Therapeutic pathways of allogeneic and autologous hematopoietic stem cell transplantation recipients: a hospital pharmacist’s perspective

E-ISSN: 2340-9894 ISSN: 0004-2927 https://revistaseug.ugr.es/index.php/ars doi: 10.30827/ars.v65i3.340246 Artículos originales Therapeutic pathways of allogeneic and autologous hematopoietic stem cell transplantation recipients: a hospital pharmacist’s perspective Trayecto terapéutico de receptores de trasplante de células madre hematopoyéticas: una perspectiva de farmacéuticos hospitalarios David Malnoë1,2,3 orcid 0000-0002-5650-852X Timothé Lamande1,a Alexia Jouvance-Le Bail1 Tony Marchand4 Pascal Le Corre1,2,3 orcid 0000-0003-4483-0957 1Centre Hospitalier Universitaire de Rennes, Pôle Pharmacie, Secteur Pharmacotechnie et Onco-Pharmacie, 35033 Rennes, France 2 Université de Rennes 1, Faculté de Pharmacie, Laboratoire de Biopharmacie et Pharmacie Clinique, 35043 Rennes, France 3Université Rennes, Inserm, EHESP, IRSET (Institut de recherche en santé, environnement et travail)-UMR_S 1085, F-35000 Rennes, France 4Université de Rennes, Service d’Hématologie Clinique, CHU de Rennes, INSERM U1236, 35000 Rennes, France a – current address : Clinique de la Côte d’Emeraude, Service de Pharmacie, 35400 Saint-Malo, France. Correspondence Pascal Le Corre pascal.le-corr[email protected] Received: 26.02.2024 Accepted: 19.05.2024 Published: 20.06.2024 Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Conflict of interest The authors declare that there is no conflict of interest. Ars Pharm. 2024;65(3):240-257 240 Resumen Introducción: Pacientes de trasplante de células madre hematopoyéticas autólogo y alogénico (Alo-TCMH y Auto-TCMH) enfrentan riesgos farmacoterapéuticos. Objetivo: Detallar el perfil terapéutico y la evolución de biomarcadores de disfunción renal, hepática e inflamatoria en pacientes de Aloy Auto-TCMH desde su ingreso hasta el alta hospitalaria, ofreciendo una perspectiva detallada del manejo farmacológico. Método: Se extrajeron datos retrospectivos de las historias clínicas de 20 pacientes de Alo-TCMH y 20 de Auto-TCMH. Se describió el trayecto terapéutico mediante el cambio de tratamientos farmacológicos, los medicamentos potencialmente inapropiados utilizando la escala GO-PIM, y la carga anticolinérgica (CA). Se evaluaron las variaciones fisiopatológicas afectando órganos de eliminación, mediante niveles de proteína C reactiva (PCR), puntuación para la enfermedad hepática en etapa terminal (puntuación MELD) y filtración glomerular (FG). Resultados: Alo-TCMH pacientes tuvieron un mayor número de fármacos iniciados durante la estancia hospitalaria, lo que llevó a una hiperpolifarmacia durante la estancia y al alta. Un 35% de los medicamentos usados eran metabolizados por CYP3A4. CA aumentó al alta en pacientes de HSCT. Los pacientes de Auto-TCMH ≥ 65 años tomaban al menos un PIM. Se informaron niveles altos de CRP en los receptores de TCMH. Puntuación MELD aumentó y la GFR disminuyó en pacientes de Alo-TCMH mientras que la FG aumentó ligeramente en pacientes de Auto-TCMH. Conclusión: El farmacéutico clínico debe enfocarse en la polifarmacia, PIM y CA, y evaluar la inflamación y las funciones renales y hepáticas para evaluar de manera reflexiva el potencial de depuración de los pacientes y sugerir dosificaciones individualizadas. Palabras clave: Trasplante de células madre hematopoyéticas; Proteína C reactiva; Insuficiencia hepática, Lista de medicamentos potencialmente inapropiados. Abstract Introduction: Patients undergoing allogeneic and autologous hematopoietic stem cell transplantation (Allo-HSCT and Auto-HSCT) are at risk of pharmacotherapy-related problems. Objective: To describe in Allo-HSCT and Auto-HSCT patients from admission to hospital discharge, their therapeutic profile, and the time-course of biomarkers of renal and