Resumen:
Objetivo: El objetivo es evaluar el grado de adhesión terapéutica a imatinib en pacientes con leucemia mieloide crónica y su relación con la respuesta terapéutica.
Material y métodos: Estudio realizado en octubre 2013-mar- zo 2014 que incluye a pacientes diagnosticados de leucemia mieloide crónica en tratamiento con imatinib en el hospital. Se evaluó la adhesión terapéutica mediante el cuestionario es- tandarizado Morisky-Green y el registro de dispensaciones de medicación. Se excluyeron aquellos que no completaron 6 me- ses de tratamiento y/o no realizaron el cuestionario. La res- puesta terapéutica se evaluó siguiendo guías clínicas. Se realizó el análisis descriptivo de variables y correlación mediante test Chi-cuadrado de Pearson.
Resultados: Se incluyeron 31 pacientes. Al evaluar el grado de asociación entre variables de respuesta y adhesión terapéutica: 1. La respuesta molecular mayor fue alcanzada por el 68,4% de los pacientes con alta adhesión y por el 75% de los pacien- tes con adhesión media. 2. La respuesta molecular completa fue alcanzada por el 57,9% de los pacientes con alta adhesión y por el 58,3% de los pacientes con adhesión media. No se observaron diferencias estadísticamente significativas en las variables de respuesta entre pacientes con adhesión terapéu- tica alta y media. No se observó asociación entre el grado de adhesión y la respuesta terapéutica.
Conclusiones: No podemos afirmar que el distinto grado de adhesión terapéutica influya en la respuesta al imatinib, aun que debería considerarse en casos de fallo terapéutico o respuesta subóptima.
Palabras clave:Leucemia mieloide crónicaLeucemia mieloide crónica,Adherencia terapéuticaAdherencia terapéutica,ImatinibImatinib.
Abstract:
Objective: To assess the level of adherence to treatment with imatinib in patients with chronic myeloid leukaemia and its association with therapeutic response.
Materials and methods: Study conducted on October, 2013 – March, 2014, including patients diagnosed with Chronic Myeloid Leukaemia on treatment with imatinib in the hospital. Therapeutic adherence was assessed through the standard Morisky-Green Questionnaire and the medication dispensing record. Those patients who did not complete 6 months of treatment and/or did not complete the questionnaire were ex- cluded. Therapeutic response was assessed following clinical guidelines. The descriptive analysis of variables and correlation was conducted through Pearsons’s ChiSquare Test.
Results: The study included 31 patients. When assessing the level of association between response variables and therapeu- tic adherence: 1. The highest molecular response was reached by 68.4% of those patients with high adherence, and by 75% of those patients with intermediate adherence. 2. Complete molecular response was achieved by 57.9% of patients with high adherence, and by 58.3% of patients with intermediate adherence. No statistically significant differences were found in response variables between patients with high and inter- mediate therapeutic adherence. No association was observed between level of adherence and therapeutic response.
Conclusions: We cannot confirm that a different level of thera- peutic adherence might have an impact on response to imatinib, though this should be taken into account in cases of therapeutic failure or sub-optimal response. Chronic Myeloid Leukaemia; Therapeutic adherence; Imatinib
Keywords: Chronic Myeloid Leukaemia, Therapeutic adherence, Imatinib.
ORIGINALES
Study on adherence to treatment with imatinib in chronic myeloid leukaemia and its association with therapeutic response
Estudio de adhesión a imatinib en la leucemia mieloide crónica y su relación con la respuesta terapéutica
Sociedad Española de Farmacia Hospitalaria
This work is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International.
Received: 02 March 2015
Accepted: 08 February 2016
Chronic Myeloid Leukaemia (CML), also known as Chronic Myelogenous or Granulocytic Leukaemia, is defined as a malignant clonal haematological neoplasia of the pluripotent stem cells1,2 included, according to the latest WHO Classification on 2008, within the group of Chronic Myeloproliferative Neoplasias (CMPN).
