Early cardiotoxicity of the CHOP regimen in aggressivenon-Hodgkin’s lymphoma

Annals of Oncology 14: 277–281, 2003 DOI: 10.1093/annonc/mdg070

Original article

Early cardiotoxicity of the CHOP regimen in aggressive non-Hodgkin’s lymphoma S. Limat1, K. Demesmay1, L. Voillat2, Y. Bernard3, E. Deconinck2, A. Brion2, A. Sabbah4, M. C. Woronoff-Lemsi1 & J. Y. Cahn2* Departments of 1Pharmacy, 2Hematology, 3Cardiology and 4Nuclear Medicine, Besançon University Hospital, Besançon, France Received 23 April 2002; revised 12 August 2002; accepted 16 September 2002

Background: To determine the incidence of early cardiotoxicity induced by the CHOP regimen in patients with aggressive non-Hodgkin’s lymphoma (NHL) and to identify associated risk factors.

Patients and methods: A retrospective analysis included 135 consecutive patients who had been treated with the CHOP (cyclophosphamide, doxorubicin, vincristin, prednisone) regimen as first-line therapy between 1994 and 2000. The cardiac evaluation was based on a determination of the resting left ventricular ejection function (LVEF) by gated blood-pool imaging. Cardiotoxicity was defined as a significant decrease in LVEF or clinical evidence of congestive heart failure (CHF). Results: Twenty-seven (20%) patients developed a cardiac event within 1 year of treatment. Among these, 14 patients had clinical signs of CHF. Three patients died suddenly from presumed cardiac causes. In multivariate analysis, a cumulative dose of doxorubicin >200 mg/m2 [odds ratio (OR) = 4.2, P = 0.005)] and age over 50 years (OR = 2.9, P = 0.03) appeared to be significant risk factors. Conclusion: Early clinical and subclinical cardiotoxicity was frequent in patients receiving the CHOP regimen. The threshold of the cumulative dose of doxorubicin appeared to be low: at doses >200 mg/m2, 27% of patients had cardiac events. Elderly patients appeared to be at higher risk. The development of cardioprotective strategies or alternative treatments are mandatory for aggressive NHL patients. Key words: cardioprotection, cardiotoxicity, congestive heart failure, doxorubicin, non-Hodgkin’s lymphoma

Introduction As anthracyclines are still accepted as highly efficient treatments for various solid tumours and haematological malignancies, cardiac toxicity remains a key problem in clinical practice [1–4]. High rates of mortality have been attributed to induced congestive heart failure (CHF) [1, 5–7]. The majority of available data have been reported in childhood and breast cancers. The definition of cardiotoxicity varied in these studies [4]. Early (within 1 year of treatment) and late onset (>1 year after treatment) cardiac abnormalities are commonly distinguished and toxic effects have been defined as clinical or subclinical heart failure [2, 4]. The cumulative incidence of CHF varied from 1.6% to 2.8% [1, 6, 8]. Using more sensitive investigations, Lipschutz et al. [9] showed that 57% of long-term survivors had cardiac abnormalities. In other studies, the incidence of left ventricular ejection (LVE) dysfunction varied from 21% to 31% [10–13].

The CHOP (cyclophosphamide, doxorubicin, vincristin, prednisone) regimen is still considered the gold standard in firstline therapy [14]. Surprisingly, there are few data concerning the cardiac consequences of the systematic use of doxorubicin for patients with aggressive non-Hodgkin’s lymphoma (NHL). Only one study has shown that cardiac toxicity was the predominant late chemotherapy-related effect in long-term survival [15]. The potential cardiotoxicity of first-line therapy should be considered, in relation to the incidence and the long-term prognosis of aggressive NHL patients. A retrospective study was carried out to analyse the early doxorubicin-induced cardiotoxicity in aggressive NHL patients. The primary objective was to determine the incidence of cardiac abnormalities within 1 year of treatment using the CHOP regimen. The secondary objective was the identification of the risk factors of cardiotoxicity.