liver dysfunction, and of inflammation to display a more specific overview of drug therapy in HSCT patients. Method: Data were retrospectively extracted from the charts of 20 Allo-HSCT and 20 Auto-HSCT patients. The therapeutic pathway was described by the turn-over of drug treatments, the potentially inappropriate medications by using the GO-PIM scale, and the anticholinergic burden. Patho-physiological variations affecting clearance organs were characterized by the C-Reactive Protein (CRP) levels, and the hepatic and renal impairment evaluation tools (Model for End-stage Liver Disease score: MELD score, and glomerular filtration rate: GFR). Results: Compared to Auto-HSCT patients, Allo-HSCT patients had a higher number of drugs initiated during hospital stay leading to hyper-polypharmacy during the stay and at discharge. Around 35 % of drugs used were metabolized by CYP3A4 in HSCT patients. Anticholinergic burden increased at discharge in HSCT patients. Auto-HSCT patients ≥ 65 years were taking at least one PIM. High CRP levels were reported in HSCT recipients. MELD score increased and GFR decreased in Allo-HSCT patients while GFR slightly increased in Auto-HSCT patients. Conclusion: Clinical pharmacist should target polypharmacy, PIM and anticholinergic burden, and evaluate inflammation and both renal and hepatic functions in order to thoughtfully assess the clearance potential of patients and to suggest individualized dosing. Keywords: hematopoietic stem cell transplantation; C-reactive protein; Hepatic Insufficiency; list of potentially inappropriate medications. Highlights Beyond general guidelines and recommendations that have defined the role of hospital pharmacists in caring for hematopoietic stem cell transplantation (HSCT) patients, this study investigated specific pharmacotherapeutic and biological features in both allogeneic and autologous HSCT patients from admission to discharge. This study emphasizes that anticholinergic burden, potentially inappropriate medication (according to the GO-PIM scale), and hepatic impairment (by using MELD-score) should be evaluated throughout the Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 241 hospitalization stay. Elevated levels of C-reactive protein raise concerns since inflammation induces metabolic down-regulation, and noteworthy of CYP3A4 which is very frequently involved in the elimination of drugs used in these patients. Clinical pharmacist should consider specificities of drug treatment and of patho-physiological variations affecting clearance organs to thoughtfully assess the clearance potential of patients and to suggest individualized dosing. Introduction Patients with hematological malignancies, especially those undergoing hematopoietic stem cell transplantation (HSCT), face a high risk of pharmacotherapy-related problems due to complex drug regimens and patho-physiological variations affecting clearance organs. Potential drug-drug interactions (DDIs) are particularly common among HSCT patients in the bone marrow transplantation unit(1). The identification and resolution of drug-related problems (DRP) constitute a very important role of the clinical pharmacist in managing drug therapy, and several general guidelines and recommendations have been provided to define the role of hospital pharmacists in caring for HSCT patients(2-6). Besides general guidelines and recommendations, clinical pharmacists should pay close attention to specific aspects of drug treatments, as exposure to polypharmacy (PP), hyper-polypharmacy (HPP, > 10 drugs), potentially inappropriate medications (PIM) including drugs with anticholinergic properties. Recently, a list of PIM specific to geriatric oncology has been proposed (Geriatric Oncology Potentially Inappropriate Medications, GO-PIM scale) based on the NCCN Clinical Practice