You are here
Age influences initial dose and compliance to imatinib in chronic myeloid leukemia elderly patients but concomitant comorbidities appear to influence overall and event-free survival
- We applied Charlson comorbidity index on CML cohort of very elderly patients.
- Comorbidities appeared to have an impact on OS, EFS and on non-CML death rate.
- Compliance was reduced in patients initially treated with reduced dose.
- For older patients, physician perception could influence initial dose.
We applied Charlson comorbidity index (CCI) stratification on a large cohort of chronic myeloid leukemia (CML) very elderly patients (>75 years) treated with imatinib, in order to observe the impact of concomitant diseases on both compliance and outcome. One hundred and eighty-one patients were recruited by 21 Italian centers. There were 95 males and 86 females, median age 78.6 years (range 75–93.6). According to Sokal score, 106 patients were classified as intermediate risk and 55 as high risk (not available in 20 patients). According to CCI stratification, 71 patients had score 0 and 110 a score ≥ 1. Imatinib standard dose was reduced at start of therapy (200–300 mg/day) in 68 patients independently from the evaluation of baseline comorbidities, but based only on physician judgement: 43.6% of these patients had score 0 compared to 34% of patients who had score ≥ 1. Significant differences were found in terms of subsequent dose reduction (39% of patients with score 0 compared to 53% of patients with score ≥ 1) and in terms of drug discontinuation due to toxicity (35% of patients with score 0 vs 65% of patients with score ≥ 1). We did not find significant differences as regards occurrence of hematologic side effects, probably as a consequence of the initial dose reduction: 39% of patients with score 0 experienced grade 3/4 hematologic toxicity (most commonly anemia) compared to 42% of patients with score ≥ 1. Independently from the initial dose, comorbidities again did not have an impact on development of grade 3/4 non-hematologic side effects (most commonly skin rash, muscle cramps and fluid retention): 62% of patients with score 0 compared to 52.5% of patients with score ≥ 1. Notwithstanding the reduced dose and the weight of comorbidities we did not find significant differences but only a trend in terms of efficacy: 66% of patients with score 0 achieved a CCyR compared to 54% of patients with score ≥ 1. Comorbidities appeared to have an impact on median OS (40.8 months for patients with score 0 vs 20.16 months for patients with score ≥ 1) on EFS and on non-CML death rate. Our results suggest that treatment of very elderly CML patients might be influenced by personal physician perception: evaluation at baseline of comorbidities according to CCI should improve initial decision-making in this subset of patients.
Keywords: Chronic myeloid leukemia, Age, Comorbidities, Overall survival.
Comorbidity, as defined by Feinstein  , is any distinct additional clinical entity that pre-exists or occurs during the clinical course of a patient with a primary disease.
Comorbidities may impact on survival and choice of treatment among cancer patients: in fact they have been identified as significant determinants of response to therapy in elderly with acute myeloid leukemia or female breast cancer or head and neck or lung cancer  . We previously evaluated the impact of comorbidities on a large series of elderly chronic myeloid leukemia (CML) patients treated with dasatinib after imatinib failure, with the aim to establish associations between comorbidities and development of pleural effusions or compliance to the drug  . For this purpose, we applied the Charlson comorbidity index (CCI) as well as adult-comorbidity evaluation-27 (ACE-27) prognostic scores: we found that the rate of drug reduction or suspension and the incidence of pleural effusions appear to be associated with the presence of comorbidities. More prone to investigate the impact of comorbidities in elderly oncologic population, we applied CCI stratification on a large cohort of CML very elderly patients (>75 years) treated with imatinib. Aim of present study was to assess the impact of concomitant diseases on both compliance and outcome and evaluate the influence of initial patient age on clinician's decision making.