Patients and methods *Correspondence to: Professor J. Y. Cahn, Department of Haematology, University Hospital of Besançon, 2 Boulevard Fleming, 25030 Besançon, France. Tel: +33-3816-684-04; Fax: +33-3816-682-15; E-mail: [email protected] © 2003 European Society for Medical Oncology

Selection All patients with a diagnosis of aggressive NHL and treated at the Besançon University Hospital between 1 January 1994 and 31 October 2000 were

278 selected. Patients treated with the CHOP regimen (cyclophosphamide 750 mg/m2 day 1, doxorubicin 50 mg/m2 day 1, vincristin 1.4 mg/m2 day 1, prednisone 100 mg/m2 days 1–5) as first-line therapy were enrolled in this study. In the case of neurotoxicity, vincristin could have been replaced by vindesin (3 mg/m2) or etoposide (100 mg/m2). During the period of the study, doxorubicin was administered as a short i.v. infusion. Patients who showed evidence of any cardiac abnormality before the start of chemotherapy were excluded from the study. All patients were informed of the potential toxicity of chemotherapy, including cardiotoxicity, at the onset of treatment. The current modalities of the cardiac function evaluation were presented. In the case of inclusion in a clinical trial, written informed consent was obtained from all patients, and studies were designed and conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committees.

Early cardiotoxicity The study was limited to the first year of follow-up, including the period of treatment [2]. According to local practice, cardiac evaluation included baseline physical examination, ECG and determination of the resting left ventricular ejection function (LVEF) by gated blood-pool (GBP) imaging (nuclear scan) [16, 17]. Gated blood-pool imaging was routinely performed before treatment, after four cycles and at the end of first-line chemotherapy (maximum of eight cycles) and/or in the case of clinical abnormalities. All examinations were performed in the same institution. The institution’s lower limit of normal value of LVEF was 53%. Cardiac events were defined as either a decline in resting LVEF of ≥15% from baseline, or a decline in LVEF of <50%, or clinical evidence of CHF. In patients receiving other treatments [salvage therapy and/or high-dose chemotherapy (HDC)] within 1 year, only cardiotoxicity detected during or at the end of CHOP therapy was considered. All potentially positive (subclinical or clinical toxicity) cases were reviewed by an independent and experienced cardiologist.

Table 1. Patient characteristics Characteristics

No. of patients (%)

Sex Male

77 (57)

Female

58 (43)

Age at treatment (years) Median

59

Range

25–79

Pre-existing cardiac risk factors No

69 (51.1)

Yes

66 (49.9)

Initial LVEF value Median

0.68

Range

0.53–0.87

Cumulative dose of doxorubicin (mg/m2) Median Range

290 50–400

LVEF, left ventricular ejection function.

seven patients had pre-existing cardiac abnormalities before the onset of chemotherapy, four patients died during the treatment without drug-related cardiotoxicity and five cardiac evaluations were unavailable or were performed by another technique. Finally, 135 patients were included in this study (Table 1). Treatments are summarised in Table 2.

Risk factors The cumulative dose of doxorubicin per square metre of body surface area was calculated for each patient. If the administration of doxorubicin was continued despite the occurrence of cardiac abnormality during the treatment, the cumulative dose at the time of the cardiac event was used in the assessment. The effects of other factors such as sex, age and pre-existing cardiac risk factors (hypertension, diabetes and hypercholesterolemia) were evaluated [2, 3].

Statistical analysis Univariate analyses were carried out using the chi-square or Fisher’s exact tests. At the outset, the cumulative dose of doxorubicin as well as the patient’s age were broken down into three and six groups (per 100 mg/m2 and 10 years), respectively. Then according to cut-off points, the number of groups could be reduced. Other parameters were considered as dichotomic qualitative variables. All variables with P <0.15 associated by univariate analysis were entered in a stepwise regression model. Interactions between candidate variables were tested. All analyses were performed using BMDP for Windows (release 7.0, BMDP Inc., Los Angeles, USA). Significance levels were set at 0.05.