Guidelines in Oncology for Older Adult Oncology(7). Some of these features of drug treatments (PP, HPP or PIM) have been associated with negative clinical outcomes in older adults with blood cancers(7). in patients with acute myeloid leukemia(8-9), non-Hodgkin’s lymphoma(10), or in patients undergoing allogeneic HSCT(11-12). Furthermore, kidney and liver impairment should be evaluated as HSCT patients are at an increased risk of developing early and late complications(13-14). Recently, Model for End-stage Liver Disease score (MELD score) has been proposed as a screening tool to identify patients with hepatic impairment (HI) who are at risk of drug safety issues(15). Inflammation has been recognized as a relevant factor that inhibits the metabolic activities of CYP450s isoforms, especially CYP3A4 and CYP2C19 thereby potentially influencing hepatic clearance and intestinal/hepatic first-pass effect(16). The purpose of this study was to describe in allogeneic and autologous HSCT patients, from admission to hospital discharge, the therapeutic profile of patients with regard to PP, HPP, GO-PIM and anti-cholinergic burden, as well as the time-course of biomarkers of renal and liver dysfunction, and inflammation status in order to bring to hospital pharmacists a more specific overview of drug therapy in these patients. Methods Study design, setting and population This retrospective, observational, single-center study (from January 2020 to December 2021) involved adult inpatients of the Clinical Hematology department of our University Hospital. Clinical data were extracted from electronic health records (EHR) using the computerized physician order entry database (CPOE, DxCare Software). Given that there was no aim of statistical comparison between allogeneic and autologous patients, clinical data 20 allogeneic and 20 autologous HSCT patients were considered representative for the descriptive study and were randomly retrieved from the database of patients. All allogeneic and autologous patients registered in the JACIE (Joint Accreditation Committee ISCT-Europa & EBMT) database were assigned unique identification numbers ranging from 1 to 127. To select a representative subset of allogeneic and autologous patients for the study, a randomization procedure was conducted for each group using Microsoft Excel’s random number generation function to randomArs Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 242 ly choose 20 patients. Following the selection of patient records, de-identification was performed to ensure confidentiality. The study received approval from the Institutional Research Ethics Committee of our University Hospital (agreement n° 23.84). It was conducted in accordance with the ethical standards set forth in the 1964 Declaration of Helsinki and its subsequent amendments, or comparable ethical standards. Due to the retrospective and non-interventional nature of the study, utilizing data from a database, a consent waiver was granted. The principles of ethical research, such as confidentiality and anonymity, were strictly followed. Patient data collection Drug treatments were documented upon hospital admission, throughout the hospital stay, and at discharge. PIM were assessed using the cancer-specific Geriatric Oncology Potentially Inappropriate Medications (GO-PIM) scale based on the NCCN Clinical Practice Guidelines in Oncology for Older Adult Oncology(7). This scale includes a list of medications commonly used for supportive care that are of concern for older adults (NCCN). The anticholinergic burden was evaluated using the Anticholinergic-Cognitive-Burden Scale (ACBS,(17)), and the Anticholinergic-Impregnation Scale (AIS,(18)) which estimates potential peripheral anticholinergic adverse effects. Information on the metabolic pathways of the drugs used was obtained from