Cardiac Systolic Function in Patients Receiving Hematopoetic Stem Cell Transplantation: Risk Factors for Posttransplantation Cardiac Toxicity

Cardiac Systolic Function in Patients Receiving Hematopoetic Stem Cell Transplantation: Risk Factors for Posttransplantation Cardiac Toxicity G.T. Sucak, Z.N. Ozkurt, Z. Akı, M. Yag៮ cı, A. Çengel, and R. Haznedar ABSTRACT One hundred eleven patients who received 125 hematopoetic stem cell transplantations (HSCT) with myeloablative conditioning regimens were retrospectively evaluated for the development of cardiac toxicity (CT). The aims of this study were to assess the frequency of cardiac complications in patients receiving HSCT and to investigate the value of pretransplantation variables to predict posttransplantation CT. Severe grade III–IV CT was not observed in this cohort, in whom pretransplantation eligibility criteria excluded the patients with a left ventricular ejection fraction (LVEF) of 50% or less. Grade I–II CT was seen in 13.4% patients. Patients with a history of previous mediastinal radiotherapy, high doses of anthracycyclines, and a longer interval between diagnosis and treatment were found to have higher risk of developing CT. Pretransplantation ferritin levels and the type of HSCT did not seem to have an effect on posttransplantation cardiac complications. Our results indicated that CT was managable in patients with a LVEF of at least 50%.

H

EMATOPOETIC stem cell transplantation (HSCT) has an expanding role in the treatment of benign and malignant hematological disorders. However, HSCT remains a procedure with high treatment-related mortality. Assessment of organ function prior to HSCT has been a routine part of the pretransplantation work-up for more than 30 years; it is considered to be an important predictor of regimen-related toxicity, which mainly includes renal, respiratory, hepatic, and mucosal injury. HSCT has been associated with a low prevalence of cardiac mortality because this treatment modality is usually confined to low– cardiac risk patients.1 Although the correlation of pretransplantation function and posttransplantation cardiac toxicity (CT) is questioned, several reports have confirmed that patients with a left ventricular ejection fraction (LVEF) of ⬍50% show a greater incidence of CT.1 Determining the risk factors and transplanting vulnerable patients with impaired cardiac function, with noncardiotoxic protocols, could further decrease CT among HSCT patients. Acute cardiac problems, such as arrhythmias, heart failure, and cardiac tamponade, have been reported in 1%–9% of patients in various studies.1–5 High-dose cyclophosphamide6,7 and/or total-body irradiation (TBI) are treatment modalities known to be associated with CT.8 High-dose anthracycline before transplantation has also been reported to be a cardiac risk factor for HSCT patients.2,3 Cardiac

performance has been assessed by measuring LVEF with echocardiography or radionuclide ventriculography (RVG). However, it remains unclear whether LVEF is sufficient to assess the risk of CT, or whether a low LVEF per se is an obstacle for HSCT.2,6,9 At the same time, the optimal method for cardiac follow-up after transplantation is still not clear; there are no adequate studies documenting long-term cardiac function. In this retrospective analysis, we sought to determine the acute CT frequency, the risk factors for CT, and changes in cardiac functions evaluated with both (RVG) and echocardiography during long-term follow-up among a cohort of HSCT patients. MATERIALS AND METHODS Patients

We retrospectively analyzed the acute CT in 111 consecutive patients (43 females, 68 males) who had undergone 125 HSCTs, including 14 patients with double/tandem transplants between September 2003 and December 2006. Of the 125 HSCT, 52 were autologous and 73 allogeneic transFrom the Department of Hematology, Gazi University Faculty of Medicine, Ankara, Turkey. Address reprint requests to Dr Gulsan T. Sucak, Gazi University Faculty of Medicine, Ankara, Cankaya 06500, Turkey. E-mail: [email protected]

0041-1345/08/$–see front matter doi:10.1016/j.transproceed.2007.11.077

© 2008 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710

1586

Transplantation Proceedings, 40, 1586 –1590 (2008)

