Single Arm, Single Centre Prospective Study to Assess the Effect of Therapeutic Drug Monitoring (TDM) Based Dosage Adjustment of Posaconazole on the Incidence of Invasive Fungal Infections (IFIs) in AML Patients on Induction Chemotherapy on Posaconazole Prophylaxis

Background:

Invasive fungal infections (IFIs) continue to remain a significant cause of morbidity and mortality in AML patients affecting the final outcomes and increasing health care costs. Posaconazole is commonly used as prophylaxis against IFIs due to its broad spectrum of action. The incidence of breakthrough IFIs is particularly high in our setting - one of the several reasons being suboptimal drug levels as we use posaconazole suspension whose absorption is influenced by multiple factors. A breakthrough IFI rate of 51% was seen in a previous study from our centre by Sengar et.al (ASH 2016) and it was seen that sub-optimal drug levels (< 700ng/ml) were associated with a higher rate of breakthrough IFI. This group can benefit from TDM but there are not many studies to guide TDM-based dosing of posaconazole. TDM is expensive and time consuming and the final impact on clinical outcomes has been difficult to estimate. To determine the impact of TDM-based dosing of posaconazole on breakthrough IFI rate, we have undertaken this prospective study.

Methods:

This prospective, single-arm interventional study included patients 16 years or older undergoing induction chemotherapy for de novo AML with no evidence of IFI at baseline (normal CT chest and normal serum galactomannan ) who have consented for participation (assent and guardian consent were used for those who were < 18 years). Concomitant drugs during induction were recorded everyday. 3 dosing schedules for posaconazole were used based on the drug level : Level 1 - 200mg TDS which was the starting dose in all patients, Level 2 - 200mg QID and Level 3 - 400mg TDS which was the maximum dose. Posaconazole was given with a high fat meal or an aerated drink to improve its absorption. The cut off for day 2 drug level (i.e after atleast 3-4 doses ) was 350ng/ml and the cut-off level at subsequent time points was 700 ng/ml. At any time point, if the patient's drug level was below this cut off, the dose was escalated and a repeat level was sent 5 complete days after the changed dose to allow for re-establishment of the pharmacokinetic equilibrium. A day 2 level was sent for all the patients and the timing of subsequent levels was based on dose escalation. Patients who were continued on the same dose due to adequate drug levels were sampled once in 7 days. QTc interval and liver function tests were monitored for posaconazole toxicity. Posaconazole was continued till the absolute neutrophil count recovered to more than 500 cells/ µl or till the patient was started on an alternative anti-fungal for a proven , possible or probable IFI or empirically at the clinician's discretion for reasons like persistent fever, cough, breathlessness, diarrhoea or vomiting. Relevant imaging and microbiological tests were performed as needed for investigating a suspected IFI. To analyze the primary endpoint chi-square test was used to compare the breakthrough IFI rate with that in a historical cohort from our own centre in which TDM based dosing was not used.

Results:

A total of 278 samples were collected for drug levels in these 90 patients with a mean of 3.08 (range 1 to 6 samples) per patient . 2 patients were proven to have mixed phenotypic acute leukemia after enrollment and hence were excluded from the analysis. 46 out of 88 patients i.e 52.3% were changed to alternative antifungals. Out of these 46, 13 patients (i.e 28.3%) had proven, probable or possible IFI as per the Revised Definitions of Invasive Fungal Disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) . The change to alternative antifungals in the remaining 33 out of 46 patients (i.e 71.7%) was empirical due to persistent fever spikes, vomiting or diarrhoea. Thus overall 13 out of 88 patients i.e 14.8% had developed a breakthrough IFI. A similar historical cohort from our own institute, without the TDM based dosing of posaconazole, had a breakthrough IFI rate of 52%. The difference between the breakthrough IFI rate of these 2 cohorts was 37.2% which was statistically significant (p < 0.0001, 95% CI 27.9% to 52.4%). 1 patient had developed grade 3 transaminitis and 1 patient had grade 3 rise in total bilirubin while on posaconazole.

Conclusion:

TDM-based dose escalation of prophylactic posaconazole in AML patients during induction leads to a significant reduction in the breakthrough IFI rates.