Drugbank 5.0(19) or relevant literature through PubMed when not available. The following laboratory parameters were retrieved upon hospital admission, the day after the bone marrow transplantation (BMT), and at discharge. • Serum creatinine levels (SCrea) for estimating glomerular filtration rate (GFR) using the CKD-EPI equation. • SCrea, bilirubin and International Normalized Ratio (INR) for calculation of the Model for End-stage Liver Disease score (MELD score), a screening tool to identify patients with hepatic impairment (HI) who are at risk of drug safety issues(15). • C-reactive protein (CRP) for estimating of the degree of inflammation. Statistical analysis No statistical comparison between allogeneic and autologous HSCT patients was performed. To assess the differences before and after allogeneic or autologous HSCT, paired t-tests were employed allowing for the comparison of means between two measurements taken on the same individuals, accounting for individual variability. A p-value less than 0.05 was considered statistically significant. All statistical analyses were performed using Microsoft Excel. Results The patient characteristics of allogeneic and autologous HSCT patients are presented in Table 1. Both myeloablative conditioning (MAC) and reduced intensity conditioning (RIC) were employed in allogeneic HSCT patients, using various drugs, which led to a high degree of heterogeneity in terms of treatment intensity and associated toxicities. In contrast, autologous patients typically received a one-drug regimen involving melphalan (140 mg/m² n=7, or 200 mg/m² n= 8) as their conditioning treatment. Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 243 Table 1. Characteristics of allogeneic and autologous HSCT patients. Allogeneic HSCT Autologous HSCT Patient demographics Number of patients 20 20 Median age (years, median (range))53.5 (25 - 67) 58.5 (19 - 69) Female 7 9 Male 13 11 Cancer type Acute Myeloid Leukemia (AML) 12 - Acute Lymphoid Leukemia (ALL) 1 - Multiple Myeloma (MM) - 15 T-Lymphoma 1 - Myelofibrosis 3 - Hodgkin Lymphoma 1 5 Chronic Myelomonocytic Leukemia (CMML) 1 - Refractory Anemia with Excess Blasts (RAEB) 1 - Conditioning treatment Myeloablative conditioning (MAC) 6 - Reduced intensity conditioning (RIC) 14 - Melphalan 15 Carmustine, Etoposide, Cytarabine, Melphalan +/- rituximab (BEAM or R-BEAM) - 3 Thiotepa, Busulfan - 2 Hospitalization Length of stay, (days, median (range))40.9 (28 - 82) 18.6 (13 - 36) Duration of aplasia (days, median (range))13 (6 - 33) 6 (4 - 11) Time from admission to BMT (days, median) 9.1 4.9 Time from BMT to discharge (days, median) 31.8 13.8 Therapeutic pathway The therapeutic pathway, excluding anticancer drug conditioning treatment, for allogeneic and autologous HSCT patients from admission to discharge, is depicted in Figure 1. HSCT patients at discharge can be categorized as having polypharmacy (PP, 5-9 drugs, nAllogeneic= 4 [20 %], nAutologous= 9 (45 %]) or hyper-polypharmacy (HPP, 10 or more drugs, nAllogeneic= 15 [75%], nAutologous= 8 [40 %]). Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 244 Figure 1. Therapeutic pathway of allogeneic (top) and autologous (bottom) HSCT patients from admission to discharge (mean number of drugs, n = 20 in each group). Renal function In patients undergoing autologous HSCT renal function significantly improved throughout the hospital stay in all patients (mean increase + 17.5 %) from admission to discharge (94.9 ± 18.7 ml/min vs 109.9 ± Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 245 17.3 ml/min, P-value: 1.66E-06). On the other hand, allogeneic HSCT patients renal function decreases from admission to discharge by 16.3 % (104.4 ± 11.4 ml/min vs 87.8 ± 24.0 mL/min, P = 2,01E-03, Figure 2). Figure 2. Evolution of glomerular filtration rate (GFR, ml/min/1.73 m2) in