CARDIAC TOXICITY

1587

plants. Twenty-six patients had acute myeloid leukemia (AML), 15 had acute lymphoblastic leukemia (ALL), 35 had multiple myeloma (MM), 10 had severe aplastic anemia (SAA), 8 had non-Hodgkin’s Lymphoma (NHL), 13 had Hodgkin’s disease (HD), and 4 patients had chronic myeloid leukemia (CML). Patient history, physical examination, prior chemotherapy radiotherapy records, transplantation protocols, and conditioning regimens were recorded per institutional practice in all patients. The median age for all subjects, for allogeneic, and for autologous recipients were 37.0, 28.0, and 52.0 years, respectively. The mean pretransplantation and 1 month posttransplantation LVEF values were 65.9 ⫾ 5.1% (range, 50 –77) and 65.1 ⫾ 5.8% (range, 40 –75), respectively. Patient characteristics are shown in Table 1. Thirty-seven patients died during posttransplantation follow-up secondary to noncardiac causes. Twelve of these patients died within posttransplantation 30 days; 6 patients, between 1 and 3 months; 11 patients, between 3 and 6 months; and 8 patients, between 6 and 12 months. Only 2 patients had a posttreatment LVEF ⬍50%. Overall there was no significant decrease in LVEF compared with baseline values (P ⬎ .05). The cumulative doses of anthracyclines were obtained from patient records and converted to the equivalent dose of doxorubicin, assuming that CT at an equal dose is 0.5 for daunorubicin, 3.4 for mitoxantrone, and 1.6 for idarubicin.2,10 The doses were categorized as low (0 –199 mg/m2) and intermediate to high (⬎200 mg/m2).

Conditioning Regimens

Busulfan-based regimens (Busulfan ⫹ cyclphosphamide ⫾ Thiotepa or ⫾ fludarabine, Busulfan ⫹ fludarabine ⫾ ATG; n ⫽ 42); TBI-based regimens (TBI ⫹ cyclphosphamide ⫾ Thiotepa, TBI ⫹ fludarabine, TBI ⫹ melphalan; n ⫽ 18); cyclophosphamide ⫾ ATG (n ⫽ 9); Melphalan (n ⫽ 35); BCNU ⫹ Etoposide ⫹ ARA-C ⫹ Melphalan (BEAM; n ⫽ 15); and other protocols (n ⫽ 6) were performed during the transplantation procedure. All paTable 1. Patient Characteristics Median age (range) Gender (male/female) Autologous transplantation Allogeneic transplantation

Primary disease MM AML ALL HD SAA NHL CML

37.0 y (16–71) 63/48 52 1MM retransplantation 73 2MM, 2 NHL, 1 HD tandem 4 SAA, 3 AML, 1 ALL retransplantation 35 26 15 13 10 8 4

tients receiving autologous HSCT were mobilized with cyclophosphamide/VP 16. Cardiac Evaluation

Electrocardiogram, two-dimensional trans-thoracic echocardiography, chest X-ray, and RVG were part of the routine evaluation before as well as 1, 3, 6, and 12 months after HSCT. In case of a discrepancy in the LVEF value between echocardiography and RVG, the results of echocardiogram were considered for the eligibility criteria. Regimen-related acute cardiotoxicity developed within 28 days as defined according to National Cancer Institute (NCI) toxicity criteria. Patients were monitored for a year in terms of cardiac and other organ toxicities per institutional practice. Statistical Analysis

Continuous variables in the 2 groups were compared using Student t test or the Mann-Whitney U test. Categorical variables were compared using the ␹2 test. Paired Student t test was used to compaire values before and after HSCT. The calculations were performed with SPSS 11.5 (SPSS Inc., Chicago, Ill, United States). RESULTS