Results

Cardiotoxicity Twenty-seven (20%) patients suffered from a cardiac event within 1 year of treatment (Table 3). Among these, 14 patients had clinical signs of CHF attributed to doxorubicin therapy after the exclusion of other potential causes (sepsis or renal failure): dyspnea, pulmonary oedema, peripheral oedema or tachycardia. Three patients died from a cardiac event during the year of treatment (two from acute pulmonary oedema associated with signs of myocardial failure, one from terminal stage CHF). Eleven patients received an appropriate cardiologic therapy such as digitalis, angiotensin-converting enzyme inhibitors or diuretics. In the 25 patients who had a decreasing LVEF, the final values and the decline of LVEF from baseline were (median and extreme values) 47.2% (62% and 20%) and –20.6% (–9% and –55%), respectively. In six cases, chemotherapy by the CHOP regimen was interrupted because of cardiac toxicity. In five additional patients with a decreasing LVEF, a cardioprotective treatment by dexrazoxane was administered and additive cycles of chemotherapy could be given.

Patients

Risk factors

At the outset, 151 patients treated with the CHOP regimen as first-line therapy were selected. Among these, 16 were excluded:

Among the patients who received involved field radiotherapy following CHOP chemotherapy at the standard dose of 40 Gy

279 Table 2. Description of CHOP therapy and disease management Patient categories

No. of patients (%)

6–8 cycles of CHOP regimen

64 (47.4)

3–4 cycles of CHOP followed by high-dose chemotherapy and hematopoietic stem cell transplantation

15 (11.1)

3–4 cycles of CHOP + involved field radiotherapy (localised stage disease)

22 (16.3)

Mediastinal radiotherapy (40 Gy)

1 (0.7)

Interruption of CHOP therapy (<6 cycles) followed by salvage therapy (refractory disease or poor response)

27 (20)

High-dose chemotherapy

11 (8.1)

Interruption of CHOP therapy (<6 cycles) related to toxic effects

7 (5.2)

Cardiotoxicity

6 (4.4)

(20 fractions), only one was administered on the mediastinum (Table 2). In a previous analysis, thresholds of toxicity were clearly identified for a cumulative dose of doxorubicin and age (Table 4). Univariate analysis showed that a cumulative dose of doxorubicin >200 mg/m2 and age >50 years were significant factors for early cardiotoxicity. No effect of sex and pre-existing cardiac risk factors was found. With multivariate analysis, the cumulative dose of doxorubicin and age remained significant risk factors of cardiotoxicity (Table 5). There was no interaction between these two factors.

clinical manifestations [16, 17]. Furthermore, GBP imaging or echocardiography are relatively insensitive when used to detect early myocardial damage [2]. The evaluation of diastolic function would probably be more sensitive [19]. The potential interest of cardiac serum troponin-T levels has been suggested, but this parameter has not yet been validated [20]. Our analysis showed that 20% of NHL patients treated by the CHOP regimen developed cardiac abnormalities within 1 year of treatment. In adult patients, available data are restricted to

Table 4. Univariate analysis

Discussion

Factors

Doxorubicin-based chemotherapy remains the gold standard first-line therapy for aggressive NHL, but cardiac effects have not been studied in these patients [14]. Our cohort was based on consecutive and unselected patients, who were representative of daily practice. All patients had not received chemotherapy and radiotherapy prior to the onset of treatment. The chemotherapy regimen and the schedule of administration of doxorubicin did not vary during the study. Thus, these confusion factors could be controlled [1, 9, 18]. In several studies, toxicity has been defined as CHF [1, 6, 8]. The use of more sensitive investigations has led to a wider definition of cardiotoxicity, including subclinical cardiomyopathy [2, 3]. As LVE dysfunction could be asymptomatic, the real clinical impact could be considered to be overestimated. However, it has been shown that a decline in LVEF values was predictive of late

Table 3. Clinical or subclinical cardiotoxicity within 1 year of treatment No. of patients (%) No cardiotoxicity Cardiac events Clinical signs of CHF without LVE dysfunction

108 (80) 27 (20)

No. of patients

No. of patients with cardiac event (%)

