Biopsychosocial vs. Biomedical Model

Cultural factors related to adherence to imatinib in CML: A Mexican perspective Olga Graciela Cantú-Rodríguez, Mónica Sánchez-Cárdenas, César Homero Gutiérrez-Aguirre, José Carlos Jaime-Pérez, Consuelo Mancias-Guerra, Oscar González-Llano, David Gómez-Almaguer

Hematology Service, Hospital Universitario, Universidad Autónoma de Nuevo León, Monterrey, Mexico

Introduction: The advent of imatinib as a therapeutic option of chronic myeloid leukemia (CML) has transformed this previously highly resistant disease into one that is susceptible to management with oral drugs that now offer high long-term survival rates. However, achieving an adequate adherence to treatment regimes is of critical importance. The characteristics of treatment compliance in Mexican patients have not been determined. Methods: We evaluated 38 CML patients, members of the Glivec® International Patient Assistance Program (GIPAP). A bimonthly simplified medication adherence questionnaire was applied and the adherence rate was calculated by direct tablet counting. Results: Two groups, one of local patients and another of out-of-town patients, were studied using an 85% adherence rate as a cut-off. The overall adherence rate was 85.9%. Fifteen patients were considered non- adherent (39.5%). The group of out-of-town patients presented a higher adherence rate of 92.8% in contrast with 76.3% in the local population (P= 0.021). The probability of achieving a complete cytogenetic response at some point of evolution after 8 years of follow-up was 93% in the adherent group vs. 58% in the group with an adherence rate <85% (P= 0.008). In patients with imatinib failure, the adherence rate was 75.8% compared to 95.5% (P= 0.008) in the optimal response group. Conclusions: In Mexican patients with CML, non-adherence to treatment is a cause of the failure to achieve remission or the subsequent loss of a complete cytogenetic and major molecular response.

Keywords: Treatment, Adherence, Imatinib

Introduction The introduction of imatinib as a therapeutic option for chronic myeloid leukemia (CML) completely revo- lutionized the natural history of the disease.1 From a progressive disease with a poor prognosis, CLM evolved into an illness that permits an adequate quality of life and an overall survival rate greater than 80% at 5 years.2 Imatinib presents the possibility of a simple, safe, oral, and highly effective outpatient treatment, but at considerable cost.3 Despite this and the fact that the disease is potentially fatal, adherence may be erratic. Although there is ample information available

related to various other pathologies, measuring adher- ence is still a challenge. The lack of standardization and the difficulty in eliminating bias contribute to ambiguity and complicate an accurate measurement of adherence.4

In Mexico, with diseases such as tuberculosis, diabetes mellitus, and hypertension, non-adherence to treatment regimes is a public health problem. In addition to the impact on the clinical course, poor adherence to treatment results in higher costs, increased hospitalization rates, and a deterioration in the doctor–patient relationship.5 In addition to this, one must consider the potential relationship between poor adherence to imatinib and the emergence of drug resistance.

In a previous study we observed a lower molecular response rate than expected to imatinib treatment, explained in part by the short follow-up period and the difficulties encountered in increasing the dose of imatinib, mainly because of side effects; however, adherence was not documented in this study.6

Mexico is a developing country with a population of 112 000 000 inhabitants, of which 37 million have an elementary school education or less. It is also impor- tant to point out that 35.6% of the population does not have health insurance.7 Because of these character- istics and the lack of data available, we considered it

Correspondence to: David Gómez-Almaguer, Servicio de Hematología del Hospital Universitario, Universidad Autónoma de Nuevo León, Madero y Gonzalitos s/n, Colonia Mitras Centro, Monterrey, Nuevo Leon CP 64460, Mexico. Email: [email protected]

© W. S. Maney & Son Ltd 2015 DOI 10.1179/1607845414Y.0000000165 Hematology 2015 VOL. 20 NO. 272

essential to assess adherence to imatinib as accurately as possible in a group of underprivileged and unin- sured patients in order to analyze the factors associ- ated with good or poor adherence and its impact on the clinical evolution of CML. A better understanding of the role of adherence will lead to the creation of pro- grams focused on strengthening adhesion to treatment and the integral management of the CML patient.