The introduction of imatinib, a tyrosine-kinase inhibitor (TKI), in the Spanish armamentarium on 2001 represented a dramatic change in CML treatment, allowing an expected survival of up to 88% at 5 years after diagnosis in patients treated in early stages of the disease. Therapy success depends on the adequate treatment compliance, and treatment discontinuations will generally lead to disease recurrence4.
Studies conducted in recent years have suggested that non-adherence to treatment with imatinib is frequent, and that it can also have an impact on the course of the disease5,6,7,8.
On a Belgian study with 169 patients, treatment adherence was assessed through the BAAS Questionnaire (Basel Assessment of Adherence Scale with Immunosu-ppressive Medication) adapted to imatinib, interviews with patients, the EVA score, and the count of surplus tablets compared with those prescribed. A third part of patients were considered non-adherent according to the BAAS Questionnaire adapted to imatinib. This question- naire consisted of a guided clinical interview including 4 questions, where one positive answer to any of the questions represented non-adherence. Only 14% of patients took adequately 100% of the imatinib doses they had been prescribed6.
In another study conducted in the United Kingdom with 87 patients, treatment adherence was assessed through the electronic record of the number of times when medication bottles were opened. Patients who took less than 90% of doses were classified as non-adherent to treatment, and reached a 26% rate of non-adherence5.
Subsequently, in a Swedish study with 38 patients, treatment adherence was assessed through an adaptation of the Morisky-Green test, and they only detected 1 non-adherent patient out of the 48 patients included in the study9.
The objective of this study was to assess the level of adherence to treatment with imatinib in patients diagnosed with CML within the healthcare area of the study, and its association with therapeutic response.
Patient data were obtained after the signature of In- formed Consents. Those protocols established by the hospital have been followed in order to access clinical record data. The information obtained has been treated in a confidential and anonymous manner, in accordance with the LOPD Act 15/99, dated December, 13th.
All patients diagnosed with CML on active treatment with imatinib at the time of study initiation were inclu- ded. The period of the study was from October, 2013 to March, 2014. Those patients who had not completed at least 6 months of treatment with imatinib at the time of the study were excluded, as well as those who did not complete the adherence questionnaire and/or did not sign the Consent Form.
Pharmacotherapeutical data were obtained from the Silicon® computer application, where the dispensing is recorded in the Hospital Pharmacy Unit, and medical data were obtained from the electronic clinical record through the IANUS® application, and from the hard-copy clinical records.
Therapeutic Adherence was measured using the me- dication dispensing record in the Hospital Pharmacy Unit collected in the Silicon® application during the whole pe- riod of the study. Medication Possession Ratio (MPR) was estimated according to the number of days of treatment dispensed divided in the number of days elapsed between consecutive appointments in the Pharmacy Unit. The mean number of dispensations per patient was 3.6, because the dispensing period included between 1 and 2 months. Those patients with a MPR between 0.8 and 1.2 were considered compliant with their therapy. Complementarily, the standard 8-question Adherence Ques- tionnaire validated by Morisky-Green10 was conducted for all patients. Interviews were conducted over the pho- ne, and there was only one interview per patient during the study period. Based on the score scale used by Morisky-Green, we established a score scale according to which those patients with a score over 2 were classified as Patients with Low Adherence; a score between 1 and 2 was classified as Intermediate Adherence, and a score of 0 determined High Adherence to Treatment.
For under-age patients, the questionnaire was answe- red by parents or guardians responsible for treatment administration.
Therapeutic response was evaluated based on national11 and international12-14 clinical guidelines for CML management, according to molecular response.
A descriptive analysis was conducted on the variables collected in the study. Qualitative variables were expressed as absolute value and percentage, while quantitative variables were described through their mean value and standard deviation, median and range. The association of therapeutic adherence and response variables (Major Molecular Response and Complete Molecular Response) was estimated with Pearson’s Chi-Square Test or Fisher’s Exact Test, as relevant. The potential correlation between the level of therapeutic adherence and response rate was studied with Spearman’s Correlation Coefficient. In the statistical tests used, values of p< 0.05 were consi- dered significant.