P value

Cumulative dose of doxorubicin (mg/m2) ≤200

50

4 (8)

>200–≤300

26

7 (26.9)

>300

59

16 (27.1)

≤200

50

>200

85

4 (8)

0.02

0.004

23 (27.1)

Age at treatment (years) ≤30

10

1 (10)

31–40

14

1 (7.1)

41–50

24

3 (12.5)

51–60

23

6 (26.1)

61–70

34

8 (23.5)

>70

30

8 (26.7)

≤50

48

5 (10.4)

>50

87

22 (25.3)

0.4

0.03

Pre-existing cardiac risks No

69

15 (21.7)

Yes

66

12 (18.2)

Male

77

15 (19.5)

Female

58

12 (20.7)

0.60

2 (1.5)

Clinical signs of CHF associated with LVE dysfunction

12 (8.9)

Asymptomatic LVE dysfunction

13 (9.6)

Sex 0.86

280 Table 5. Multivariate analysis Risk factors

Odds ratio

95% CI

P value

Cumulative dose of doxorubicin >200 mg/m2

4.2

1.3–13.5

0.005

Age >50 years

2.9

1–8.5

0.03

CI, confidence interval.

metastatic breast cancer patients. Early cardiotoxicity has been estimated as between 21% and 31%, but many women had received prior chemotherapy and/or radiotherapy [12, 13]. Among the 27 cases of cardiotoxicity which are reported in our study, 14 (52%) had clinical symptoms of CHF within 1 year of treatment. These results are in agreement with those of two recent population-based studies [21, 22], which showed that among the patients who had LVE dysfunction, ∼50% had clinical symptoms of CHF. In our study, three (21%) of the symptomatic patients died from a cardiac event within 1 year. In Von Hoff et al.’s study [1], the mortality attributed to anthracycline-induced CHF was estimated as ∼40%. In more recent studies, this rate has varied from 8% to 23% [5, 6, 8]. All these reports, including our study, were based on retrospective data, and mortality related to doxorubicin cardiotoxicity could have been influenced by other confounding factors. However, 1230 cases of cardiomyopathy were recently analysed, with an overall mortality rate of 34% [7]. The doxorubicin-related form appeared to have especially poor longterm prognosis compared to idiopathic cardiomyopathies (relative risk = 3.4, P <0.001). The cumulative dose of doxorubicin is the major factor of cardiotoxicity [1, 5]. An empirical threshold of 500 mg/m2 is commonly used and the risk appeared to be limited below 300 mg/m2 [3, 6, 13, 23]. With the CHOP regimen, our study identified a lower threshold for cumulative doses. Over 200 mg/m2, 27% of patients had subclinical or clinical cardiomyopathy. If these results are confirmed, the hypothesis can be made that the cardiotoxicity of doxorubicin could be increased by the presence of associated drugs. Cyclophosphamide is commonly associated with doxorubicin in childhood and breast cancers. Lipschultz et al. [5] suggested that cardiac effects could be increased by high doses of corticosteroids. Additionally, the potential role of vincristin should be investigated. Another finding of our study was that patients over 50 years of age had a significantly increased risk of cardiotoxicity. The specific effect of age has been demonstrated in children, but the results cannot be transposed to adults [1, 5, 9]. CHOP is now considered to be the standard regimen in all patients with aggressive NHL, including elderly patients [24, 25]. Using the equation of the logistic regression model, the probability of developing a cardiac event in patients over 50 years of age and receiving more than 200 mg/m2 of doxorubicin could be estimated at 33% within 1 year of treatment. These patients should be considered as a high-risk population. A limiting factor of our study was the short follow-up period. Several studies have reported late clinical and subclinical cardiomyopathy occurring in previously asymptomatic children [6, 9–11].

