Leukemia (2001) 15, 256–263  2001 Nature Publishing Group All rights reserved 0887-6924/01 $15.00 www.nature.com/leu High-dose + melphalan (MITO/L-PAM) as conditioning regimen supported by peripheral blood progenitor cell (PBPC) autograft in 113 lymphoma patients: high tolerability with reversible cardiotoxicity C Tarella1, F Zallio1, D Caracciolo1, A Cuttica1, P Corradini1,2, P Gavarotti1, M Ladetto1, V Podio3, A Sargiotto4, G Rossi5, AM Gianni6 and A Pileri1

1Dipartimento di Medicina e Oncologia Sperimentale, Divisione Universitaria di Ematologia; Serv. 3Univ. e 4Osped di Medicina Nucleare; 5Divisione Universitaria di Radioterapia-Azienda Ospedaliera S Giovanni Battista di Torino, Torino; and 6Unita` Trapianto Midollo, Ist Naz. Tumori, Milano, Italy

Hematological and extrahematological toxicity of high-dose The ease deriving from the excellent tolerability of PBPC (hd) mitoxantrone (MITO) and melphalan (L-PAM) as condition- procedures probably slowed down the efforts aimed to look ing regimen prior to peripheral blood progenitor cell (PBPC) autograft was evaluated in 113 lymphoma patients (87 at dis- for innovative and possibly more effective conditioning regi- ease onset). Autograft was the final part of a hd-sequential mens. In fact, most autograft programs employed nowadays (HDS) program, including a debulkying phase are based on conditioning regimens designed several years (1–2 APO ± 2 DHAP courses) and then sequential adminis- ago, such as TBI + (CY), or regimens con- tration of hd-cyclophosphamide, (or Ara-C) and taining nitrosurea or melphalan (L-PAM), such as BEAM, CBV , at 10 to 30 day intervals. Autograft phase included: or BEAC.17–21 Looking for a better conditioning regimen, one (1) hd-MITO, given at 60 mg/m2 on day −5; (2) hd-L-PAM, given at 180 mg/m2 on day −2; (3) PBPC autograft, with a median of should not only consider the antitumor activity of a given drug 11 × 106 CD341/kg, or 70 × 104 CFU-GM/kg, on day 0. A rapid combination, but should be concerned with its toxicity as hematological recovery was observed in most patients, with well. Today, hematological toxicity is not as severe as it used ANC .500/µL and Plt .20 000/µl values reached at a median to be, due to the availability of mobilized PBPC. Thus, efforts of 11 and 10 days since autograft, respectively. The good hem- should rather be addressed to reduce extra-hematological opoietic reconstitution allowed the delivery of consolidation toxicity of intensive programs. radiotherapy (RT) to bulky sites in 53 out of 57 candidate patients, within 1 to 3 months following autograft; five of these Lymphoma patients seem to benefit from - patients required back-up PBPC re-infusion due to severe post- containing conditioning regimens. Unfortunately, their use at RT pancytopenia. Few severe infectious complications were high dosage has been hampered by the risk of cardiac tox- recorded. There was one single fatal event due to severe pancy- icity.22–24 Mitoxantrone (MITO), a molecule derived from topenia following whole abdomen RT. Cardiac toxicity was anthracenedione, has been proposed as an alternative. Several evaluated as left ventricular ejection fraction (LVEF), monitored studies have shown that this drug can be safely used alone or by cardiac radionuclide scan. LVEF prior to and after autograft in combination, at doses significantly higher than those com- was significantly reduced (median values: 55% vs 46%) in 58 25–30 evaluated patients; however, a significant increase to a median monly employed in conventional protocols. In spite of value of 50% was observed in 45 patients evaluated at 1 to 3 consistently reported good tolerability, there is presently a years since autograft. At a median follow-up of 3.6 years, 92 lack of studies conducted on wide and homogeneous groups patients are alive, with a 7-year overall survival projection and of patients and with an appropriately long follow-up. More- 6.7-year failure-free survival projection of 77% and 69%, over, few studies have been specifically addressed to analyze respectively. We conclude that a conditioning regimen with hd- MITO/L-PAM fits well within the HDS program. It implies good cardiac function prior and following high-dose MITO. tolerability and reversible cardiotoxicity and it may have con- In this study we investigated the tolerability of a combi- tributed to the good long-term outcome observed in this series nation of high-dose MITO and L-PAM in 113 lymphoma of patients. Leukemia (2001) 15, 256–263. patients. Autograft was the final part of a high-dose (hd) Keywords: mitoxantrone; conditioning regimen; autograft; cardi- sequential chemotherapy program.9,10,15 The study was aimed otoxicity; HDS; lymphoma to evaluate at short, medium and long term: (1) hematological toxicity; (2) extra-hematological toxicity and in particular car- diac function evolution; (3) life expectancy, of patients com- Introduction pleting the whole program including autograft. The results indicate that MITO