Elderly Medication Adherence Intervention Using the My Interventional Drug-Eluting Stent Educational App: Multisite Randomized Feasibility Trial

Background A lifesaving treatment for myocardial infarction is the placement of a stent in a closed or obstructed coronary artery. The largest modifiable risk factor after receiving a stent is medication adherence to Dual AntiPlatelet Therapy, a combination of P2Y12 inhibitors and aspirin. Objective This study aimed to determine the acceptability of a protocol and an intervention using the My Interventional Drug-Eluting Stent Educational App (MyIDEA) and to evaluate medication adherence using the proportion of days covered (PDC) and platelet activation tests in a multisite randomized controlled trial. Methods Potential participants who received a post percutaneous coronary intervention (PCI) procedure with a drug-eluting stent were approached. All patients older than 50 years and who spoke English were recruited. Participants were recruited, baseline demographics were collected, and the Hospital Anxiety and Depression Scale (HADS), Rapid Estimate of Adult Literacy in Medicine-Short Form, Burden-Benefit questionnaire, 36-Item Short Form Health Survey, and PCI knowledge questionnaire were administered. Block randomization was used to randomize participants to either usual care or MyIDEA supplementation. MyIDEA is a personalized educational intervention based on the Kolb experiential learning theory using patient narratives for education. During the visits, participants’ blood was collected to measure platelet suppression from medication. During the second and third encounters, the Morisky medication adherence score and cardiology outcomes were measured. The study was conducted at the University of Illinois Hospital and John H Stroger Jr Cook County Hospital with appropriate ethical approvals. Platelet suppression was measured through aspirin reactive units and P2Y12 reactive units. Medication adherence was measured using the PDC. The analysis team was blinded to the participants’ group membership. The primary outcome was a feasibility analysis of recruitment and retention. Results The mean age of participants was 60.4 years (SD 7.1); the majority of patients were black and non-Hispanic. The majority of patients’ reading levels were seventh grade or above, and they were not very familiar with other electronic devices for information and communication. The number of control subjects was 21, and the number of participants in the interventional arm was 24. The interventional group was able to use MyIDEA in both the hospital and outpatient setting. However, there was no significant difference in platelet suppression or medication adherence between groups. There were also differences between the groups in terms of depression and anxiety, initially, as measured by HADS. No documented adverse event associated with the intervention was found. Conclusions Elderly patients are willing to use tablet devices to be educated about health conditions. Additional studies are required to measure the effectiveness and determine the most suitable timing and location for patient education. Trial Registration ClinicalTrials.gov NCT04439864; https://clinicaltrials.gov/ct2/show/NCT04439864

Elderly Medication Adherence Intervention Using the My Interventional Drug-Eluting Stent Educational App: Multisite Randomized Feasibility Trial (Preprint)

BACKGROUND A lifesaving treatment for myocardial infarction is the placement of a stent in a closed or obstructed coronary artery. The largest modifiable risk factor after receiving a stent is medication adherence to Dual AntiPlatelet Therapy, a combination of P2Y12 inhibitors and aspirin. OBJECTIVE This study aimed to determine the acceptability of a protocol and an intervention using the My Interventional Drug-Eluting Stent Educational App (MyIDEA) and to evaluate medication adherence using the proportion of days covered (PDC) and platelet activation tests in a multisite randomized controlled trial. METHODS Potential participants who received a post percutaneous coronary intervention (PCI) procedure with a drug-eluting stent were approached. All patients older than 50 years and who spoke English were recruited. Participants were recruited, baseline demographics were collected, and the Hospital Anxiety and Depression Scale (HADS), Rapid Estimate of Adult Literacy in Medicine-Short Form, Burden-Benefit questionnaire, 36-Item Short Form Health Survey, and PCI knowledge questionnaire were administered. Block randomization was used to randomize participants to either usual care or MyIDEA supplementation. MyIDEA is a personalized educational intervention based on the Kolb experiential learning theory using patient narratives for education. During the visits, participants’ blood was collected to measure platelet suppression from medication. During the second and third encounters, the Morisky medication adherence score and cardiology outcomes were measured. The study was conducted at the University of Illinois Hospital and John H Stroger Jr Cook County Hospital with appropriate ethical approvals. Platelet suppression was measured through aspirin reactive units and P2Y12 reactive units. Medication adherence was measured using the PDC. The analysis team was blinded to the participants’ group membership. The primary outcome was a feasibility analysis of recruitment and retention. RESULTS The mean age of participants was 60.4 years (SD 7.1); the majority of patients were black and non-Hispanic. The majority of patients’ reading levels were seventh grade or above, and they were not very familiar with other electronic devices for information and communication. The number of control subjects was 21, and the number of participants in the interventional arm was 24. The interventional group was able to use MyIDEA in both the hospital and outpatient setting. However, there was no significant difference in platelet suppression or medication adherence between groups. There were also differences between the groups in terms of depression and anxiety, initially, as measured by HADS. No documented adverse event associated with the intervention was found. CONCLUSIONS Elderly patients are willing to use tablet devices to be educated about health conditions. Additional studies are required to measure the effectiveness and determine the most suitable timing and location for patient education.