Methods Patients We evaluated 38 patients with a confirmed diagnosis of CML who were being treated with imatinib and beneficiaries of the Glivec® International Patient Assistance Program (GIPAP). Adherence to treatment was assessed at each medical visit between 1 August 2011 and 15 June 2013. The GIPAP program provided the necessary medication for treatment of CML based on imatinib. Each patient was instructed to keep the empty blisters and return them when he or she received a new box; the empty boxes returned were registered. Approval for the study was obtained from the Ethics Committee of the School of Medicine and the University Hospital of the Universidad Autónoma de Nuevo León. Patients gave informed consent in accordance with the Declaration of Helsinki.

Adherence assessment Adherence was determined for each patient using two methods that were different and complementary. During the follow-up period, the adherence rate was calculated by the direct counting of tablets (tablets consumed/tablets prescribed) based on GIPAP program guidelines and considering medication adjustments. A bimonthly, simplified medication adherence questionnaire (SMAQ) was also applied. This adherence test has been used and validated for different pathologies with characteristics similar to CML.8 Each patient was classified as adherent/non- adherent according to the guidelines established for the interpretation of the questionnaire. Regarding exclusion criteria, those who did not undergo at least two evaluations with the SMAQ adherence test were eliminated from analysis.

Response to treatment Before the start of this study, the degree of cytogenetic response achieved was registered. At the end of the follow-up period, cytogenetic and molecular response was determined for each patient in order to detect changes in disease status and the relationship of these changes with the adherence rate. Achieved response was categorized according to criteria estab- lished by the international guidelines of the European Network of Chronic Myelogenous Leukemia. Imatinib failure was defined as progression

from the chronic phase, loss of hematologic, cytoge- netic, or molecular response, and death from disease. Optimal response was defined as the presence of at least a stable molecular response.

Statistical analysis Groups were compared using the Chi square test/ Fisher’s exact test for categorical data and Student’s t-test/Mann–Whitney U-test for quantitative data. The probability of a complete cytogenetic response (CCyR) was calculated using the cumulative incidence procedure and the long rank test was used to compare the groups. A P value <0.05 was considered statisti- cally significant and all tests were two-sided. Statistical analysis was carried out using SPSS soft- ware version 20.0.

Results We evaluated 38 patients (19 men and 19 women) with a median age of 42 years (range 21–79) for the study. At diagnosis, 92.1% of patients were in the chronic phase of the disease and three patients were diagnosed in the accelerated phase (7.9%). After a median follow- up of 241 days (29–301), the mean adherence rate (MAR) was 85.9% (Table 1). Prior to this study, our cohort received imatinib for the management of CML for a median of 41.3 months (92–127). In patients who had undergone more than 24 months of treatment, the MAR was significantly lower (80.8%)

Table 1 Patient demographic and clinical characteristics

Adherence

(≥85%) (<85%)

Variable N= 23 (60.5%)

N= 15 (39.5%) P

Age, median (range) 44 (21–79) 37 (26–65) 0.3 Gender

F/M 12/11 7/8 0.7 Education (years). Mean

(SD) 8.18 (3.7) 7.75 (3.8) 0.7

Distance to hospital 0.09 Local 30% 60% Foreign 70% 40%

Diagnostic blood count. Mean (SD) HB (g/dl) 10.2 (2.2) 9.2 (1.9) 0.2 WBCs (103/μl) 192 (12) 332 (24) 0.05 PLTs (103/μl) 501 (46) 380 (37) 0.4

Adverse events record GI 30% 60% 0.09 Anemia 4% 0% 1.0 Leukopenia 9% 13% 1.0 Thrombocytopenia 35% 13% 0.25 Cutaneous 30.% 33% 1.0 Neurological 4.% 20% 0.28 General 61% 73% 0.50