Before HSCT, the LVEF evaluated by echocardiography was ⱖ50% in all patients (range, 50 –77; mean %, 65.9 ⫾ 5.1) and the mean LVEF evaluated with RVG was 55.6 ⫾ 7.4 (range, 36.0 –74.0; n ⫽ 92). Pretransplantation LVEF values measured with echocardiography significantly correlated with RVG (P ⫽ .02). CT was diagnosed in 17 (13.4%) patients. There was grade I CT in 11 patients (8.7%), and grade II in 6 patients (4.7%). Three patients had asymptomatic, repeating, temporary arrythmias; 6 patients had systolic dysfunction that did not cause heart failure; 3 patients had nonspecific ST-T wave abnormalities, and 5 patients had self-limited pericardial effusion. Grades III and IV CT were not observed. LVEF values evaluated prior to HSCT were not different both by echocardiography and RVG (P ⬎ .05) among groups with versus without CT. Characteristics of the patients with versus without acute CT are shown in Table 2. Anthracycline exposure was present in 96 patients (86.5%) during pretransplantation treatment. Anthracycline types were doxorubicin in 59 patients; idarubicin in 21 patients; daunorubicin in 3 patients; mitoxantrone in 2 patients; daunorubicin ⫹ idarubicin in 5 patients; idarubicin ⫹ doxorubicin in 1 patient; and idarubicin ⫹ daunorubicin in 1 patient. Cumulative dose of anthracyclines before HSCT were higher among patients with (193.3 ⫾ 116.3 mg) compared with those without CT (126.1 ⫾ 101.0 mg; P ⫽ .014; Fig 1). CT was more frequent among patients who previously received a total of at least 200 mg anthracyclines (8/18, 47.1% vs 9/90, 9.9%; P ⫽ .008) and in patients who received mediastinal radiotherapy (5/12, 29.4% vs 6/102, 5.5%; P ⫽ .007; Fig 2). The interval between the diagnosis and transplantation was longer in

1588

SUCAK, OZKURT, AKI ET AL Table 2. Patient Characteristics and Cardiac Toxicity Acute CT n ⫽ 125 HSCT

Present (n ⫽ 17) 13.4%

Absent (n ⫽ 108) 86.6%

8/9 36,5 8/9 46/62 1398.9 581 193.4 9 8 5/12

69/39 37 44/64 14/3 1306.2 305 126.1 90 18 6/102

.193 .494 .732 .002* .85 .026 .014* .008*

65.1 ⫾ 5.8 52.7 ⫾ 9.5 3/14 8/9 36/72

66.1 ⫾ 5.0 56.1 ⫾ 6.9 14/94 52/56 6/11

.329 .279 .702 .571 .538

Gender (male/female) Median age (y) Stem cell (auto/allo) Pretransplantation CT cycles ⱖ2/⬍2 Pretransplantation ferritin levels Interval between diagnosis and transplantation Cumulative dose of anthracycline Low (0–200 mg/m2) Intermediate or high (⬎200 m2) Mediastinal radiotherapy LVEF Ecocardiography RVG TBI-containing regimen (⫹/⫺) Cyclophasphamide-containing regimen (⫹/⫺) Busulfan-containing regimen (⫹/⫺)

P

.007*

*Indicates numbers with statistical significance.

patients who developed CT than those without CT (P ⫽ .026; Fig 3). There was a significantly positive correlation between diagnosis-transplantation period and cumulative anthracycline doses (P ⬍ .001; r ⫽ 0.387). There was no difference between CT frequency according to the conditioning regimen and transplantation type (P ⬎ .05). There were also no differences between LVEF before and at 1, 3, 6, or 12 months posttransplantation with both techniques (P ⬎ .05). A decrease in LVEF ⬎10% with respect to the basal values in the first month posttransplantation was more frequent among patients with (3/9) than

Fig 1. Cumulative anthracycline doses (mg) and cardiac toxicity.

those without CT (3/91; P ⫽ .002). However, during 3- and 6-month evaluations posttransplantation, only patients with CT showed percentage reduction of LVEF upon echocardiography to be significantly lower than the values before transplantation (P ⬍ .05; P ⫽ .05), whereas LVEF values simultaneously determined with multiple-gated acquisition were not different (P ⬎ .05). LVEF values evaluated with both techniques at 12 months after SCT were not different (P ⬎ .05). Among 125 HSCT, 77/125 patients’ stem cells were cryopreserved with DMSO and 48/125 patients’ stem cells were infused as a fresh product. CT was seen in 11/77 patients whose stem cells were cryopreserved (14.3%),

Fig 2.

Mediastinal radiotherapy and cardiac toxicity.