The analysis was conducted with the SPSS 19.0 Program for Windows.
Forty-three (43) patients diagnosed with CML were identified as being on treatment with imatinib at the time of the study. Three patients were excluded because they had not completed 6 months of treatment at the time of the study, and 9 patients were excluded because they did not answer the questionnaire of treatment adherence. Finally, 31 patients were included in the study.
The mean age of patients at the time of the study was 58 ± 18 years. Out of them, 51.6% were male and 48.4% were female.
A 67.6% of patients had received treatment with hydroxyurea before receiving imatinib; 35.5% had received interferon, and 3.2% had been treated with busulfan.
The mean time of treatment with imatinib up to the time of the study was 56.6 ± 44.4 months.
In terms of molecular response, the lowest value detected for mean BCR-ABL/ABL transcript ratio was 0.326 ± 1.79%, with a minimum of 0.00% (undetectable) and a maximum of 10.0% (in the case of 1 patient who refused to increase the dose above 200 mg). The median value was 0.00%; that is to say, undetectable. Thus, 61.3% of patients reached at least an undetecta- ble value of BCR-ABL/ABL transcript ratio at some point during their treatment with imatinib.
The proportion of patients who reached MMR (Major Molecular Response) during treatment with imatinib was 71.0%, and the mean time elapsed until achieving said level of response was 35.9 ± 34.0 months.
The proportion of patients who reached CMR (Com- plete Molecular Response) during treatment with imati- nib was 58.1%, and the mean time elapsed until achie- ving said stage was 41.1 ± 38.0 months (Table 1).
The adherence measurement conducted through Pharmacy Dispensing Records over the 6 months of the study classified all patients as having good treatment ad- herence, because the mean MPR was 1.
Demographical Data
When assessing treatment adherence through the Morisky-Green Questionnaire, 61.3% of patients presented high adherence, 38.7% showed intermediate adherence, and there were no patients with low adherence.
Thus, only one patient answered that he occasionally forgot to take his medication, and three patients ack- nowledged that they had forgotten to take some doses during the past two weeks. No patients acknowledged they had discontinued treatment without discussing it with their doctors; however, six patients stated they had been willing to discontinue treatment at some point. Six patients stated that sometimes they forgot to take their treatment when they went out or travelled, and only one patient acknowledged that he usually had difficul- ties to remember taking all his medication.
The assessment of the association between response variables (MMR and CMR) and therapeutic adherence offered the following results:
From those patients who presented high therapeutic adherence according to the Morisky-Green Questionnaire, 68.4% reached MMR; and from those who presented intermediate therapeutic adherence, 75% reached MMR. CMR was reached by 57.9% of patients with high therapeutic adherence and by 58.3% of patients with intermediate therapeutic adherence.
No statistically significant differences were observed in any of the response variables within the group of patients with high and intermediate therapeutic adherence. Thus, no association was observed between the level of treatment adherence according to the Adherence Questionnaire, or according to MPR, and the level of response to treatment.
Regarding the minimum value of BCR-ABL/ABL transcript expression for each patient, and based on the level of adherence, it was observed that the level of response was similar for patients with high and intermediate adherence, except for one patient who refused to take the optimal dose (classified as having good adherence to the 200mg dose by the Adherence Questionnaire and MPR, but underdosed), and who did not reach therapeutic response.
The demographical profile of the patients in the study is similar to the one found in bibliography, regarding mean age, gender, and mean time on treatment with imatinib at the time of the study; however, this study did not assess the education level of patients, the level of knowledge they had about their disease, or the level of involvement that they considered having in terms of clinical decisions for their treatment5,6,9.