allogeneic (left) and autologous (right) HSCT patients at admission, the day after the BMT, and at discharge (median, Q1-Q3, and min-max, n = 20 in each group). Inflammation Allogeneic and autologous HSCT recipients had CRP levels peaking around 131 mg/L and 117 mg/L, respectively, after transplantation. At discharge, CRP levels were 10 to 4-times lower than peak levels in allogeneic and autologous HSCT recipients but they remained 2 to 6-times higher than levels at admission (Fig. 3). Two patients in the autologous group had CRP levels > 100 mg/L at discharge (Figure 3). Figure 3. Evolution of C-reactive protein (CRP in mg/L) in allogeneic (left) and autologous (right) HSCT patients at admission, the day after the BMT, at the peak during hospitalization, and at discharge (median, Q1-Q3, and min-max, n = 20 in each group). Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 246 Liver function & MELD score The mean MELD score was lower than 7.5 in both allogeneic and autologous HSCT patients at admission. It was not significantly different from admission to discharge for autologous HSCT patients. On the other hand, the mean MELD score significantly increased to 9.0 for allogeneic patients (P = 2,89E03) with one third of patients having a MELD score above 10 (corresponding to a Child Pugh Score B, Figure 4). Figure 4. Evolution of model for end-stage liver disease (MELD) score in allogeneic (top) and autologous bottom) HSCT patients at admission, the day after the BMT, and at discharge (median, Q1-Q3, and min max, n = 20 in each group) and Child Pugh liver function estimation through MELD score. Anticholinergic burden The central anticholinergic burden, measured by ACB scores, at admission and discharge for allogeneic and autologous HSCT patients, is low and doesn’t show any differences throughout hospitalization. The peripheral anticholinergic burden (AIS scale) is higher at discharge compared to admission in both allogeneic (P-value: 6,13E-03) and autologous (P-value: 4,33E 03) HSCT patients. It is slightly higher in allogeneic HSCT patients compared to autologous HSCT patients (Figure 5, Table 2). Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 247 Figure 5. Anticholinergic burden estimated according by the anticholinergic-impregnation scale (AIS for peripheral effects) and by the ACB score (anticholinergic cognitive burden, central effects) measured at admission and at discharge in allogeneic (top) and autologous (bottom) HSCT patients. Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 248 Drugs Frequency (%) Metabolic pathway Reference VANCOMYCINE 50 almost not metabolized * FUROSEMIDE 50 CYP2C11, 2E1, 3A1, and 3A2 Yang 2009 METOCLOPRAMIDE 50 CYP2D6, CYP3A4 and CYP1A2 * MORPHINE 50 UGT2B7 * LANSOPRAZOLE 45 CYP3A4 and CYP2C19 * LETERMOVIR 35 UGT1A1 and UGT1A3: Minimal * FILGRASTIM 35 non CYP450 * ACICLOVIR 35 minimal, via alcohol dehydrogenase and aldehyde dehydrogenase * CETIRIZINE 35 minor Renwick 1999 POSACONAZOLE 30 primarily glucuronidation * HYDROXYZINE 30 hydrolysis and N-acetylation * METHYLPREDNISOLONE 30 ND * PANTOPRAZOLE 30 CYP2C19, sulfation and CYP3A4 * PREDNISONE 30 ND * DEXCHLORPHENIRAMINE 25 CYP2D6, CYP3A4, and glucuronidation or sulfation * CLORAZEPATE POTASSIQUE 25 CYP 2C19 and 3A4 Riss 2008 CHLORPROMAZINE 20 CYP2D6 (major pathway), CYP1A2 and CYP3A4 * VORICONAZOLE 15 Extensive via CYP2C19, CYP2C9 and CYP3A4 * CASPOFUNGINE 15 independent of CYP450, hydrolysis and N-acetylation * NICARDIPINE 15 extensive via CYP2C8, CYP2D6, and CYP3A4 * VALGANCICLOVIR 15 esterases * Li XQ, Björkman A, Andersson TB, et al. Identification of human cytochrome P(450)s that metabolise anti-parasitic drugs and predictions of in vivo drug hepatic clearance from in vitro data. Eur J Clin Pharmacol. 