Dose of imatinib 1.0 400 mg/day 70% 67% >400 mg/day 30% 33%

Time of prescription 0.001 <24 months 52% 0% >24 months 48% 100%

Cantú-Rodríguez et al. Treatment adherence on Mexican patients in CML

Hematology 2015 VOL. 20 NO. 2 73

compared to those patients with shorter prescriptions (96.9%) (P= 0.008). Regarding the actual dosage, the median was

400 mg (300–800 mg). During follow-up, seven patients (18.4%) required dose adjustment. In three, it was necessary to reduce the dose due to the presence of adverse effects; in four patients the dose was increased due to failure to respond (MAR 70.4%). Patients receiving 800 mg daily had the lowest adher- ence rate (MAR 65.9%). The group of patients that received ≤4 tablets/day achieved a greater adherence rate (MAR 87.6%) than patients taking 5–8 tablets/ day (MAR 82.2%), with no statistically significant difference (P= 0.6). To create contrasting groups, we established 80%, 85%, and 90% adherence rates as cut-off levels. Using Cox regression, 85% was estab- lished as the cut-off, derived from its predictive ability in terms of the degree of response achieved. Group 1 (39.5% of patients) had a MAR of 68.1%, in contrast with Group 2 (MAR 97.5%, N= 23), with no statistically significant differences in gender (P= 0.5), age (P= 0.3), or years of education (P= 0.07). Local patients had a MAR of 76.35% compared to out-of-town patients 92.85% (P= 0.021). Throughout their clinical course, 65.8% of patients

presented some adverse event associated with imatinib, with gastrointestinal symptoms representing the most common manifestations. The non-adherent group had higher rates of gastrointestinal events in relation to adherents (60% vs. 30.4%). Before the start of this study, the degree of response achieved was registered to detect which patient failed to show adequate response levels and the relationship of these response levels to drug adherence. At baseline, 31.6% of patients (n= 12) had failed to achieve any degree of response; of these patients, only three (25%) achieved complete molecular response (MAR 93.6%). During follow-up, four patients lost the degree of

response achieved (MAR 81.72%). Of 38 patients, one was in an accelerated phase and one died because of disease activity. Currently, 13 patients (34%) are non-responders

(imatinib failure), 21 patients (55%) have optimal response (stable MMR), and 4 patients (14%) were non-evaluable. In patients with imatinib failure, the MAR was 75.8%, in contrast with the group with optimal response (MAR 95.5%). This difference was statistically significant (P= 0.004). The impact of adhesion on achieving CCyR at some

point during the course of the disease was determined. In the adherent group (adherence≥ 85%) after 8 years of follow-up, there was a 93% probability of achieving a response (HR 3.1; CI 1.2–7.5, P= 0.01), in contrast with the group categorized as non-adherent with 58% (P= 0.008) (Fig. 1). An adherence rate of less than 95% was associated with an HR of 2.5 (CI 1.1–5.7,

P= 0.026), while values less than 80% were linked to an HR of 6.1 (CI 1.4–26.8, P= 0.016).

We applied 161 SMAQ as an adherence test (median 4; 2–7). A patient was considered non-adher- ent when at least one questionnaire item indicated that the drug had not been taken as prescribed. With regard to the adherence rate established by the SMAQ, the non-adherent group presented a MAR of 71.7%, in contrast to the adherent group with 96.2%, a statisti- cally significant difference (P< 0.001). We analyzed the correlation between the two methods (adherence rate obtained by direct tablet counting and the SMAQ) and obtained a kappa index of 0.728 (P≤ 0.001). This showed that there was good concordance between the two methods.

Discussion During this study, the MARwas 85.9% after a median of 3.4 years of treatment based on imatinib; we found lower rates of adherence related to longer use of the drug. This behavior mirrors the role of chronic drug use in pathologies such as hypertension or diabetes.9

The approximate cost of medication with a standard dose of 400 mg/day is 3800 US$ per month. Our insti- tution is part of the GIPAP program so that the patients received the drug for free. As a result, we were able to analyze the factors that could affect a lack of adherence after eliminating economic limit- ations. In contrast with previous evaluations where younger individuals displayed lower adherence rates, in our cohort characteristics such as age or gender were not related to adherence rate.10 We found that patients who had to travel from their hometowns to the medical center to receive the drug (distances of 100–600 km) had a significantly higher adherence rate than local patients (P= 0.02). This could be because out-of-town patients are afraid of being far

Figure 1 Cumulative probability of complete cytogenetic response according to level of adherence.