can be safely employed in conditioning In the last decade high-dose chemotherapy and autograft have regimens before PBPC autograft, with an acceptable hematol- significantly improved the outcome of lymphoma patients. ogical and extra-hematological toxicity; there was a certain This approach is presently the first choice for relapsed high cardiac toxicity but this was unexpectedly reversible, at least grade non-Hodgkin’s and Hodgkin’s lymphoma.1–4 Moreover, in part. Finally, the prolonged long-term survival indicates that in the last few years evidence has been produced pointing the combination MITO/L-PAM may contribute to the effective- towards a role for autograft as first-line treatment for both low ness of hd-sequential chemotherapy programs. and high grade poor prognosis non-Hodgkin’s lymphoma.5–12 The wide applicability of this procedure has been possible due to the advantages offered by peripheral blood progenitor cells Patients and methods (PBPC) in reducing hematological toxicity.13–16 Patients Correspondence: C Tarella, Cattedra di Ematologia, Via Genova 3, 10126 Torino, Italy; Fax: 0039–011–696.3737 A retrospective analysis on overall tolerability and cardiac tox- 2Present address: Unita` Trapianto, Ist Sc. S. Raffaele, Milano, Italy icity of the combination of hd-MITO + L-PAM was carried out Received 8 June 2000; accepted 13 October 2000 by collecting data from 113 lymphoma patients treated at the Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 257 University Division of Hematology of Turin, Italy between ing prior to the hd-phase with four APO + two DHAP 1991 and 1998. All patients were part of study protocols (dexametasone/hd-Ara-C/) courses. In the subsequent based on the use of the original or modified high-dose sequen- hd-phase, the CY/VP16 sequence was inverted, in order to tial (HDS) chemotherapy regimens. They received hd-MITO/L- schedule PBPC harvest at the end of the hd-scheme. In PAM followed by PBPC autograft as the conclusive phase of addition, a chemotherapy-free interval of 30 days was the program. The protocols, including the final PBPC auto- included prior to hd-CY in order to allow adequate marrow graft, were approved by the local Ethical Committee and all repopulation and optimal progenitor cell mobilization.31 In patients gave written informed consent to both chemotherapy the chemotherapy-free interval, a total of three hd-dexame- program and autograft. All enrolled patients had a biopsy- thasone courses (dexamethasone at 40 mg/day for 4 consecu- proven diagnosis of lymphoma. They received the HDS pro- tive days) were administered every 10 days. G-CSF (Filgrastim) gram either at disease onset, if they presented with advanced- was given at 5 µg/kg/ day following CY, VP16, Ara-C and stage and poor prognostic features, or at disease relapse. Eligi- autograft, until myelopoietic recovery or completion of har- bility criteria included normal heart, renal and liver functions vesting procedures. The HDS chemotherapy schedules, ie as well as negative tests for HIV, HBsAg and HCV. All patients original HDS, i-HDS and C-HDS, are detailed in Figure 1. All underwent routine staging procedure before starting the patients presenting with bulky disease received additional chemotherapy program; their main clinical features are shown radiotherapy (RT) to bulky sites 1 to 2 months after autograft. in Table 1. Twenty-six patients received the HDS program as salvage after disease recurrence. Their previous treatments consisted of MACOP-B (four patients), CHOP (seven patients), MINE (one HDS treatment schedules patient), (two patients) and ABVD or MAMA hybrid schemes (12 patients). Only six patients had also All patients received either the original (64 patients) HDS received involved field radiotherapy. or one of the two ‘second-generation’ schedules, named i-HDS and C-HDS.9,10,15 The original HDS regimen included the sequential administration of: (1) hd-CY Collection and evaluation of hemopoietic progenitors given at 7 g/m2 i.v., with PBPC harvest at hemopoietic recov- ery; (2) methotrexate (8 g/m2) plus (2 mg i.v.) PBPC were mobilized and collected following hd-CY or, in a approximately at day 16; (3) hd-etoposide (VP16) at 2 g/m2, few patients, following hd-Ara-C or hd-VP16. A BM harvest approximately at day 23; (4) myeloablative treatment with PBPC autograft approximately between days 48 and 52. In 21 high-risk patients, hd-Ara-C replaced hd-methotrexate (C- HDS). To minimize hematological toxicity, a small amount of PBPC (1–2 × 106 CD34+ cells/kg) were re-infused following hd-Ara-C. In addition, a debulkying phase including three APO courses (A = 75 mg/m2;O= Vincristine 1.2 mg/m2, both drugs given at 15–21 day intervals; P = prednisone at 50 mg/m2/day for 15–21 days), was also added. The i-HDS was specifically designed for patients presenting with involvement and it was deliv- ered to 49 patients with low-intermediate grade lymphoma. This latter regimen includes an even more intensive debulky-