When analyzing the level of therapeutic adherence, we observed that in the Swedish study assessing the level of therapeutic adherence through the 9-question MMAS questionnaire developed upon the 8-question Morisky-Green Questionnaire, the rate of patients clas- sified as having good treatment adherence was 97.4%, significantly higher than that detected in our study population through the 8-question Morisky-Green Questionnaire (61.3% with high adherence and 38.7% with intermediate adherence). It is worth pointing out that patients in our study were classified according to three levels of treatment adherence: high, intermediate and low, unlike the Swedish study, where patients were only differentiated in terms of having or not having treatment adherence, without specifying their level of adherence. The higher adherence among the Swedish population regarding other European studies could be determined, in the authors’ opinion, by differences in their healthcare system organization, and by demographic and socioeco- nomic factors9.
In the British study by Marin and cols. which evaluated treatment adherence through the electronic record of medication bottle opening and tests to determine the plasma levels of imatinib, the rate of patients with treatment adherence equal or lower to 90% was around 26%, and only 14.2% of patients were classified as having perfect adherence and taking 100% of the pres- cribed dose5.
In the Belgian study by Noens and cols., which evaluated treatment adherence according to interviews with patients and their relatives, visual analogue scale, compliance with appointments and tablet count, two third of patients were classified as having good adherence to treatment6, a value similar to the population in our study.
We consider that the rate of patients with high treatment adherence observed in our study is low, and below desirable levels, if we take into account the severity of the condition studied. We must take into account that the duration of treatment with imatinib is prolonged over time, because patients had been on treatment with imatinib for a mean 56.6 months at the time of the study. Previous studies have shown superior adherence rates, around 70% for drugs within the same therapeutic class15,16 , and other studies for adherence to chronic non-neoplastic therapies have shown even lower rates, around 40-50%17.
In the present study, all patients were classified as having good adherence to treatment when assessing MPR. This percentage is very superior to the one observed when analyzing adherence through the Morisky-Green Questionnaire, where only 61.3% of patients were classified as having good treatment adherence. Therefore, we believe that the Pharmacy Dispensation Record and the MPR estimation does not allow to obtain an accurate measure of treatment adherence, and the Morisky-Green Questionnaire is a better indicator for adherence.
We consider that treatment adherence questionnaires should always complement the information obtained through the Pharmacy Dispensing Records at the time of assessing adherence to a specific treatment. All measures could complete the assessment of treatment adherence, such as tablet count, the use of a visual analogue scale, and even testing in order to determine plasma levels of imatinib, such as conducted by other authors5,6.
Even though it has been observed in previous studies that there is a strong correlation between the level of treatment adherence and the rate of response to treatment5,6, no correlation between both variables has been observed in this study when comparing the response rates in the group of patients with high and intermediate treatment adherence (no patient without adherence), possibly due to the small sample size.
It can’t be confirmed that the different level of treatment adherence observed in the group of patients in the study can have any influence on treatment response. However, we believe that at the time of assessing a potential treatment failure or sub-optimal responses to treatment, low treatment adherence should be one more factor to consider among other potential mechanisms of resistance.
More studies are required to allow determining the level of association between adherence and treatment response in chronic therapies including oral cytostatic medications.
Contribution to scientific literature: The present article shows the outcomes of a study on therapeutic adherence to imatinib conducted throu- gh the Morisky-Green Questionnaire and the medication dispensing record. It included 31 patients diagnosed with chronic myeloid leukaemia in a third level hospital, and the level of therapeutic adherence was assessed, as well as its association with response to treatment. It can- not be concluded that a different level of therapeutic ad- herence will determine response to treatment; however, we think it should be one more factor to be taken into account in cases of therapeutic failure or sub-optimal response.
How to cite this article: García-Queiruga M, Martínez-López LM, Martín-Herranz MI, Seoane-Pillado T,Fernández-Gabriel E, Calleja-Chuclá T. Study on adherence to treatment with imatinib inchronic myeloid leukaemia and its association with therapeutic response.Farm Hosp. 2016;40(3):141-149.
Demographical Data