2. Patients and methods
Present retrospective analysis included 181 CML chronic phase (CP) patients aged > 75 years treated with imatinib in 21 Italian Centers. Participating centers were asked to send data on all consecutive CP-CML patients who fulfilled the following inclusion criteria, irrespective of response or median time on treatment, after appropriate informed consent: (1) age > 75 years; (2) being in CP at the time of start of imatinib. No criteria of exclusion were considered. All patients were followed and monitored according to European LeukemiaNet recommendations  . For the analysis, the following parameters were considered: age, sex, Sokal risk, starting dose of imatinib, all grade side effects, grade of toxicity, reasons for discontinuation, number of days of discontinuation. All side effects were graded according to WHO classification. Comorbidities were evaluated at baseline, before imatinib treatment by medical staff. All the information about concomitant diseases and drugs used were reported on case history and then used for this retrospective evaluation. As reported, CCI is a list of 19 comorbid conditions: each condition has a weight assigned from 1 to 6, which was derived from relative risk estimates of a proportional hazard regression model using clinical data  and .
Statistical analysis was carried out using the SPSS software package. In particular, responses and frequency of toxicities were compared using X2 or Fisher's exact test. With a Cochran–Armitage test of trend we assessed the presence of an association between variables with several categories. p Values were considered significant when were < 0.05. Overall survival (OS) was estimated by the Kaplan Meier method and compared by the log–rank test. Events were defined as: treatment failure or permanent discontinuation of imatinib for any reason (including toxicity, patient refusal, or loss to follow-up).
Ethical Committee approval was obtained.
Table 1 shows clinical features of patients enrolled in the study. There were 95 males and 86 females, median age 78.6 years (range 75–93.6). According to Sokal score, 106 patients were classified as intermediate risk and 55 as high risk (not available in 20 patients). One hundred and ten patients presented 1 or more comorbidities at baseline. One hundred and forty-three patients received imatinib as first line whereas 38 patients received the drug as second line. According to CCI stratification, 71 patients had score 0, 50 patients had a score 1, 37 patients had score 2 and 23 patients had score ≥ 3. Imatinib standard dose was reduced at start of treatment in 68 patients independently from the evaluation of baseline comorbidities but based only on physician judgement. We did not identify statistical significant differences in terms of cytogenetic efficacy in patients stratified according to CCI, nor in terms of achievement of molecular responses ( Table 2 ), but an evident trend was shown for a reduction of response rates in patients with CCI score ≥ 1 (54% CCyR in patients with score ≥ 1 compared to 66% in patients with score 0, p = 0.06) and for MMR (24.5% in patients with score ≥ 1 compared to 32% for patients with score 0). We also stratified patients according to physician perception to start with a standard or reduced dose, not based on CCI score: reduced rate of molecular response was observed in patients with reduced dose but a CCI score of 0. As shown in Table 3 , initial dose was reduced (200–300 mg/day) in 31 or 43.6% of patients with score 0 compared to 37% or 34% of patients with score ≥ 1. Significant differences were however found between patients without or with comorbidities in terms of subsequent dose reduction when therapy was started at reduced dose (no patients with score 0 compared to 35% of patients with score ≥ 1). No significant differences in drug reduction were revealed in patients initially treated with standard dose. Discontinuations due to toxicity occurred in 44% of patients with score 0 and in 60% of patients with score ≥ 1, independently from the initial dose. Higher rate of discontinuations due to toxicity was revealed in patients started at a reduced dose (35% in patients with score 0 compared to 65% in patients with score ≥ 1, p = 0.04) compared to patients treated with full dose (50% in patients with score 0 vs 57% in patients with score ≥ 1, p = 0.43).
