia. Blood 99; 8: 58.. Meloni G, Vignetti M, Andrizzi C et al. Interleukin treatment in acute myeloblastic leukemia in different phases of disease. Blood 99; 8 (Suppl > 58 (Abstr). Received December 995; accepted September 99. Correspondence to: Giovanna Meloni, MD Hematology, Department of Human Biopathology Universita 'La Sapienza' via Benevento 00 Roma Italy Downloaded from on 09 April 08