Table 1 Main clinical features of 113 lymphoma patients treated with hd-MITO/L-PAM and autograft

Parameter n %

Age, median (range) years 44 (16–61) Sex M/F 64/49 57/43 Diagnosis/Relapsea 87/26 77/23 Histology small lymphocytic 10 9 follicle-center 24 21 mantle cell 7 6 diffuse large cellb 49 43 peripheral T cell 9 8 CD30+ 33 Hodgkin 11 10 Stage III/IV 83 75 BM involvement 42 37 LDH .450 56 49 c > aaIPI 27969Figure 1 Scheme of the original and modified HDS regimens. i-HDS, intensified HDS; C-HDS, hd-Ara-C-HDS; CY, cyclophospham- aTwenty-two patients received HDS as salvage therapy for refrac- ide; MTX, methotrexate; VP16, etoposide; MITO, mitoxantrone; tory or recurrent disease following first-line conventional treatment. L-PAM, melphalan; APO, adriamycin–prednisone–vincristine; DHAP, bFive of them had transformed histology. dexamethasone–hd-Ara-C–cisplatin; CDDP, cisplatin; PBPC, periph- cAge-adjusted International Prognostic Index score. eral blood progenitor cells.

Leukemia Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 258 was performed at the end of the hd-phase in those patients Outcome parameters with low amounts of harvested PBPC (see below). To predict number and timing of leukaphereses, circulating CD34+ cells Physical examination + cell blood count (CBC) and common along with cell blood counts were evaluated daily starting biochemical blood tests were performed daily during recovery from day +9 following chemotherapy administration to har- following autograft. Thereafter, all patients had a complete vesting procedure completion. CD34+ cell evaluation was car- restaging assessment every 3 months during the first year, and ried out according to published procedures.31 Peripheral every 6 months afterwards. A thorough reassessment of renal, blood buffy-coat cells were collected when the WBC count hepatic and cardiac functions were scheduled prior to auto- was at least 1000/µl and PB CD34+ cells .10/µl. Leukapher- graft and at given intervals thereafter. In particular, cardiac eses were performed using continuous-flow blood cell separ- function was monitored by careful clinical examination, EKG ators (Cobe-Spectra, Lakewood, CO, USA or Fresenius, and chest X-ray. No patients had to be excluded from auto- Schweinfurt, Germany) and 7–13 blood liters (median 8.9 l) graft due to cardiac dysfunction. A cardiac radionuclide scan were processed for each procedure. was performed at the start of treatment and prior to autograft Progenitor estimation in leukapheresis products was carried in all patients; in addition, 58 patients had their cardiac func- out by evaluating both CD34+ cells and CFU-GM. The in vitro tion evaluated prior to autograft and within 6 months after colony-forming assay was performed as previously autograft; 45 of these 58 patients were subsequently studied described.31 Values of 30 × 104 CFU-GM/kg and 5 × 106 at 1 to 3 years since autograft. Their clinical characteristics CD34+ cells/kg were taken as the minimal required dose of did not differ from those of the other patients. Thus, this group harvested progenitors for patients to be enrolled in the auto- was highly representative of the whole series of 113 patients. grafting program with PBPC only. In 12 patients where lower Long-term outcome of patients completing the whole pro- amounts of progenitors were harvested, additional BM harvest gram was evaluated as overall survival and failure-free sur- was combined with leukapheresis collection. vival. Survival duration was measured from treatment start to the day of death or last follow-up evaluation. Failure-free sur- vival duration was calculated from autograft up to the first MITO/L-PAM administration and post-autograft adverse event, ie failure to reach complete remission (CR), supportive care relapse, treatment-related or disease-related deaths or the date the patient was last known to be in remission. The actuarial The combined administration of hd-MITO and hd-L-PAM was duration of overall survival (OS) and failure-free survival (FFS) used as conditioning regimen in all patients.25 MITO was were plotted as curves according to Kaplan and Meier pro- given i.v. at 60 mg/m2 in three divided doses of 1 h each every duct-limit method.33 The closing date for analysis was 30 2 h on day −5: this schedule allowed an adequate interval for June 1999. mitoxantrone clearance from plasma prior to PBPC reinfusion, in accordance with reported pharmacokinetic data.25,27,28 L- PAM was given i.v. at 180 mg/m2 in three divided doses of Results 30 min each every 2 h on day −2; PBPC were reinfused 2 days later. Since 1996, G-CSF was added following autograft in Haematological toxicity order to accelerate hemopoietic recovery.32 G-CSF was given i.v. over 1 h at 5 µg/kg/day starting on day +1 until neutrophil Median number of grafted progenitors and time to post-graft count reached 500/µl on 2 consecutive days. The entire pre- hemopoietic recovery are reported in Table 2. There were six transplant program was carried out in ordinary, non-protected patients with delayed recovery; three of them had prolonged rooms, whereas for the autograft phase patients were generally requirement of transfusion support; a rapid hematological admitted to a dedicated inpatient bone marrow transplant recovery was the rule in all remaining patients and values of unit. During the recovery phase following autograft patients ANC .500/µL and Plts .20 000/µl were reached at a median were managed with a common protocol for prophylaxis and of 11 (range 8–31) and 10 (range 7–60) days since autograft, therapy of pancytopenia-related complications. Briefly, anti- respectively. No statistically significant differences were biotic prophylaxis included oral ciprofloxacin (500 mg twice observed in short-term hemopoietic recovery between original a day) + fluconazole (150 mg) + i.v. acyclovir (250 mg three and modified HDS regimens. Indeed, platelet and RBC trans- times a day); in the case of fever .38°C, blood culture and chest X-ray were performed, then patients were empirically treated with i.v. mezlocillin + amikacin + vancomycin; this Table 2 Median number of grafted progenitors and time to post- graft hemopoietic recovery latter is now added only in the case of positive surveillance + culture for gram cocci; if the fever was of undetermined ori- Median Range gin and persisted after 36–48 h, mezlocillin was replaced by imipenem; i.v. amphotericin was added in a few cases with Grafted progenitors fever persisting for 36–48 more hours. Antibiotics were con- CD34+ cells ×106/kg 11.2 1.8–47.6 tinued until the temperature reverted to normal values for at CFU-GM ×104/kg 70.3 3.6–368 least 2 consecutive days along with the return of the ANC to Days to reach .500/µl. Parenteral antibiotics were also started in the WBC .1000/µl 11 (8–40) . µ absence of fever whenever patients were unable to continue ANC 1000/ l 12 (8–60) Plts . 50 000/µl 14 (9–.100)a oral prophylactic antibiotics, due to mucositis. Irradiated, leu- Transfusion requirement kocyte-filtered, single-donor platelet concentrates or, less fre- No. of RBC transfusions (U) 3 (0–120)b quently, multiple-donor irradiated platelet concentrates were No. of Plt transfusions (U) 2 (1–120)b given when platelet count was less than 20 000/µl; irradiated, leukocyte-filtered packed RBCs were given if hemoglobin was aSix patients had delayed plt recovery. less than 8.0 g/dl. bThree patients had prolonged transfusion requirement.

Leukemia Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 259 Table 3 Median cell blood counts recorded at 1, 3 and 6 months Extrahematological toxicity and cardiac function since autograft evaluation