|Age (years)||78.6 (75–93.6)|
|Phase of disease|
|Dose of imatinib at start|
|CCI score||No. pts||CCyR (%)||p||MMR (%)||p||CMR (%)||p|
|0||71||47 (66%)||0.06||23 (32%)||0.05||11 (15%)||0.6|
|≥1||110||60 (54%)||27 (24.5%)||13 (12%)|
|0 (reduced dose)||31/71||18 (58%)||0.07||7 (22%)||0.05||3 (9.6%)||0.04|
|0 (standard dose)||40/71||29 (62%)||16 (40%)||8 (20%)|
|≥1 (reduced dose)||37/110||20 (54%)||0.12||8 (22%)||0.23||4 (11%)||0.54|
|≥1 (standard dose)||73/110||40 (58%)||19 (26%)||9 (12%)|
|CCI score||No. pts||Initial dose reduction by physician choice No. (%)||Reduction during treatment in patients started at reduced dose No. (%)||p||Reduction during treatment in patients started at standard dose No. (%)||p||Discontinuation during treatment in patients started at reduced dose No. (%)||p||Discontinuation during treatment in patients started at standard dose No. (%)||p|
|0||71||31 (43.6%)||–||0.001||28 (39%)||0.07||11/31 (35%)||0.04||20/40 (50%)||0.43|
|≥1||110||37 (34%)||13/37 (35%)||39/73 (53%)||24/37 (65%)||42/73 (57%)|
According to CCI stratification, we detected a statistically significant difference in terms of OS (from 40.8 months from patients with score 0 to 20.1 months from patients with score ≥ 1, p = 0.02), in terms of EFS (89% for patients with score 0 to 58% for patients with score ≥ 1, p = 0.04). Comorbidities influenced death rate: from 18% in patients with score 0 to 33% in patients with score ≥ 1, even when we considered physician perception and patients who started at reduced dose (p = 0.01, Table 4 ).
|CCI score||No. pts||Median OS (months)||p||EFS||p||Death rate||p|
|0 (reduced dose)||31/71||38.2||0.18||70%||0.02||22%||0.07|
|0 (standard dose)||40/71||42.6||91%||15%|
|≥1 (reduced dose)||37/110||18.2||0.08||48%||0.04||30%||0.06|
|≥1 (standard dose)||73/110||23.6||68%||36%|
We did not find significant differences as regards occurrence of hematologic side effects, probably due to the initial reduction of the dose: grade 3/4 hematologic toxicity (most commonly anemia) was experienced by 39% of patients with score 0 compared to 42% of patients with score ≥ 1. Independently from the initial dose, comorbidities again did not show an impact on development of grade 3/4 non-hematologic side effects (most commonly skin rash, muscle cramps and fluid retention): these occurred in 62% of patients with score 0 compared to 52.5% of patients with score ≥ 1 ( Table 5 ).
|CCI score||No. pts||Hematological toxicities||Non-hematological toxicities|
CML incidence rate varies from 0.6 to 2 cases per 100,000 inhabitants/year and increases with age, with a male prevalence  . A recent German study  , aimed to determine population-based age and gender-specific incidence of CML, reported that the median age in CML patients was 60.3 years, with male/female ratio of 1.66. The crude incidence for CML was 0.79, whereas age-specific incidence was 0.57 for patients aged less than 65 years, and 1.91 for patients aged > 65 years. The impact of age as a poor prognostic factor in CML has been well demonstrated  . In the interferon era, elderly patients diagnosed as having chronic phase chronic myeloid leukemia (CP-CML) had shorter survival compared to younger patients , , and . With the advent of target therapy with imatinib the outcome of elderly CML-CP patients has been extensively investigated. Several reports described improved responses in elderly patients treated with imatinib after IFN failure, with similar overall survival compared to younger population  and . In newly diagnosed older patients imatinib showed similar rate of cytogenetic and molecular responses compared to younger patients , , and . Although significant data were