In NHL patients, late LVE dysfunction has been identified in 14% of long-term survivors, but the study included both children and adults and the chemotherapy regimens varied [15]. Nevertheless, early identification of a population at risk is important for the short-term management and the long-term follow-up of patients receiving doxorubicin. Furthermore, in two recent studies [6, 8], 76% and 89%, respectively, of all cases of CHF have been reported within 1 year of treatment. As early cardiotoxicity would appear to be a severe problem in aggressive NHL patients, preventive strategies need to be considered. In adults, it has been suggested that a prolonged infusion over 48 or 96 h may be less toxic than a short infusion [18]. However, this finding has not been confirmed and doubt concerning the preservation of antitumoral activity remains [2]. The use of epirubicin in aggressive NHL has not been conclusive and its advantage over doxorubicin on an equimolar basis is unclear [3, 26]. Although randomised trials have shown that dexrazoxane provides an effective cardioprotection in breast cancer, their design has been questioned and the role of this agent remains unclear [12, 27]. There are no available data for NHL patients and the recent guidelines of the American Society of Clinical Oncology are highly cautious concerning the use of dexrazoxane for tumours other than breast cancer [23]. The use of liposomeencapsulated doxorubicin could be considered to be a more promising approach [13]. Despite the usual limitations of retrospective analyses, our pragmatic study showed that early cardiac abnormalities and symptomatic forms are frequent in patients receiving the CHOP regimen. In addition, hematopoietic stem cell transplantation plays a major and increasing role in the treatment of these patients [28, 29]. The cardiotoxicity of HDC is probably dependent on regimen, but a history of declining LVEF values would appear to be a predictive factor for cardiac toxicity in NHL patients undergoing HDC [30]. At this point in time, the role of anthracyclines in first-line therapy remains unquestionable. Thus, the development of cardioprotective strategies or alternative treatments are mandatory for aggressive NHL patients.

References 1. Von Hoff D, Layard M, Basa P et al. Risk factors for doxorubicininduced congestive heart failure. Ann Intern Med 1979; 91: 710–717. 2. Shan K, Lincoff M, Young J. Anthracycline-induced cardiotoxicity. Ann Intern Med 1996; 125: 47–58. 3. Singal P, Iliskovic N. Current concepts: doxorubicin-induced cardiomyopathy. N Engl J Med 1998; 339: 900–905. 4. Green M. Anthracycline cardiotoxicity, no longer an issue? Ann Oncol 1998; 9: 691–693.

281 5. Lipschultz S, Lipsitz S, Mone S et al. Female sex and higher drug dose as risk factors for late cardiotoxic effects of doxorubicin therapy for childhood cancer. New Engl J Med 1995; 332: 1738–1743. 6. Kremer L, Van Dalen E, Offringa M et al. Anthracycline-induced clinical heart failure in a cohort of 607 children: long term follow-up study. J Clin Oncol 2001; 19: 191–196. 7. Felker M, Thompson R, Hare J et al. Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med 2000; 342: 1077–1084. 8. Krisher J, Epstein S, Cuthbertson D et al. Clinical cardiotoxicity following anthracycline treatment for childhood cancer: the Pediatric Oncology Group experience. J Clin Oncol 1997; 15: 1544–1552. 9. Lipschultz S, Colan S, Gelber R et al. Late cardiac effects of doxorubicin therapy for acute lymphoblastic leukemia in childhood. N Engl J Med 1991; 324: 808–815. 10. Steinherz L, Steinherz P, Tan C et al. Cardiac toxicity 4 to 20 years after completing anthracycline therapy. JAMA 1991; 266: 1672–1677. 11. Sorensen K, Levitt G, Bull C et al. Anthracycline dose in childhood acute lymphoblastic leukemia: issues of early survival versus late cardiotoxicity. J Clin Oncol 1997; 15: 61–68. 12. Swain S, Whaley F, Gerber M et al. Cardioprotection with dexrazoxane for doxorubicin-containing therapy in advanced breast cancer. J Clin Oncol 1997; 15: 1318–1332. 13. Batist G, Ramakrishnan G, Sekhar Rao C et al. Reduced cardiotoxicity and preserved antitumor efficacy of liposome-encapsulated doxorubicin and cyclophosphamide compared with conventional doxorubicin and cyclophosphamide in a randomized multicenter trial of metastatic breast cancer. J Clin Oncol 2001; 19: 1444–1454. 14. Messori A, Vaiani M, Trippoli S. Survival in patients with intermediate or high-grade non-Hodgkin’s lymphoma: meta-analysis of randomized studies comparing third generation regimens with CHOP. Br J Cancer 2001; 84: 303–307. 15. Haddy T, Adde M, McCalla J et al. Late effects in long-term survivors of high-grade non-Hodgkin’s lymphomas. J Clin Oncol 1998; 16: 2070– 2079. 16. Alexander J, Dainiak N, Berger H et al. Serial assessment of doxorubicin cardiotoxicity with quantitative radionuclide angiocardiography. N Engl J Med 1979; 300: 278–283. 17. Schwartz R, McKenzie W, Alexander J et al. Congestive heart failure and left ventricular dysfunction complicating doxorubicin therapy. Seven-year experience using serial radionuclide angiocardiography. Am J Med 1987; 82: 1109–1118.