Parameter Time since autograft Autograft was relatively well tolerated, with few severe infec- tious complications, as shown in Table 5. Mucositis of grades 1 month 3 months 6 months III–IV was the most frequent complication, occurring in 35 patients. Fever was also quite common, occurring in the Hb g/dl majority of patients. Mild and reversible elevation of liver median 10.4 11.2 12.4 range (5.8–12.8) (6.6–14.3) (7.2–15.9) enzymes was observed in 10 patients, and bilirubin elevation WBC ×103/µl up to five-fold normal value occurred in 36 patients. How- median 3.1 3.1 4.1 ever, no long-term irreversible liver damage ever occurred. range (0.3–12.0) (0.5–10.8) (1.4–12.0) No evidence of hepatic venous occlusive disease was ever ANC ×103/µl observed nor severe toxicities of the and the urinary median 1.7 1.7 2.4 tract. range (0.1–6.8) (0.2–5.2)a (0.6–6.7) Plts ×103µl There were no severe acute cardiac toxicities during early median 100 77 116 recovery following autograft, with the exception of a transient range (7–494) (9–342)b (15–305)c recurrence of atrial fibrillation. During the first 100 days post- autograft, four patients complained of palpitation, associated aANC were ,0.5 × 103/µl in one patient. with sinus tachycardia, requiring medical therapy in three of bPlts were ,20 × 103/µl in six patients. them (propanolol in two, digoxin in one). Two of these cOne patient showed values of Plts ,20 × 103/µl. patients had disease recurrence and required further chemo- therapy; the other two patients are long-term, disease-free sur- vivors and are no longer symptomatic. One more patient had an overt heart failure occurring immediately after mediastinal fusion requirement was similar for the three regimens RT delivered at 2 months post-autograft. She was an obese employed. patient who had an early disease recurrence, with a rapid and Hemopoietic reconstitution was stable in most patients. This fatal progression. is shown by median CBCs recorded at 1, 3 and 6 months after In the long term, two patients developed symptomatic heart autograft (see Table 3). The stable hemopoietic reconstitution failure: one is a 55-year-old man treated with HDS for disease is further proved by RT feasibility at 1 to 3 months following recurrence after CHOP-based first-line therapy; the other is a autograft. In fact, RT could be delivered to bulky sites in 53 32-year-old obese subject who is now in continuous out of 57 candidate patients. Five more patients required rein- remission of high-grade lymphoma 5 years after HDS + fusion of back-up PBPC due to severe pancytopenia following mediastinal RT. In both patients, symptoms subsided after RT. These patients had been transplanted with a number of medical therapy. These two patients had a marked left ven- progenitor cells that was not out of the range of the whole tricular ejection fraction (LVEF) reduction, as detected by patient population. Unfortunately, there was a toxic death in radionuclide cardiac-scan after autograft. However, both a patient with persistent, severe pancytopenia following whole patients improved their LVEF a few months later, similarly to abdomen RT. The pancytopenia persisted after reinfusion of that observed in the whole patient population (see below). back-up PBPC. This was the single toxic event occurring in One more patient had a transient atrial fibrillation occurring at this series of patients. Long-term engraftment was durable in 4 years after autograft + mediastinal RT which spontaneously all evaluable patients, as shown by CBCs recorded at 1, 3 and subsided. No other symptomatic cardiac disfunctions have 6 years after autograft (Table 4). been recorded among 92 survivors. Thirteen patients had to be rescued due to disease recurrence following HDS. Salvage therapy consisted of hd-chemotherapy and a second autograft in five patients and allogeneic transplant in four patients. All Table 4 Median cell blood counts recorded at 1, 3 and 6 years were able to receive retreatment without major cardiac dys- since autograft functions. Cardiac function was investigated by radionuclide cardiac- Parameter Time since autograft

1 year 3 years 6 years Table 5 Extra-hematological toxicity in 113 patients

Hb g/dl Parameter n % median 12.9 13.7 13.8 range (7.7–15.9) (11–16.2) (9.2–15.4) Fever 97 86 WBC ×103/µl Severe infectionsa 87 median 4.3 5.3 4.8 Grades 3–4 oral mucositis 35 31 range (1.5–1.9) (2.9–10.9) (4.1–11.0) Liver enzyme increaseb 10 9 ANC ×103/µl Hyperbilirubinemiab 36 32 median 2.5 2.9 2.9 Toxic deathc 1 0.9 range (0.6–5.5) (1.7–7.9) (1.8–9.0) Days under i.v. morphine Plts ×103/µl median (range) 6 (0–27) median 134 117 213 a range (12–352) (57–337) (123–414) aFive pneumonia, three CMV reactivations. bLiver enzyme increase was transient and never exceeded three aPlts ,20 000/µl in one patient with secondary myelodysplasia, and times normal value. in one with bone marrow relapse of NHL. cPersistent pancytopenia following post-graft abdominal RT.

Leukemia Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 260

Figure 2 LVEF evaluated in 58 patients by cardiac radionuclide scan immediately prior to and within 6 months after hd-MITO/L-PAM Figure 4 LVEF evaluated by cardiac radionuclide scan within 6 and PBPC autograft. Values are expressed as box plot. months and at 1 to 3 years after hd-MITO/L-PAM and PBPC autograft in 32 patients with marked post-graft LVEF reduction (,40%). Values are expressed as box plot.

scan. Virtually all patients were evaluated at the start of treat- ment and before autograft. LVEF value decreased following played a significant improvement at .1 year following auto- the hd-phase, although the difference was not statistically sig- graft (Figure 4). nificant (data not shown). Three patients underwent autograft As shown in Table 6, the rate of post-graft LVEF reduction with LVEF values slightly below 40%. LVEF values prior to was similar between patients receiving the original or the autograft and within 6 months after autograft could be com- modified HDS schemes as well as between patients receiving pared in 58 patients. Ten of them received mediastinal RT a total anthracyclin dose lower or higher than 200 mg/m2.A after autograft. A LVEF decrease was recorded in 51 patients significant LVEF improvement was recorded in all these sub- and median values recorded prior to and after autograft were groups at .1 year following autograft. In addition, other para- significantly different, as shown in Figure 2. A marked meters, including sex, age (less or more than 45 years old), reduction of LVEF was seen in four out of the 10 evaluable mediastinal RT, previous treatments, did not significantly patients who received mediastinal RT. Finally, 45 out of 58 influence post-graft LVEF reduction and its subsequent patients could be subsequently assessed at 1 to 3 years there- improvement. after. Thirty-four patients displayed a spontaneous improve- ment of their LVEF and median LVEF value recorded at long- term was significantly higher compared to the post-graft evalu- Overall outcome ation (Figure 3). LVEF improvement was not confined to patients with mild or moderate cardiac function impairment; To date, 92 patients are alive at 1 to 7 years since disease in fact, patients with post-graft LVEF lower than 50% also dis- onset. At a median follow-up of 3.8 years, the 7-year OS curve

Table 6 Changes of LVEF at 3–6 months and at .1 year after auto- graft according to treatment schedule and previous anthracyclin dose

Parameter LVEF %

Time since autograft 3–6 months 1–3 years median (range) median (range)

HDS scheme original HDS (n = 14) 45 (38–63) P = 0.017a 50 (45–57) i-HDS and C-HDS (n = 32) 46 (30–59) P = 0.018 50 (35–64) Previous anthracyclin dose <200 mg/m2 (n = 11) 47 (35–66) P = 0.07 50 (39–56) .200 mg/m2 (n = 35) 46 (30–56) P = 0.001 51 (35–64)

aSignificant differences were always found between LVEF values obtained at 3–6 months and those at .1 year. No significant differ- ences were found between patients receiving the original or the Figure 3 LVEF evaluated in 45 patients by cardiac radionuclide modified HDS schedules as well as between patients who had scan within 6 months and at 1 to 3 years after hd-MITO/L-PAM and received less or more than 200 mg/m2 . PBPC autograft. Values are expressed as box plot. i-HDS, intensified HDS; C-HDS, HDS with hd-Ara-C.