Interest and use of alternative/holistic therapies amongst cancer patients at a large urban safety-net hospital.

e18289 Background: Patient interest in alternative/holistic therapies during cancer treatment is common in oncology. Studies have shown that 1 in 3 cancer patients in Europe turn to alternative therapies in combination with conventional medicine. Observational studies have shown some benefit of herbal medications in treating chemotherapy-associated complications; however, there remains skepticism in clinical practice. This study was undertaken to evaluate the interest and use of alternative treatments amongst cancer patients at a large urban safety-net hospital. Methods: An anonymous optional survey was offered in a random fashion to malignant hematology/oncology patients at Cook County Hospital in downtown Chicago. Results: One hundred seventy patients completed the survey comprised of 51% men, 78% being over age 50, and 67% with a high school or higher level of education. Responses included 35% African Americans, 34% Hispanics, and 25% Caucasians. At the time of the survey 16% of patients were currently using alternative treatments, while another 12% were planning to. Eighty-seven percent stated that they believed alternative treatments were very/somewhat beneficial, 77% expressed being very/somewhat interested in their use, and 77% stated they were very/somewhat likely to use such treatments. The treatments of interest included herbal supplements/oils in 40%, meditation/prayer in 29%, and cleanses/diets in 26%. Media and internet were the primary sources of information for 41% patients, while 36% heard about these treatments from family/friends. Although 51% of patients described having some safety concerns of such therapies, 77% had not discussed with their doctor. Conclusions: Our data showed that 28% of patients were either already taking or planning to take alternative/holistic therapies in conjunction with conventional medicine for their cancer care. Data also shows that an even larger number of patients are interested in such options and perceive them to be beneficial; however, there is a general reluctance to discuss it with the treating physician. The use of complementary therapies needs to be more openly discussed in order to better guide patients and assure safety.[Table: see text]

Real world outcomes of upfront docetaxel for hormone naïve metastatic prostate cancer in an ethnically diverse inner-city population.

359 Background: Upfront docetaxel (UD) in castration-sensitive metastatic prostate cancer (CSPC) has improved failure-free and overall survival in the CHAARTED, GETUG-AFU 15 and STAMPEDE trials, resulting in a paradigm shift in practice patterns. However, the impact of docetaxel-based chemotherapy regimens in minorities and in real-world practice remains to be described. The objective of this study is to evaluate tolerability and response to UD in an inner-city ethnically diverse CSPC cohort. Methods: We retrospectively reviewed clinical data for CSPC patients at Cook County Hospital. Patients treated with UD and androgen deprivation (UD, n = 49, 2013-17) were compared to those receiving androgen deprivation alone (ADT, n = 42, 2010-17) using descriptive statistics and Cox Proportional Hazards analysis. Results: Median age was 59 and 60 years in UD and ADT, respectively. African Americans, Hispanics and Caucasians formed 69%, 18%, and 10% in UD and 64%, 9.5% and 26% in ADT. Median PSA at diagnosis was: UD 536 [72-1110] ng/ml, ADT 229 [54-999] ng/ml (p = 0.3). Gleason score > 7 was present in 89.8% [UD] and 92.8% [ADT]. UD and ADT had similar frequency of bone (91% vs. 95%, p = 0.5), visceral (12.2% vs. 11.9%, p = 0.9) and retroperitoneal nodal metastases (55% vs. 45.2%, p = 0.3). UD was initiated a median of 8 weeks from diagnosis. 94% in UD received > 3 cycles of chemotherapy. CTCAE grade > 3 events included anemia (6%); neutropenia (6%); infection (4%); diarrhea (4%); peripheral neuropathy (6%), and fatigue (2%). Median PSA nadir was 3.2 [0.5-40.1] vs. 3.45 [0.1-36.4] ng/ml and time to nadir was 20 [13-27] vs. 41 [19-69] weeks in UD vs. ADT. Biochemical progression was observed in 34.7% [UD] vs. 40.5% [ADT]. Median time to castration-resistance was UD: 39 [31-67] vs. ADT: 65 [34-76] weeks [HR: 0.9; 95% CI 0.4-1.8; p = 0.7]. There were 12 deaths [UD: 4; ADT: 8] during a median follow-up of 46 [23-86, UD] and 111 [68-157, ADT] weeks. Conclusions: In a diverse underserved population with CSPC, UD was well tolerated but was not associated with improvement in time to castration compared to ADT. These results warrant validation and underscore the importance of ensuring accrual of minorities in clinical trials.