Cantú-Rodríguez et al. Treatment adherence on Mexican patients in CML

Hematology 2015 VOL. 20 NO. 274

from their doctors and therefore more motivated to carefully follow the instructions, on the other hand, Jonsson et al. determined a high level of adherence to imatinib associated with factors such as frequent contact with a single hematologist, involvement in decision making, and appropriate information about the disease.11 These variables were not assessed in our study. It is important to note that the educational level of our cohort (mean 8 years) probably had an important impact on our findings.12

Adherence to imatinib using an electronic monitor- ing system13 was evaluated in a group of British patients with CML. Adherence rates in this very differ- ent population (97.6%) were higher than those in our study; however, the authors recognize the inability to completely eliminate measurement bias. In this setting, measurement of imatinib blood levels may represent the most accurate method, but it is certainly difficult and expensive to implement.14 A weakness of our study is the short time period in which adherence was measured with relation to time since diagnosis, which could imply an inaccurate reflection of the characteristics of attachment present along the course of the disease. However, the characteristics of drug adherence tend to remain stable over time, allow- ing us to assume the presence of such adherence pat- terns for extended periods of time. The SMAQ has been widely used in the assessment of adherence to various drugs.8 It was initially used in the field of anti- retroviral drugs and it has demonstrated high levels of detection of patients with poor adherence. In our population, we were able to detect that 95% of patients had an adherence rate below 85%. We retrospectively evaluated the moment when our

patients achieved a CCyR or MMR. After 7 years of follow up, the likelihood of such a response was clearly influenced by the degree of adherence (P= 0.008). After 60 months of imatinib use, we found that our non-adherent population did not exceed a 60% probability of achieving that level of response in contrast to the adherent group, in which this prob- ability was higher than 95%. Despite population differences, other authors have

reported similar data, reinforcing the importance of adherence to achieve response goals.13,15 Ibrahim et al.13 determined the role of adherence in the loss of CCyR once it has been reached. In our population, patients with imatinib failure had significantly lower adherence rates (P= 0.008) compared to the group with optimal response, strengthening the role of adherence. The limitations of our study include bias in the

assessment of adherence and a relatively short follow-up. Ideally the population should be evaluated from the moment that the diagnosis is established, with periodic up-dates on the degree of response.

Currently imatinib constitutes the first line of treat- ment in the management of CML. A decade after its introduction we still do not know the impact of its chronic use and if it is realistic to talk of a cure. Additionally, the constant risk of drug resistance by new mutations is well known.16,17 If we consider the economic impact of tyrosine kinase inhibitors, it is essential to ensure that they are prescribed and administered optimally and with complete adherence.5

In the majority of Mexican CML patients imatinib is almost the only choice to obtain sustained remission and long-term disease-free survival, rendering adher- ence to this drug essential.

Disclaimer statements Contributors DG-Awas responsible for designing pro- tocol and writing the manuscript. OGCR was respon- sible for designing and writing the protocol and the manuscript.MSCwas responsible for patients’ attention, follow-up. CHG-A was responsible for study drug administration and data analysis. JCJP was responsible for writing the manuscript. OGL was responsible for data analysis and Table 1. CMG was responsible for data analysis and Fig. 1.

Funding None.

Conflicts of interest None.

Ethics approval Approval for the study was obtained from the Ethics Committee of the School of Medicine and the University Hospital of the Universidad Autónoma de Nuevo León. Patients gave informed consent in accordance with the Declaration of Helsinki.

References 1 Jabbour EJ, Kantarjian H, Eliasson L, Cornelison AM, Marin D. Patient adherence to tyrosine kinase inhibitor therapy in chronic myeloid leukemia. Am J Hematol. 2012;87:687–91

2 Goldman JM,Melo JV. Chronic myeloid leukemia – advances in biology and new approaches to treatment. N Engl J Med. 2003; 349:1451–64.