CARDIAC TOXICITY

Fig 3. Cardiac toxicity interval between diagnosis and transplantation (days).

whereas CT was seen in 6/48 patients whose stem cells were infused as a fresh product (12.5%; P ⬎ .05). While all autologous HSCT patients received cryopreserved stem cells, 26/73 allogeneic HSCT patients received cryopreserved stem cells, and 47/73 allogeneic HSCT patients received fresh stem cells. There was no difference among allogeneic stem cell transplant recipients whose stem cells had been exposed to dimethyl sulfoxide (DMSO; P ⬎ .05). Another factor, which is well known to have a negative impact on transplant-related mortality during HSCT, is iron overload.11 In this respect pretransplantation ferritin levels were analyzed in patients with (mean, 1398.9 ng/mL) versus without (1306.2 ng/mL) CT. There was no difference in ferritin levels between the 2 groups (P ⬎ .05). DISCUSSION

High-dose cytotoxic chemoradiotherapy is a major challenge for organ functions. Lung, kidneys, liver, and gastrointestinal mucosa are particularly vulnerable to the myeloablative conditioning protocols of HSCT. CT is relatively less frequent in HSCT patients compared with other major organ toxicities. Similarly, we have not seen grade III–IV CT in our cohort of 125 HSCT, whereas grade I–II toxicities were seen in 13.4% of patients. Considering the fact that most patients received high-dose cyclophosphamide during conditioning (48%) and all of the autologous HSCT patients were mobilized with high-dose cyclophosphamide an agent frequently associated with CT,5 this relatively high ratio of CT, although not severe, was found to be acceptable. One possible explanation for the absence of severe CT was the exclusion of patients with LVEF of ⱕ50% from our myeloablative HSCT protocols. Similarly Bearman et al4 reported 4% grade III–IV CT among patients whose pre-

1589

transplantation LVEF was ⱖ50%. Hertenstein et al, similar to our results, found even less (1%) major CT among their cohort of 148 patients.1 Fujimaki et al, in their report of 80 HSCT patients evaluated for CT, found a significant association between pretransplantation LVEF and early CT.9 Whereas 43% of patients with a LVEF of ⬍55% developed severe CT, none with a LVEF ⬎55% developed major CT. Although pretransplantation LVEF was not different among patients with versus without posttransplantation CT in our study group, it should be noted that all the patients in our cohort had LVEF ⬎50%. The absence of severe acute CT might indicate that a meticulous pretransplantation cardiac evaluation may save high-risk patients from severe toxic events. However, it should be noted that LVEF alone might not be sufficient to determine all high-risk patients. Patients with Thallassemia, Sickle cell disease, and autoimmune disorders usually have diastolic malfunction, which might be overlooked while only assessing systolic function. Coghlan et al recommend a stepwise approach to assess pretransplantation cardiac risk factors in a recently published review.12 They suggested that N-terminal pro-brain natriuretic peptide (NT-proNP), troponin, Holter monitoring, and even cardiac catheterization should be included in the diagnostic algorithm according to risk status. However, because none of our patients had an underlying disease with a high risk of diastolic malfunction, we believe that none of our patients required further testing according to this stepwise approach. In the current study, there was no significant difference between the basal LVEF values evaluated with echocardiography and RVG for patients with versus without CT. Patients who were exposed to high-dose anthracycline, who had a long interval between diagnosis and transplantation, who underwent more chemotherapy cycles, and who had a history of mediastinal radiotherapy were observed to experience an increased acute CT frequency during HSCT. Similarly, Sakata-Yanagimoto et al2 reported cumulative pretransplantation anthracycline doses to be the most potent predictor of CT. On the other hand, Lehmann et al13 reported no difference between the LVEF values evaluated with RVG performed between 22 and 220 days after transplantation among patients who did versus did not receive high-dose anthracycline. Similarly, in our retrospective analysis, with regular follow-up, we observed no difference in LVEF measured with RVG during the posttransplantation 12 months. An additional cause of CT may have been the infusion of freshly thawed DMSO-containing stem cells. Cardiac complications, usually mild, such as bradycardia and hypertension, have been reported with DMSO.5,14 However, there are some anecdotal reports that have associated CT and death induced by DMSO-containing grafts.15 We have not observed any difference between patients receiving cryopreserved or freshly thawed stem cells. The difference between allogeneic and autologous HSCT was also investigated. There are previous reports that CT