Regorafenib tolerance and outcomes in inner-city minority colorectal cancer population.

722 Background: Regorafenib (REG) is an oral multikinase inhibitor used for treatment of metastatic colon cancer after progression on fluorouracil, oxaliplatin and irinotecan therapy. The FDA approval was based on CORRECT trial showing improvements in progression-free survival (PFS) and overall survival (OS). Another trial (CONCUR) showed similar results. However, these trials included predominantly Whites and Asians. Our study aimed to evaluate treatment outcomes in the underserved, mainly African Americans and Hispanics. Methods: Cook County hospital is the 3rd largest public hospital in the US. All patients with an order for REG were identified from our pharmacy database. Charts were retrospectively reviewed for cancer stage, age, ethnicity, previous treatments, PFS and OS. Patients with non-colorectal cancer, incomplete data, no evidence of REG usage were excluded. Statistical analysis was done by t-test on subgroups. Results: A total of 42 patients were screened and 30 were included in the study. A comparison of outcomes with CORRECT and CONCUR trials is presented in Table 1. Our study patients, despite being less heavily pre-treated, had a worse OS compared to the clinical trials, with similar PFS. There was a trend towards better PFS and OS in women in our study of 4.3 and 4.8 months compared to men with 1.9 and 3.3 months respectively (p = NS). Six (20%) patients discontinued REG due to intolerance within a median of 1.2 months. The intolerant patients had a significantly higher median BMI of 31 compared to 24.9 for patients who continued the medication to progression (p = 0.019). Conclusions: Compared to CORRECT and CONCUR trials, population of predominantly AA and Hispanics showed a trend towards worse OS despite being less heavily pre-treated indicating worse outcomes in minorities treated with REG. Women showed better PFS and OS compared to men in our study. Patients unable to tolerate REG had significantly higher BMI, indicating a need to study dosing in this subgroup. [Table: see text]

Distribution of Cytogenetic Abnormalities in African American Multiple Myeloma Patients May be Unique for Different Geographic Regions

BACKGROUND: Multiple myeloma (MM) is the most common hematologic malignancy in the African American population, with an incidence more than 2 times higher than Caucasian population [Landgren O et al Blood 2006]. Historically, the African American MM patients have had better outcomes compared with other races [Ailawadhi S et al Br J Haematol 2012], but no biologic explanations exist for this observation. Greenberg et al [Blood Cancer J 2015] have recently reported on differences in commonly observed baseline cytogenetic abnormalities (CA) between African American and Caucasian MM patients seen at Mayo Clinic (Rochester, MN), Cook County Hospital (Chicago, IL) and University of Maryland (Baltimore, MD). We examined the MM cohort at our referral center to validate these observations. PATIENTS & METHODS: The Levine Cancer Institute MM database was interrogated for all patients presenting with MM between January 2012 and April 2015. Baseline clinical and pathology variables were compared between the African American and Caucasian cohorts. Continuous variables were compared using nonparametric rank tests, while incidences and proportions (e.g. CAs including t(11;14), t(4;14), monosomy13/del13q and del17p) were compared using Fisher's exact tests. RESULTS: A total of 662 patients were identified; excluding those with MGUS classification, 368 patients were included in the analysis (African Americans n = 130, Caucasian n = 238). The median age of African American MM patients was significantly younger than Caucasian MM patients (median age 60 years vs. 65 years, p=0.010), with similar gender distribution. There was a numerically larger proportion of African American patients with anemia (40.8% vs 30.8%, p =0.166), however, there was no significant difference in degree of BM plasmacytosis amongst the two groups. The overall distribution of MM patients by IMWG risk stratification (Chng et al, Leukemia 2013) was also similar between the two groups. The African American MM patients had a numerically higher incidence of a metaphase abnormality on conventional cytogenetics (21.7% vs. 13.9%, p =0.154). They had a significantly lower incidence of t(11;14) [7.7% vs. 16%, p=0.024], a numerically higher incidence of t(4;14) [6.2% vs. 3.8%, p=0.309], and similar incidence of deletion 13/del13q [22.3% vs. 18.9%, p=NS ] and del17p [7.7% vs. 7.6%, p=NS ]. CONCLUSIONS: The present dataset is the largest single institution report on CA racial differences in MM patients. We found that unlike previous reports of lower incidence of t(4;14) or del17 p in African American MM patients by Greenberg et al, we have observed a higher incidence of t(4;14) and similar incidence of del17p in our experience compared to Caucasian MM patients. The different pattern of CA distribution compared to published literature may represent geographic heterogeneity and potentially influence survival outcomes. Disclosures Cogdill: Celgene: Speakers Bureau; Onyx: Speakers Bureau; Millennium: Speakers Bureau; Novartis: Speakers Bureau. Usmani:Celgene: Honoraria, Speakers Bureau; Onyx: Honoraria, Research Funding, Speakers Bureau; Janssen Oncology: Honoraria, Research Funding; Sanofi: Honoraria, Research Funding; Pharmacyclics: Research Funding; Millennium: Honoraria, Speakers Bureau; Array BioPharma: Honoraria, Research Funding.