Leukemia Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 261 is projected to 77%. Eighty-two patients are surviving in con- The prolonged follow-up enabled us to monitor the long- tinuous complete remission, at 0.5 up to 6.7 years after auto- term CBC pattern. This analysis was obviously carried out only graft. At a median follow-up of 3 years, the 6.5-year FFS curve in disease-free patients, thus in the absence of any further is projected to 69%. Besides the single toxic death, 17 patients cytotoxic treatment. There was no graft failure ever, and CBC had a short-lasting PR or disease progression following HDS; values were constantly good; even those few patients with 13 more patients reached CR and subsequently relapsed. All values lower than normal in the first months following trans- these latter patients were able to receive salvage therapy, con- plant showed a progressive improvement of their CBCs. The sisting of conventional therapy (one patient), RT (three long-term pattern of CBC is consistent with our previous patients), second high-dose chemoradiotherapy with PBPC or observations on PBPC stable and durable engraftment, con- BM autograft (five patients) or allogeneic transplantation (four firming those data on a wider number of cases and with a patients). Salvage therapy was safely delivered without any definitely longer follow-up.38 It has always been our policy to fatal toxicity and 11 out of 13 patients were able to reach a use large numbers of PBPC for autograft. Also in this study prolonged second remission. most patients received .5 × 106 CD34+/kg or .30 × 104 CFU- GM/kg. These values are in line with those previously reported in a study conducted on a large number of patients trans- Discussion planted with PBPC.39 Thus, graft failure is quite unlikely if such PBPC numbers are employed. On the other hand, the The association MITO/L-PAM was first described in 1989.25 long-term CBC outcome in subjects receiving fewer PBPC Several aspects make this drug combination particularly suit- remains unpredictable. able for lymphoma patients: it includes a high-dose anthra- There were no severe extra-hematological complications. cenedione derivative, closely related to the anthracycline fam- Advanced-grade oro-pharyngeal mucositis was very common. ily, a keystone in the treatment of lymphoma; L-PAM is an This complication was probably more pronounced with alkylating-agent particularly effective in lymphoid malig- MITO/L-PAM compared to other schemes. Moreover, low- nancies as well; finally, the absence of TBI makes the scheme grade liver toxicity was also commonly observed in about very practical and easy to use. In spite of these considerations, 30% of cases, with liver function tests going back to normal MITO/L-PAM has not been paid particular attention to date values within a few days. A severe toxic left over was never and, following the original observation, only one other study observed and MITO/L-PAM seems globally to share an over- has evaluated the toxicity and efficacy of this regimen, on a imposable toxicity pattern with other conditioning schemes relatively small series of patients, mostly affected by solid commonly employed in patients with lymphoma. tumor.30 The present study was addressed to evaluate short- Due to the high MITO dose employed, the potential cardiac and long-term outcome in a large series of 113 lymphoma toxicity was carefully evaluated. From a clinical point of view patients receiving hd-MITO/L-PAM followed by PBPC auto- there was no particular adverse event with the exception of graft. All patients have been followed for a minimum of 1 year one case of atrial fibrillation and two episodes of heart failure, and up to 7 years. in any case promptly resolved following medical treatment. Post-transplant engraftment was fast, as usually observed in Heart function was evaluated by cardiac radionuclide scan, autograft with PBPC. The use of G-CSF following PBPC commonly accepted as the most reliable test to assess drug- infusion may have contributed to the rapid ANC recovery as induced cardiac toxicity.40,41 In the present study, LVEF was already reported by others and our group.32,34,35 Blood counts systematically evaluated before and after transplant in as were satisfying in the weeks immediately following short-term many as 58 patients. This function assessment was signifi- engraftment. This is a crucial issue, since it implies the possi- cantly reduced from 55% to 46% median values. Such a bility of delivering full-dose consolidation radiotherapy in reduction is in line with the one mentioned in a previous study most patients requiring it. RT on bulky sites is a very effective on hd-MITO.28 Only very few patients showed post-graft LVEF additional measure in chemotherapy-treated high grade lym- values below 40%. Moreover, we should not forget that in the phoma patients.36 RT constitutes an integral part of our HDS present study the MITO/L-PAM conditioning was preceded by programs, and has presumably contributed to the very good several other cytotoxic treatments, including a preliminar results reported in poor-risk lymphoma patients.10,37 Hence, debulkying with doxorubicin, and a sequence of hd drugs maintainance of good blood counts in the first months follow- such as CY and, in most patients, followed by RT, often deliv- ing autograft is of definite importance in every program ered to the thoracic region. Thus, in spite of other potentially including RT following hd-chemotherapy. cardiotoxic treatments, LVEF reduction following hd-MITO Unfortunately, one patient had a prolonged pancytopenia was globally acceptable and not higher than the one observed following abdominal RT and subsequently died of . with other commonly employed regimens such as CY/TBI.42 This is the only toxic death observed in this series of patients. Post-graft LVEF reduction had little clinical impact. There It is possible that whole abdomen RT following HDS is poorly were only two cases of limited-grade heart failure. In fact, car- tolerated; for this reason we no longer use this approach, and diac function was shown to improve with the passage of time. rather, we prefer limited field RT in the case of abdominal LVEF was better in most cases 2–3 years after transplant, as bulky masses. Four patients could not undergo RT due to low assessed by cardiac radionuclide scan. Heart failure has been CBC and five more patients experienced severe pancytopenia reported as a possible late complication in approximately following consolidation RT. In these patients a further back- 14% of long-term surviving lymphoma patients treated with up PBPC re-infusion was necessary. Thus, following MITO/L- conventional-dose anthracycline-containing regimens.43 PAM, hematological reconstitution is usually good, apart from Anthracycline-linked heart damage dramatically increases a few exceptions raising a certain problem in delivering event- with cumulative doses over 550 mg/m2 and there is little evi- ual consolidation RT in a small though not negligible fraction dence of its reversibility.22–24,43–47 From this viewpoint our of patients. This issue should not be disregarded and we sug- data may be of interest. We are well aware that hd-MITO gest collecting as many PBPC as possible in order to have implies a certain cardiac toxicity, however, this toxicity has extra PBPC if needed. little clinical impact. In addition, the use of a conditioning