2003;59(5-6):429-42. doi: 10.1007/s00228-003-0636-9. Mittur A. A Simultaneous Mixed-Effects Pharmacokinetic Model for Nefopam, N-desmethylnefopam, and Nefopam N-Oxide in Human Plasma and Urine. Eur J Drug Metab Pharmacokinet. 2018;43(4):391404. doi: 10.1007/s13318-017-0457-3. Becquemont L, Mouajjah S, Escaffre O, et al. Cytochrome P-450 3A4 and 2C8 are involved in zopiclone metabolism. Drug Metab Dispos. 1999 Sep;27(9):1068-73. Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 255 Yang KH, Choi YH, Lee U, et al. Effects of cytochrome P450 inducers and inhibitors on the pharmacokinetics of intravenous furosemide in rats: involvement of CYP2C11, 2E1, 3A1 and 3A2 in furosemide metabolism. J Pharm Pharmacol. 2009;61(1):47-54. doi: 10.1211/jpp/61.01.0007. Renwick AG. The metabolism of antihistamines and drug interactions: the role of cytochrome P450 enzymes. Clin Exp Allergy. 1999;29 Suppl 3:116-24. doi: 10.1046/j.1365-2222.1999.0290s3116.x. Riss J, Cloyd J, Gates J, Collins S. Benzodiazepines in epilepsy: pharmacology and pharmacokinetics. Acta Neurol Scand. 2008;118(2):69-86. doi: 10.1111/j.1600-0404.2008.01004.x. Table 3. Ranking of drugs administered to autologous HSCT patients during hospitalization estimated by the frequency of patients that received the drugs, and their metabolic pathways (Informations retrieved from Drugbank (*), and when not available retrieved from literature). Drugs Frequency (%) Metabolic pathway Reference VALACICLOVIR 100 esterase * PENTAMIDINE ISETHIONATE 95 CYP1A1 Li 2003 NEFOPAM 95 CYP1A2, CYP2C19 and CYP2D6 Mittur 2018 CEFEPIME 95 almost not metabolized * ALIZAPRIDE 90 ND * ALPRAZOLAM 90 3A4 extensive * RACECADOTRIL 90 ND * ZOPICLONE 90 CYP3A4 and CYP2C8 Becquemont 1999 MACROGOL 80 not metabolized * PEGFILGRASTIM 80 non CYP450 * ONDANSETRON 65 CYP1A2, CYP2D6 and CYP3A4 * TRAMADOL 65 extensive CYP2D6 and CYP3A4, CYP2B6 * SULFAMETHOXAZOLE and TRIMETHOPRIME 60 NAT and CYP2C9 // CYP2C9, CYP3A4 and CYP1A2 * ACICLOVIR 50 minimal, alcohol dehydrogenase and aldehyde dehydrogenase * FLUCONAZOLE 50 minimal * MORPHINE 50 UGT2B7 * PHLOROGLUCINOL 50 ND * PHYTOMENADIONE 50 CYP4F2 * AMOXICILLINE or AMOXICILLINE/ CLAVULANATE 40 ND but minimal * Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 256 Drugs Frequency (%) Metabolic pathway Reference FILGRASTIM 40 non CYP450 * AMPHOTERICINE B 35 not metabolized * CHLORPROMAZINE 35 CYP2D6 (major pathway), CYP1A2 and CYP3A4 * METOCLOPRAMIDE 25 CYP2D6, CYP3A4 and CYP1A2 * DEXCHLORPHENIRAMINE 25 CYP2D6, CYP3A4,and glucuronidation or sulfation * LANSOPRAZOLE 25 CYP3A4 and CYP2C19 * VANCOMYCINE 25 almost not metabolized * CLORAZEPATE POTASSIQUE 20 CYP 2C19 and 3A4 Riss 2008 MEROPENEM 20 almost not metabolized * OXYCODONE 15 CYP3A4 and CYP2D6 extensive * PANTOPRAZOLE 15 CYP2C19, sulfation and CYP3A4 * PARACETAMOL 15 conjugation and CYP2E1 * PREGABALINE 15 almost not metabolized * APREPITANT 10 CYP3A4 major and CYP1A2 and CYP2C19 * VORICONAZOLE 5 CYP2C9, CYP2C19, and CYP3A4 * Li XQ, Björkman A, Andersson TB, et al. Identification of human cytochrome P(450)s that metabolise anti-parasitic drugs and predictions of in vivo drug hepatic clearance from in vitro data. Eur J Clin Pharmacol. 2003;59(5-6):429-42. doi: 10.1007/s00228-003-0636-9. Mittur A. A Simultaneous Mixed-Effects Pharmacokinetic Model for Nefopam, N-desmethylnefopam, and Nefopam N-Oxide in Human Plasma and Urine. Eur J Drug Metab Pharmacokinet. 2018;43(4):391404. doi: 10.1007/s13318-017-0457-3. Becquemont L, Mouajjah S, Escaffre O, et al. Cytochrome P-450 3A4 and 2C8 are involved in zopiclone metabolism. Drug Metab Dispos. 1999 Sep;27(9):1068-73. Riss J, Cloyd J, Gates J, Collins S. Benzodiazepines in epilepsy: pharmacology and pharmacokinetics. Acta Neurol Scand. 2008;118(2):69-86. doi: 10.1111/j.1600-0404.2008.01004.x. creative-commons BY-NC-SA 4.0 Ars Pharm. 2024;65(3):240-257 Malnoë D, Lamande T, Jouvance-Le Bail A, et al. 257