3 Richardson JL, Shelton DR, Krailo M, Levine AM. The effect of compliance with treatment on survival among patients with hematologic malignancies. J Clin Oncol. 1990;8:356–64.

4 Lee JK, Grace KA, Foster TG, Crawley MJ, Erowele GI, Sun HJ, et al.How should we measure medication adherence in clini- cal trials and practice? Ther Clin Risk Manag. 2007;3:685–90.

5 Darkow T, Henk HJ, Thomas SK, Feng W, Balandi JF, Goldberg GA, et al. Treatment interruptions and non-adherence with imatinib and associated healthcare costs: a retrospective analysis among managed care patients with chronic myelogenous leukaemia. Pharmacoeconomics 2007;25:481–96.

6 Gutierrez-Aguirre CH, García-Rodríguez F, Ortiz-Galvez V, Cantú-Rodríguez OG, Salazar-Riojas R, Martínez-González O, et al. Lower than expected cytogenetic and molecular response to imatinib in Mexican patients with chronic myelogenous leukemia. Hematology 2013;18:224–9.

7 INEGI. México en cifras. Available from: http://www3inegi orgmx/sistemas/mexicocifras/2013

8 Ventura-Cerda JM, Minguez-Gallego C, Fernandez-Villalba EM, Alós-Almiñana M, Andrés-Soler J. Simplified scale for

Cantú-Rodríguez et al. Treatment adherence on Mexican patients in CML

Hematology 2015 VOL. 20 NO. 2 75

medication adherence related problems in anti-retroviral therapy. Farm Hosp. 2006;30:171–6.

9 Yeaw J, Benner JS, Walt JG, Sians S, Smith DB. Comparing adherence and persistence across 6 chronic medication classes. J Manag Care Pharm. 2009;15:728–40.

10 Marin D, Bazeos A, Mahon FX, Eliasson L, Milojkovic D, Bua M, et al. Adherence is the critical factor for achieving molecular responses in patients with chronic myeloid leukemia who achieve complete cytogenetic responses on imatinib. J Clin Oncol. 2010; 28:2381–8.

11 Jonsson S, Olsson B, Soderberg J, Wadenvik H. Good adherence to imatinib therapy among patients with chronic myeloid leukemia – a single-center observational study. Ann Hematol. 2012;91:679–85.

12 Eliasson L, Clifford S, Barber N, Marin D. Exploring chronic myeloid leukemia patients’ reasons for not adhering to the oral anticancer drug imatinib as prescribed. Leuk Res. 2011;35:626–30.

13 Ibrahim AR, Eliasson L, Apperley JF, Milojkovic D, Bua M, Szydlo R, et al. Poor adherence is the main reason for loss of

CCyR and imatinib failure for chronic myeloid leukemia patients on long-term therapy. Blood 2011;117:3733–6.

14 Zhong JS, Meng FY, Xu D, Zhou HS, Dai M. Correlation between imatinib trough concentration and efficacy in Chinese chronic myelocytic leukemia patients. Acta Haematol. 2012; 127:221–7.

15 Ganesan P, Sagar TG, Dubashi B, Rajendranath R, Kannan K, Cyriac S, et al. Nonadherence to imatinib adversely affects event free survival in chronic phase chronic myeloid leukemia. Am J Hematol. 2011;86:471–4.

16 Noens L, van Lierde MA, De Bock R, Verhoef G, Zachée P, Berneman Z, et al. Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood 2009;113:5401–11.

17 Ruiz-Argüellez GJ, Tarin-Arzaga LC, Gonzalez-Carrillo ML, Gutierrez-Riveroll KI, Rangel-Malo R, Gutierrez-Aguirre CH, et al. Therapeutic choices in patients with Ph-positive CML living in Mexico in the tyrosine kinase inhibitor era: SCT or TKIs? Bone Marrow Transplant 2008;42:23–8.

Cantú-Rodríguez et al. Treatment adherence on Mexican patients in CML

Hematology 2015 VOL. 20 NO. 276

Copyright of Hematology is the property of Maney Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.