Leukemia Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 262 regimen such as MITO/L-PAM following an intensive high- cation with autologous bone-marrow transplantation in relapsed dose sequential schedule does not preclude at least a partial Hodgkin’s disease: results of a BNLI randomised trial. Lancet recovery of cardiac function. A more prolonged follow-up is 1993; 341: 1051–1054. 4 Goldstone AH, McMillan AK. The place of high-dose therapy with required to assess if LVEF recovery can be complete with time. haemopoietic stem cell transplantation in relapsed and refractory Additional studies might verify whether reversibility selec- Hodgkin’s disease. Ann Oncol 1993; 4 (Suppl. I): 21–27. tively applies to MITO-induced cardiotoxicity or whether it 5 Bastion Y, Brice P, Haioun C, Sonet A, Salles G, Marolleau JP, also concerns the damage linked to the other potentially cardi- Espinouse D, Reyes F, Gisselbrecht C, Coiffier B. Intensive therapy otoxic drugs included in the programs. with peripheral blood progenitor cell transplantation in 60 patients Besides tolerability, any conditioning regimen needs careful with poor-prognosis follicular lymphoma. Blood 1995; 86: 3257–3262. evaluation of efficacy as well. Unfortunately, this aspect could 6 Freedman AS, Gribben JG, Neuberg D, Mauch P, Soiffer R, Ander- not be adequately assessed in this study. Autograft was the son K, Pandite L, Robertson M, Kroon M, Ritz J, Nadler R. High- final part of a very intensive hd-program and it is very difficult dose therapy and autologous bone marrow transplantation in to distinguish between MITO/L-PAM activity and the activity patients with follicular lymphoma during first remission. Blood of the hd-phase itself. Moreover, most patients undergoing 1996; 88: 2780–2786. autograft were already in clinical remission. Long-term out- 7 Bierman PJ, Vose JM, Anderson JR, Bishop MR, Kessinger A, Armit- come of grafted patients was very good anyway, in terms of age JO. High-dose therapy with autologous hematopoietic rescue for follicular low-grade non-Hodgkin’s lymphoma. J Clin Oncol both overall and failure-free survival. Thus we can at least 1997; 15: 445–450. conclude that advanced-stage lymphoma patients completing 8 Haas R, Moos M, Mohle R, Dohner H, Witt B, Goldschmidt H, the whole hd-program including autograft have a high chance Murea S, Flentje M, Wannenmacher M, Hunstein W. High-dose of prolonged disease-free survival. The final autograft phase therapy with peripheral blood progenitor cell transplantation in seems to fit well into the hd-sequential program and probably low-grade non-Hodgkin’s lymphoma. Bone Marrow Transplant contributes in a non-measurable way to the good clinical out- 1996; 17: 149–155. 9 Corradini P, Astolfi M, Cherasco C, Ladetto M, Voena C, Caracci- come observed in these patients. In addition, due to its good olo D, Pileri A, Tarella C. Molecular monitoring of minimal tolerability, autograft did not eventually prevent the adminis- residual disease in follicular and mantle cell non-Hodgkin’s lym- tration of rescue treatments in those patients who sub- phomas treated with high-dose chemotherapy and peripheral sequently relapsed. Most of these patients were able to blood progenitor cell autografting. Blood 1997; 89: 724–731. undergo a second-line high-dose regimen with autograft or 10 Gianni AM, Bregni M, Siena S, Brambilla C, Di Nicola M, Lom- allogeneic transplantation, often with achievement of a bardi F, Gandola L, Tarella C, Pileri A, Ravagnani F, Valagussa second remission. Thus, feasibility of effective salvage ther- P, Bonadonna G. High-dose chemotherapy and autologous bone marrow transplantation compared with MACOP-B in aggressive apies may have further improved the overall life expectancy B-cell lymphoma. N Engl J Med 1997; 336: 1290–1295. of this series of patients. 11 Pettengell R, Radford JA, Mogenstern GR, Scarffe JH, Harris M, Dose size is of critical importance in the autograft setting, Woll PJ, Deakin DP, Ryder D, Wilkinson PM, Crowther D. Sur- since the whole antitumor effect relies on the drugs delivered. vival benefit from high-dose therapy with autologous blood pro- Thus, efforts have been directed in the last years to developing genitor-cell transplantation in poor-prognosis non-Hodgkin’s lym- new conditioning regimens including both good tolerability phoma. J Clin Oncol 1996; 14: 586–592. 12 Cortelazzo S, Rossi A, Viero P, Bellavita P, Marchioli R, Marfisi and drug dose intensification.48–50 We think the association of + RM, Rambaldi A, Barbui T. BEAM chemotherapy and autologous hd-MITO L-PAM fits well in this new line of conditioning haemopoietic progenitor cell transplantation as front-line therapy regimens and seems promising for patients with lymphoma. for high-risk patients with diffuse large cell lymphoma. Br J Haematol 1997; 99: 379–385. 13 Gianni AM, Bregni M, Siena S, Villa S, Sciorelli GA, Ravagnani Acknowledgements F, Pellegris G, Bonadonna G. Rapid and complete hemopoietic reconstitution following combined transplantation of autologous blood and bone marrow cells. A changing role for high-dose We thank the laboratory staff and nurses of the Divisione chemo-radiotherapy? Hematol Oncol 1989; 7: 139–148. Universitaria di Ematologia and Centro Dipartimentale Trapi- 14 Gianni AM, Siena S, Bregni M, Tarella C, Stern AC, Pileri A, Bona- anto Midollo – S Giovanni Hospital, Torino for help and donna G. Granulocyte–macrophage colony-stimulating factor to patient care. This work was supported in part by Consiglio harvest circulating haematopoietic stem cells for autotransplan- Nazionale delle Ricerche, Rome, Italy (special project ACRO, tation. Lancet 1989; ii: 580–585. grant No. 96.00742.PF39 to TC and No. 96.00615.PF39 to 15 Tarella C, Gavarotti P, Caracciolo D, Corradini P, Cherasco C, Castellino C, Gallo E, Pileri A. Haematological support of high- AP) and by Associazione Italiana Ricerca sul Cancro, Milan, dose sequential chemotherapy: clinical evidence for reduction of Italy. toxicity and high response rate in poor risk lymphomas. Ann Oncol 1995; 6 (Suppl. 4): 3–8. 16 Schmitz N, Linch DC, Dreger P, Goldstone AH, Boogaerts MA, References Ferrant A, Demuynck HMS, Link H, Zander A, Barge A, Borkett K. Randomised trial of filgrastim-mobilised peripheral blood pro- 1 Philip T, Guglielmi C, Hagenbeek A, Somers R, Van der Lelie H, genitor cell transplantation versus autologous bone-marrow trans- Bron D, Sonneveld P, Gisselbrecht C, Cahn JY, Harousseau JL. plantation in lymphoma patients. Lancet 1996; 347: 353–357. Autologous bone marrow transplantation as compared with sal- 17 Tannir NM, Spitzer G, Zander AR, Jagannath S, Kanojia M, Velle- vage chemotherapy in relapses of chemotherapy-sensitive non- koop L, McLaughlin P, Hagemeister FJ, Dicke KA. High-dose Hodgkin’s lymphoma. N Engl J Med 1995; 333: 1540–1545. chemoradiotherapy and bone marrow transplantation in patients 2 Shipp MA, Abeloff MD, Antman KH, Carrol G, Hagenbeek A, with refractory lymphoma. Eur J Clin Oncol 1983; 19: Loeffler M, Montserrat E, Radford JA, Salles G, Schmitz N, Symann 1091–1096. M, Armitage JO, Ciffier B, Philip T. International Consensus Con- 18 Appelbaum FR, Thomas ED. Review of the use of marrow trans- ference on high-dose therapy with hematopoietic stem cell trans- plantation in the treatment of non-Hodgkin’s lymphoma. J Clin plantation in aggressive non-Hodgkin’s lymphoma: report of the Oncol 1983; 1: 440–447. jury. J Clin Oncol 1999; 17: 423–429. 19 Wheeler C, Antin JH, Churchill WH, Come SE, Smith BR, Bubley 3 Linch DC, Winfield D, Goldstone AH, Moir D, Hancock B, GJ, Rosenthal DS, Rappaport JM, Ault KA, Schnipper LE. Cyclo- McMillan A, Chopra R, Milligan D, Hudson GV. Dose intensifi- phosphamide, and etoposide with autologous bone

