Real-World Clinician Adherence to Screening and Preventive Practices Guidelines for Subsequent Neoplasms Among Allogeneic Hematopoietic Cell Transplant Survivors in the United States

Background: Prior studies have reported variable adherence to screening/prevention guidelines from allogeneic hematopoietic cell transplant (HCT) survivors’ perspective (Khera N et al. Biol Blood Marrow Transplant. 2011 Jul;17(7):995-1003). However, no prior study has assessed clinician adherence to these guidelines. To address this knowledge gap, we aimed to examine clinician adherence to the recommended assessments for subsequent neoplasms (SN) screening according to the 2011 guidelines (Majhail NS et al. Biol Blood Marrow Transplant. 2012 Mar;18(3):348-71).

Methods: A retrospective, multi-center cohort study of patients (pts) who underwent HCT for hematologic malignancies between 01/01/2016-12/31/2016 was conducted. Pts who were alive and disease free for at least 3 years post-HCT were included. Data on key screening practices focusing on breast, skin, and cervical neoplasm were abstracted from electronic medical records (EMR). “Adherence” was defined by evidence of an order or test result for each screening test between 1-3 years post-HCT. Two separate multivariable logistic regression models were created to assess the factors associated with adherence to SN screening. Model 1 included a best subset of patient-, disease-, and treatment-related factors such as patient age, sex, race/ethnicity, highest education, preferred language, disease diagnosis, HCT-comorbidity index (HCT-CI) score, history of prior malignancy, conditioning intensity, total body irradiation (TBI) use, history of acute and/or chronic GVHD, and high-dose corticosteroid use, and post-HCT follow-up care providers. Model 2 consisted of HCT-center related factors such as type of center (academic vs. non-academic), center volume, availability of survivorship clinic, provision of survivorship care plan to pts, referring providers, or primary care providers. Odds ratio (OR) and 95% confidence interval (CI) were provided.

Results: A total of 440 pts from 19 HCT centers with a median age at HCT of 46 (range 0-77) years were included in the analysis. Forty two percent were females, 18% were of non-White race, 12% were of Hispanic or Latino ethnicity, and 5% preferred a language other than English. A quarter of recipients had an HCT-CI score ≥3 and 7% had a prior history of a solid tumor. AML (43%), ALL (22%), and MDS (11%) were the most common indications for HCT. Seventy one percent received myeloablative conditioning and TBI was used in 42% of the patients. History of acute and chronic GVHD was noted in 57% and 52% of the patients, respectively. Half of the pts had a history of high dose corticosteroid exposure.

Adherence to breast cancer, skin cancer, and cervical cancer screening were noted in 49%, 34%, and 41% of the pts, respectively. Increasing pt age was significantly associated with increase in breast cancer screening adherence [OR 1.08, 95% CI 1.03-1.12, p<0.001]. No other clinical or HCT-center factors were associated with breast cancer screening adherence. Compared to those with HCT-CI score of 0, the odds of skin cancer screening adherence were lower with HCT-CI score of 1 [0.29, 0.11-0.77, p=0.012] and 2 [0.14, 0.03-0.56, p=0.006], but not ≥3 [0.73, 0.31-1.74, p=0.48]. Exposure to high dose corticosteroids was also associated with lower odds of skin cancer adherence [0.47, 0.26-0.83, p=0.01], whereas exposure to TBI was associated with higher skin cancer screening adherence [1.93, 1.05-3.55, p=0.03]. Increasing annual number of transplants [1.01, 1.00-1.01, p<0.001] and provision of survivorship care plan to pts [2.53, 1.45-4.42, p<0.001] were associated with increased skin cancer screening adherence. While a similar association between higher annual number of HCT and cervical cancer screening adherence was noted [1.01, 1.00-1.02, p=0.008], an inverse association was noted with the provision of survivorship care plan to pts [0.15, 0.03-0.80, p=0.03]. History of chronic GVHD was also associated with higher odds of cervical cancer screening.

Conclusions: To our knowledge, this is the first report of EMR-based clinician adherence to screening/prevention guidelines for HCT survivors in the United States. More than half of the eligible survivors did not undergo subsequent neoplasm screening. We found several factors associated with adherence, which will provide an opportunity to develop and test interventions to improve clinician adherence to screening and preventive practice guidelines.