Chronic Opioid Use Is Highly Prevalent in Patients Undergoing Allogeneic Transplant and Impacts Long Term Outcomes

Opioid analgesics are used to treat cancer-related pain and improve quality of life, however overuse of these high-risk drugs has been associated with significant public health implications as exhibited by the national attention received during the opioid pandemic. In addition, pain associated with hematologic malignancies is frequent yet not well understood. There is no data examining opioid use and outcomes in patients with hematologic malignancies and recipients of allogeneic hematopoietic stem cell transplantation (allo-HSCT). Therefore, we sought to describe chronic opioid use (COU) and its impact on long-term outcomes in patients undergoing allo-HSCT.

We analyzed outcomes of adult patients ≥ 18 years (n=159) diagnosed with hematologic malignancies (n=151) or benign hematologic disorders (n=8) (excluding sickle cell disease) who received allo-HSCT between 2012 and 2019 at a single urban medical center. COU was defined as having a documented active prescription for 3 consecutive months. Daily opioid doses were converted to morphine milligram milliequivalents (MME) using standard conversion factors.

Among the entire cohort, the median age was 55.4 (19.4-74.3) years, 94 (59.1%) were male, 70 (44%) were White, 42 (26.4%) were Hispanic, and 26 (16.4%) were Black. The majority of patients were diagnosed with acute leukemia, with 89 (56%) having AML and 16 (10%) having ALL patients. Of the remaining patients, 14 (8.8%) were CML-BP or CML-AP, 16 (10%) NHL/HD, and 15 (9.4%) MPDs. 33.5% of patients were found to have high or very high DRI prior to allo-HSCT. Most patients received either a myeloablative (n=102, 64.2%) or reduced-intensity (n=55, 34.6%) conditioning regimen.

At baseline (immediately prior to allo-SCT), COU was observed in 38 (23.9%) patients, 20 (52.6%) of which were diagnosed with AML. Only 23 (60.5%) patients with COU had a documented indication for opioid analgesia. The baseline median MME per day was 23.5mg (range: 2-150). In logistic regression analysis including demographic and social factors such as insurance type (i.e. Medicare, Medicaid, private payer), education level, smoking, alcohol, illicit drug and/or benzodiazepine use, and employment status, only older age (RR 0.97, 95% CI 0.94-0.99; p=0.026) was associated with a lower likelihood of baseline COU.

In total, 149 (93.7%) patients received opioids during the allo-HSCT admission, while 45 (28.3%) were discharged on opioids. The reason for being discharged on opioids was related to musculoskeletal pain (n=19), residual mucositis-related pain (n=7), and headache (n=3). This was found to persist over time, with 35 (92.1%) of the 45 patients discharged on opioids remaining on opioids at 180 days after allo-HSCT thus meeting the definition for COU. The only factor found to predict for COU at 6 months was discharge from initial transplant hospitalization with an opioid (RR 2.24, 95% CI 1.16-4.32, p=0.016). Not only did a significant number of post-transplant survivors meet the definition for COU, but the MME was found to be relatively high with a median of 50mg (range: 10-540) at discharge and 30mg (range: 4.5-202) on days +30, 90, 180 as well as at 1 and 5 years after allo-HSCT.

In a multivariable modified Poisson regression with robust standard errors for binary outcomes, when adjusted for established prognostic characteristics (i.e., disease, DRI, HCT-CI), we found that COU prior to admission strongly predicted for worse overall survival (HR 2.99, 95% CI 1.59-5.64; p=0.001), progression free survival (HR 2.72, 95% CI 1.48-4.99), and GVHD free, relapse-free survival (HR 1.78, 95% CI 1.05-3.03; p=0.033).

This study is the first to describe patterns of COU, an important public health problem, among patients undergoing allo-HSCT, and in particular in long term survivors. We demonstrate high rates of baseline COU in patients undergoing allo-HSCT as well as persistent long term COU, which is linked to prescribing patterns after the initial transplant hospitalization. In addition, we show the negative impact of baseline COU on overall survival, even when adjusted for disease- and transplant-related factors. These data highlight the need to improve understanding and management of pain in hematologic malignancies as well as to reinforce the need for continuous reassessment of the use of opioids prior to and after allo-HSCT.