published, the choice to treat elderly patients, outside clinical trials, is still influenced by clinical perception of ability by the patient to tolerate therapy. We observed here that imatinib standard dose was reduced at start in 68 out of 181 elderly patients independently from the evaluation of baseline comorbidities but only especially based on physician judgement: 43.6% of patients with score 0 were started at a reduced dose (200–300 mg/day) compared to 33.6% of patients with score ≥ 1. Comparing CCI stratification and physician perception of ability to tolerate the drug (independently from comorbidities evaluation) we found a reduced rate of molecular responses obtained in patients without comorbidities but who started improperly with a reduced dose. Concomitant diseases maintained a role for a higher death rate independently from the dose used based on physician judgement. Comorbidities appeared to have an impact on EFS (89% in patients with score 0 vs 58% in patients with score ≥ 1) and influenced median OS (40.8 months in patients with score 0 vs 10.6 months in patients with score ≥ 1), but not efficacy. We found that compliance was reduced in patients initially treated with reduced dose. We recently showed how comorbidities might help in treatment decision making in a cohort of elderly patients treated with dasatinib  : application at baseline of CCI or ACE-27 identifies patients at high risk of developing pleural effusions and of low compliance to the drug. The results of this study demonstrates how should be reasonable to define an elderly patient according to reproducible tools of fragility (such as comorbidity indexes) rather than simply according to physician's perception. We showed here as treatment of very elderly CML patients might be influenced by personal physician perception: evaluation at baseline of comorbidities according to CCI should improve initial decision-making in this subset of patients.
Conflict of interest
All authors have no conflict of interest to report.
None. No funding to declare. MB followed patients, analysed data and wrote the paper; LL, RL, FC, DF, FC, MM, MA, FS, MT, GB, EC, PM, AG, LC, EM, AI, SR, MC, ARR, PP, ME, AS, MM, GG, FC, FS, SS, AD, GRC, SS, EA, GS, GR, followed patients and GA critically revised the paper.
-  A.R. Feinstein. The pre-therapeutic classification of co-morbidity in chronic disease. J Cron Dis. 1970;23:455-468 Crossref
-  O. Birim, A.P. Kappetein, A.J.J.C. Bogers. Charlson comorbidity index as a predictor of long-term outcome after surgery for non-small cell lung cancer. Eur J Cardiothorac Surg. 2005;28(5):759-762 Crossref
-  M. Breccia, R. Latagliata, F. Stagno, L. Luciano, A. Gozzini, F. Castagnetti, et al. Charlson comorbidity index and adult comorbidity evaluation-27 scores might predict treatment compliance and development of pleural effusions in elderly patients with chronic myeloid leukemia treated with second-line dasatinib. Haematologica. 2011;96:1457-1461 Crossref
-  M. Baccarani, G. Saglio, J.M. Goldman, A. Hochhaus, B. Simonsson, F. Appelbaum, et al. Evolving concepts in the management of chronic myeloid leukemia: recommendations from an expert panel on behalf of the European LeukemiaNet. Blood. 2006;108(6):1809-1820 Crossref
-  M. Breccia, A.M. Frustaci, L. Cannella, C. Stefanizzi, R. Latagliata, C. Cartoni, et al. Comorbidities and FLT3-ITD abnormalities as independent prognostic indicators of survival in elderly acute myeloid leukemia patients. Hematol Oncol. 2009;27(3):148-153 Crossref
-  S.J. Harrison, P.R.E. Johnson, T.L. Holyoake. The Scotland Leukemia Registry audit of incidence, diagnosis and clinical management of new patients with chronic myeloid leukemia in 1999 and 2000. Scott Med J. 2004;49:87-90