Leukemia Reversible cardiotoxicity following HDS and autograft with high-dose mitoxantrone C Tarella et al 263 marrow transplantation in refractory Hodgkin’s disease and non- ery following autologous peripheral-blood stem-cell transplan- Hodgkin’s lymphoma: a dose-finding study. J Clin Oncol 1990; 8: tation in patients with lymphoproliferative disorders: evidence for 648–656. clinical and economic benefits. J Clin Oncol 1997; 15: 451–457. 20 Mills W, Chopra R, McMillian A, Pearce R, Linch DC, Goldstone 36 Shipp MA, Klatt MM, Yeap B, Jochelson Ms, Mauch PM, Rosenthal AH. BEAM chemotherapy and autologous bone marrow transplan- DS, Skarin AT, Canellos GP. Patterns of relapse in large-cell lym- tation for patients with relapsed or refractory non-Hodgkin’s lym- phoma patients with bulk disease: implications for the use of adju- phoma. J Clin Oncol 1995; 3: 588–595. vant radiation therapy. J Clin Oncol 1989; 7: 613–618. 21 van Besien K, Tabocoff J, Rodriguez M, Andersson B, Mehra R, 37 Tarella C, Caracciolo D, Corradini P, Zallio F, Ladetto M, Cuttica Przepiorka D, Dimopoulos M, Giralt S, Suki S, Khouri I. High-dose A, Rossi G, Novero D, Gavarotti P, Pileri A. Long-term follow-up chemotherapy with BEAC regimen and autologous bone marrow of advanced-stage low-grade lymphoma patients treated upfront transplantation for intermediate grade and immunoblastic lym- with high-dose sequential chemotherapy and autograft. Leukemia phoma: durable complete remissions, but a high-rate of regimen- 2000; 14: 740–747. related toxicity. Bone Marrow Transplant 1995; 15: 549–555. 38 Siena S, Bregni M, Di Nicola M, Ravagnani F, Peccatori F, Gan- 22 Von Hoff DD, Layard MW, Basa P, Davis HL Jr, Von Hoff AL, dola L, Lombardi F, Tarella C, Bonadonna G, Gianni AM. Dura- Rozencweig M, Muggia FM. Risk factors for doxorubicin-induced bility of hematopoiesis following autografting with peripheral congestive heart failure. Ann Intern Med 1979; 91: 710–717. blood hematopoietic progenitors. Ann Oncol 1994; 5: 935–941. 23 Ettinghausen SE, Bonow RO, Palmieri ST, Seipp CA, Steinberg SM, 39 Weaver CH, Hazelton B, Birch R, Palmer P, Allen C, Schwartzberg White DE, Rosenberg SA. Prospective study of cardiomiopathy L, West W. An analysis of engraftment kinetics as a function of induced by adjuvant doxorubicin therapy in patients with soft- the CD34 content of peripheral blood progenitor cell collection tissue sarcomas. Arch Surg 1986; 121: 1445–1451. in 692 patients after the administration of myeloablative chemo- 24 Bielack SS, Erttmann R, Kempf-Bielak B, Winkler K. Impact of therapy. Blood 1995; 86: 3961–3969. scheduling on toxicity and clinical efficacy of doxorubicin: what 40 Alexander J, Dainiak N, Berger H, Goldman L, Johnstone D, do we know in the mid-nineties? Eur J Cancer 1996; 32: 1652– Reduto L, Duffy T, Schwartz P, Gottschalk A, Zaret BL. Serial 1660. assessment of doxorubicin cardiotoxicity with quantitative radio- 25 Mulder POM, Sleijfer DT, Willemse PHB, de Vries EGE, Uges nuclide angiocardiography. N Engl J Med 1979; 300: 278–283. DRA, Mulder NH. High-dose cyclophosphamide or melphalan 41 Schwartz RG, McKenzie WB, Alexander J, Sager P, D’Souza A, with escalating doses of MITO and autologous bone marrow trans- Manatunga A, Schwartz PE, Berger HJ, Setaro J, Surkin L, Wackers plantation for refractory solid tumors. Cancer Res 1989; 49: FJ, Zaret B. Congestive hearth failure and left ventricular dysfunc- 4654–4658. tion complicating doxorubicin therapy. Seven-year experience 26 LeMaistre CF, Herzig R. Mitoxantrone: potential for use in inten- using serial radionuclide angiocardiography. Am J Med 1987; 82: sive therapy. Semin Oncol 1990; 17 (Suppl. 3): 43–48. 1109–1118. 27 Canal P, Attal M, Chatelut E, Guichard S, Huguet F, Muller C, 42 Kupari M, Volin L, Suokas A, Timonen T, Hekali P, Ruutu T. Car- Schlaifer D, Laurent G, Houin G, Bugatt R. Plasma and cellular diac involvement in bone marrow transplantation: electrocardio- of mitoxantrone in high-dose chemotherapeutic graphic changes, arrhythmias, heart failure and autopsy findings. regimen for refractory lymphomas. Cancer Res 1993; 53: 4850– Bone Marrow Transplant 1990; 5: 591–598. 