1590

was more common among autologous HSCT patients.1 More intensive chemoradiotherapy prior to HSCT was stated to be the cause of more frequent CT in autologous HSCT patients. On the other hand, cyclosporine (CsA), an agent used for GVHD prophylaxis, is well known to have long-term cardiac complications, such as hypertension and hyperlipidemia,16 which may contribute to CT. However, we did not observed any difference in acute CT between allogeneic HSCT patients receiving CsA and autologous patients who do not. In conclusion, acceptable CT occurs with myeloablative conditioning regimens among patients with LVEF ⱖ50% before HSCT. CT of HSCT is self limited and manageable if patients are scrutinized prior to transplantation. MUGA has no additional benefits to estimate and follow CT in the pretransplantation and posttransplantation periods. An increased cumulative dose of anthracycline, previous mediastinal radiotherapy, and longer diagnosis-transplantation intervals were risk factors for CT. Switching to induction chemotherapy regimens without or with lower doses of anthracylines and considering transplantation earlier in the disease course whenever possible could decrease the CT associated with HSCT. Strategies such as nonmyeloablative and noncardiotoxic conditioning regimens may be suggested for patients with LVEF ⬍50% prior to transplantation.

REFERENCES 1. Hertenstein B, Stefanic M, Schmeiser T, et al: Cardiac toxicity of bone marrow transplantation: predictive value of cardiologic evaluation before transplant. J Clin Oncol 12:998, 1994 2. Sakata-Yanagimoto M, Kanda Y, Nakagawa M, et al: Predictors for severe cardiac complications after hematopoietic stem cell transplantation. Bone Marrow Transplant 33:1043, 2004

SUCAK, OZKURT, AKI ET AL 3. Murdych T, Weisdorf DJ: Serious cardiac complicarions during bone marrow transplantation at the University of Minnesota, 1977–1997. Bone Marrow Transplant 28:283, 2001 4. Bearman SI, Peterson FB, Schor RA, et al: Radionuclide ejection fractions in the evalution of patients being considered for bone marrow transplantation: risk for cardiac toxicity. Bone Marrow Transplant 5:173, 1990 5. Cazin B, Gorin NC, Laporte JP, et al: Cardiac complications after bone marrow transplantation. A report on a series of 63 consecutive transplantations. Cancer 57:2061, 1986 6. Braverman AC, Antin JH, Plappert MT, et al: Cyclophosphamide cardiotoxicity in bone marrow transplantation: a prospective evalution of new dosing regimens. J Clin Oncol 9:1215, 1991 7. Goldberg MA, Antin JH, Guinan EC, et al: Cyclophosphamide cardiotoxicity: an analysis of dosing as a risk factor. Blood 68:1114, 1986 8. Stewart GR, Fajardo LF: Radiation-induced heart disease: an update. Prog Cardio Dis 17:173, 1984 9. Fujimaki K, Maruta A, Yoshida M, et al: Severe cardiac toxicity in hematological stem cell transplantation: predictive value of reduced left ventricular ejection fraction. Bone Marrow Transplant 27:307, 2001 10. Herait P, Poutignat N: Early assessment of a new anticancer drug analogue-are the historical comparisons obsolete? The French experience with pirarubicin. Eur J Cancer 28A:1670, 1992 11. Armand P, Kim HT, Cutler CS, et al: Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation. Blood 109:4586, 2007 12. Coghlan JG, Handler CE, Kottaridis PD, et al: Cardiac assessment of patients for haematopoietic stem cell transplantation. Best Pract Res Clin Haematol 20:247, 2007 13. Lehmann S, Isberg B, Ljungman P, et al: Cardiac systolic function before and after hematopoietic stem cell transplantation. Bone Marrow Transplant 26:187, 2000 14. Davis JM, Rowley SD, Braine HG, et al: Clinical toxicity of cryopreserved bone marrow graft infusion. Blood 75:781, 1990 15. Zenhausern R, Tobler A, Leoncini L, et al: Fatal cardiac arrhythmia after infusion of dimethyl sulfoxide-cryopreserved hematopoietic stem cells in a patient with severe primary cardiac amyloidosis and end-stage renal failure. Ann Hematol 79:523, 2000 16. Rabkin JM, Corless CL, Rosen HR, et al: Immunosuppression impact on long-term cardiovascular complications after liver transplantation. Am J Surg 183:595, 2002