-  U. Berger, O. Maywald, M. Pfirmann, et al. Gender aspects in chronic myeloid leukemia: long-term results from randomized studies. Leukemia. 2005;19:984-989 Crossref
-  H. Kantarjian, M. Keating, K. McCredie, et al. Old age is a sign of poor prognosis in patients with chronic myelogenous leukemia. Southern Med J. 1988;80:1233-1235
-  J. Cortes, H. Kantarjian, S. O’Brien, et al. Result of interferon-alpha therapy in patients with chronic myelogenous leukemia 60 years of age and older. Am J Med. 1996;100:452-455 Crossref
-  W. Hilbe, U. Apfelbeck, M. Fridrik, et al. Interferon-alpha for the treatment of elderly patients with chronic myeloid leukemia. Leuk Res. 1998;22:881-886 Crossref
-  U. Berger, G. Engelich, O. Maywald, et al. Chronic myeloid leukemia in the elderly: long-term results from randomized trials with interferon alpha. Leukemia. 2003;17:1820-1826 Crossref
-  J. Cortes, M. Talpaz, S. O’Brien, et al. Effects of age on prognosis with imatinib mesylate therapy for patients with Philadelphia chromosome-positive chronic myelogenous leukemia. Cancer. 2003;98:1105-1113 Crossref
-  G. Rosti, I. Iacobucci, S. Bassi, et al. Impact off age on the outcome off patients with chronic myeloid leukemia in late chronic phase: results off a phase II study off the GIMEMA CML Working Party. Haematologica. 2007;92:101-105 Crossref
-  R. Latagliata, M. Breccia, I. Carmosino, et al. Real-life” results of front-line treatment with imatinib in older patients (>65 years) with newly diagnosed chronic myelogenous leukemia. Leuk Res. 2010;34:1472-1475 Crossref
-  F.M. Sanchez-Gujio, S. Duran, J. Galende, et al. Evaluation of tolerability and efficacy on imatinib mesylate in elderly patients with chronic phase CML: ELDERGLI study. Leuk Res. 2011;35:1184-1187
-  G. Gugliotta, F. Castagnetti, F. Palandri, et al. Frontline imatinib treatment off chronic myeloid leukemia: no impact off age on outcome, a survey by the GIMEMA CML Working Party. Blood. 2011;117:5591-5599 Crossref
a Dipartimento di Biotecnologie Cellulari ed Ematologia, Università “La Sapienza”, Via Benevento 6, Roma 00161, Italy
b Ematologia, Università di Torino, Torino, Italy
c Ematologia, Università di Ferrara, Ferrara, Italy
d Ematologia, Ospedale Sant’Eugenio, Roma, Italy
e Ematologia, Ospedale Cardarelli, Napoli, Italy
f Ematologia, Ospedale Ferrarotto, Catania, Italy
g Ematologia, Università di Udine, Udine, Italy
h Ematologia, Università di Padova, Padova, Italy
i Dipartimento Onco-Ematologico, IRCCS-CROB, Rionero in Vulture, Italy
j Ematologia, Università di Firenze, Firenze, Italy
k Ematologia, Università di Perugia, Perugia, Italy
l Ematologia, Ospedale Sant’Andrea, Roma, Italy
m UOC Oncoematologia, Fondazione IRCSS Cà Granda Policlinico Milano, Milan, Italy
n Ematologia, Università di Messina, Messina, Italy
o Ematologia, Ospedale S. Giovanni, Roma, Italy
p Ematologia con Trapianto, Università di Bari, Bari, Italy
q Ematologia, Ospedale S Giovanni Battista, Torino, Italy
r Ematologia, Centro Trapianti Cremona, Cremona, Italy
s Ematologia, IFO Regina Elena, Roma, Italy
t Ematologia, Ospedale Civile di Campobasso, Campobasso, Italy
u Ematologia, Ospedale Belcolle, Viterbo, Italy
v Ematologia, Università Cattolica del Sacro Cuore, Roma, Italy
w Onco-Ematologia, Università Cattolica Giovanni Paolo II, Campobasso, Italy
x Ematologia, Ospedale Villa S Pietro, Roma, Italy
y Ematologia, Polo Universitario ASO San Luigi Gonzaga, Orbassano, Italy
z Ematologia, Università “Federico II”, Napoli, Italy
aa Ematologia, Università di Bologna, Bologna, Italy
bb Ematologia, Pesaro, Italy
© 2014 Elsevier Ltd, All rights reserved.