18. Legha S, Benjamin R, Mackay B et al. Reduction of doxorubicin cardiotoxicity by prolonged continuous intravenous infusion. Ann Intern Med 1982; 96: 133–139. 19. Tjeerdsma G, Meinardi M, Van der Graf W et al. Early detection of anthracycline induced cardiotoxicity in asymptomatic patients with normal left ventricular systolic function: autonomic versus echocardiographic variables. Heart 1999; 81: 419–423. 20. Herman E, Zhang J, Lipschultz E et al. Correlation between serum levels of cardiac troponin-T and the severity of the chronic cardiomyopathy induced by doxorubicin. J Clin Oncol 1999; 17: 2237–2243. 21. McDonagh T, Morrison C, Lawrence A et al. Symptomatic and asymptomatic left-ventricular systolic dysfunction in an urban population. Lancet 1997; 350: 829–833. 22. Davies M, Hobbs F, Davis R et al. Prevalence of left-ventricular systolic dysfunction and heart failure in the Echocardiographic Heart of England Screening Study: a population based study. Lancet 2001; 358: 439–444. 23. Hensley M, Schuchter L, Lindley C et al. American Society of Clinical Oncology clinical practice guidelines for the use of chemotherapy and radiotherapy protectants. J Clin Oncol 1999; 17: 3333–3355. 24. Tirelli U, Errante D, Van Glabbeke M et al. CHOP is the standard regimen in patients ≥70 years of age with intermediate grade or high grade non-Hodgkin’s lymphoma: results of a randomized study of the European Organization for Research and Treatment of Cancer Lymphoma Cooperative Study Group. J Clin Oncol 1998; 16: 27–34. 25. Coiffier B, Lepage E, Brière J et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. New Engl J Med 2002; 346: 235–242. 26. Nair R, Ramakrishnan G, Nair N et al. A randomized comparison of the efficacity and toxicity of epirubicin and doxorubicin in the treatment of patients with non-Hodgkin’s lymphoma. Cancer 1998; 82: 2282–2288. 27. Phillips K, Tannock I. Design and interpretation of clinical trials that evaluate agents that may offer protection from the toxic effects of cancer chemotherapy. J Clin Oncol 1998; 16: 3179–3190. 28. Shipp M, Abeloff M, Antman K et al. International consensus conference on high dose therapy with hematopoietic stem cell transplantation in aggressive non-Hodgkin’s lymphoma: report of the jury. J Clin Oncol 1999; 17: 423–429. 29. Haioun C, Lepage E, Gisselbrecht C et al. Survival benefit of high-dose therapy in poor-risk aggressive non-Hodgkin’s lymphoma: final analysis of the prospective LNH-87 protocol—a Groupe d’Etude des Lymphomes de l’Adulte study. J Clin Oncol 2000; 18: 3025–3030. 30. Fujimaki K, Maruta A, Yoshida M et al. Severe toxicity in hematological stem cell transplantation: predictive value of reduced left ventricular ejection fraction. Bone Marrow Transplant 2001; 27: 307–310.