4854. 43 Haddy TB, Adde MA, McCalla J, Domanski MJ, Datiles III M, Mee- 28 Attal M, Canal P, Schlaifer D, Chatelut E, Dezeuze A, Huguet F, han SC, Pikus A, Shad AT, Valdez I, Lopez Vivino L, Magrath IT. Payen C, Pris J, Laurent G. Escalating dose of mitoxantrone with Late effects in long-term survivors of high-grade non-Hodgkin’s high-dose cyclophosphamide, carmustine, and etoposide in lymphomas. J Clin Oncol 1998; 16: 2070–2079. patients with refractory lymphoma undergoing autologous bone 44 Freter CE, Lee TC, Billingham ME, Chak L, Bristow MR. Doxorub- marrow transplantation. J Clin Oncol 1994; 12: 141–148. icin cardiac toxicity manifesting seven years after treatment. Am J Med 1986; 80: 483–485. 29 Feldman EJ. High-dose mitoxantrone in acute leukemia: New York 45 Moreb JS, Oblon DJ. Outcome of clinical congestive heart failure Medical College experience. Eur J Cancer Care 1997; 6 (Suppl. 1): induced by anthracycline chemotherapy. Cancer 1992; 70: 27–32. 2637–2641. 30 Ballestrero A, Ferrando F, Garuti A, Basta P, Gonella R, Esposito 46 Sorensen K, Levett G, Bull C, Chessells J, Sullivan I. Anthracycline M, Vannozzi MO, Sorice G, Friedman D, Puglisi M, Brema F, dose in childhood acute lymphoblastic leukemia: issues of early Mela GS, Sessarego M, Patrone F. High-dose mitoxantrone with survival versus late cardiotoxicity. J Clin Oncol 1997; 15: 61–68. peripheral blood progenitor cell rescue: toxicity, pharmacokinet- 47 Nysom K, Holm K, Lipsitz SR, Mone SM, Colan SD, Orav EJ, Sal- ics and implications for dosage and schedule. Br J Cancer 1997; lan SE, Olsen JH, Hertz H, Jacobsen JR, Lipshultz SE. Relationship 76: 797–804. between cumulative anthracycline dose and late cardiotoxicity in 31 Tarella C, Caracciolo D, Gavarotti P, Bondesan P, Cherasco C, childhood acute lymphoblastic leukemia. J Clin Oncol 1998; 16: Omede` P, Bregni M, Siena S, Gianni M, Pileri A. Circulating pro- 545–550. genitors following high-dose sequential (HDS) chemotherapy with 48 Heyll A, Sohngen D, Kobbe G, Schneider P, Bauser U, Thiele KP, G-CSF: short intervals between drug courses severely impair pro- Wehmeier A, Sudhoff T, Quenzel EM, Rieth C, Werneth P, Fischer genitor mobilization. Bone Marrow Transplant 1995; 16: 223– J, Frick M, Aul C. , melphalan and cyclophosphamide: 228. an intensified high-dose regimen for the treatment of myeloma 32 Tarella C, Castellino C, Locatelli F, Caracciolo D, Corradini P, patients. Leukemia 1997; 11 (Suppl. 5): S32-S34. Falda M, Novarino A, Tassi W, Pileri A. G-CSF administration fol- 49 Benedetti Panici P, Pierelli L, Scambia G, Foddai ML, Salerno MG, lowing peripheral blood progenitor cell (PBPC) autograft in lymph- Menichella G, Vittori M, Maneschi F, Caracussi U, Serafini R, oid malignancies: evidence for clinical benefits and reduction of Leone G, Mancuso S. High-dose , etoposide and mel- treatment costs. Bone Marrow Transplant 1998; 21: 401–407. phalan (CEM) with peripheral blood progenitor cell support as late 33 Kaplan EL, Meier P. Nonparametric estimation from incomplete intensification for high-risk cancer: non-haematological, haemato- observation. J Am Stat Assoc 1958; 53: 457–481. logical toxicities and role of growth factor administration. Br J Can- 34 Klumpp TR, Mangan KF, Goldberg SL, Pearlman ES, MacDonald cer 1997; 75: 1205–1212. JS. Granulocyte colony-stimulating factor accelerates neutrophil 50 Stiff PJ, Dahlberg S, Forman SJ, McCall AR, Horning SJ, Nade- engraftment following peripheral-blood stem-cell transplantation: manee AP, Blume KG, LeBlanc M, Fisher R. Autologous bone mar- a prospective, randomized trial. J Clin Oncol 1995; 13: 1323– row transplantation for patients with relapsed or refractory diffuse 1327. aggressive non-Hodgkin’s lymphoma: value of augmented pre- 35 McQuaker IG, Hunter AE, Pacey S, Haynes AP, Iqbal A, Russell parative regimens - a Southwest Oncology Group trial. J Clin NH. Low-dose Filgrastim significantly enhances neutrophil recov- Oncol 1998; 16: 48–55.

Leukemia