Concurrent chemoradiotherapy with 5-fluorouracil and mitomycin-C for invasive anal carcinoma in HIV-positive patients receiving highly active anti-retroviral therapy versus non-HIV patients.

e14590 Background: In the non-HIV population, radiotherapy (RT), fluorouracil (5FU) with mitomycin (MMC) has become the standard in the non-metastatic setting for anal carcinomas (AC). To date, most studies with AC in HIV patients (pts) are small case series where multiple chemoradiation (CRT) regimens were used and analyzed as one cohort. In addition, little data exists on the inner city HIV population. Cook County Hospital (CCH) is the largest health provider for HIV pts in Chicago and together with its outpatient clinic, the Ruth M. Rothstein CORE Center (CC) 5,500 HIV+ pts are treated per year. The County Hospital (CCH) AIDS Malignancy Project (CHAMP Study) is a retrospective study of all HIV cancer pts treated for the past 14 years. Methods: We identified all HIV+ pts with invasive AC in CHAMP cohort. We analyzed HIV characteristics, overall survival (OS), PFS and pt demographics and compared it to a HIV- cohort from the same institution. All AC treated without MMC/5FU/RT were excluded. Statistics: Time to local recurrence (TLR), time to distant metastasis (TDM), and OS data was analyzed using Kaplan-Meier analysis and a Cox Proportional Hazards model. Results: 35 HIV + and 52 HIV - pts were included. Of the HIV+ vs. HIV- pts, 89 vs. 52% were male, 82 vs. 48% were AA and the average age in HIV+ vs. HIV- was 44 vs. 52 yrs. 45 % of the HIV pts presented with stage IIIA or IIIB disease vs. 46% in HIV- pts. 15 % HIV- pts had stage IV vs. 0% HIV +. The median survival in the HIV+ vs. HIV- was 34 vs. 39 mo (p>0.5). In the HIV- population, 22% survived 120 months, while no HIV pt survived over 90 months. TLR was 20 months shorter in the HIV+ arm (p<0.5). OS based on CD4 count did not differ. Conclusions: HIV associated AC is an AA male disease compared to HIV- pts in the inner city. More stage IV disease was reported in the HIV- cohort, but the median survival was equal with no long-term survivors in the HIV+ arm, possibly due to TLR, which was 20 months shorter, implying more aggressive disease. Both inner city groups present late but tolerate chemotherapy equally well. Education is needed in both HIV+ and HIV- pts to diagnose the cancers early so the OS can match the national average.

Geographic variation of extranodal sites of mucosal associated lymphoid tissue lymphoma (MALToma): Analysis of series from two institutions.

e18534 Background: MALToma represents about 5% of Non-Hodgkin lymphomas (NHL). Prior studies report stomach to be the most common site of involvement (35-50%), while skin (16%) or salivary glands (26%) were reported to be the 2nd most frequent site in 2 different studies. Geographical variation may affect the proportions of primary sites of MALToma, which could provide further clues to investigating their causes. Methods: The data was obtained on 91 consecutive patients from the University of Arizona Cancer Center (UACC) clinical and pathological databases, and 25 patients from the Cook County Hospital (CCH) database, seen between 2005 and 2011.The frequency of clinical characteristics and their relation to survival were analyzed and compared between UACC and CCH. Results: UACC and CCH cohorts were balanced for gender (M:F ratio 0.65) and age (median 63 y (range, 26-88), but not for race or/and ethnicity. While at CCH 36% of pts were African-American (AA), 28% Hispanic (any race), 12% Caucasian and 20% unknown; at UACC 88% of pts were Caucasian, 10% Hispanic and 1% AA (p<0.0001). UACC had 8% pts with > 1 extranodal site, while CCH had none. The primary sites of MALToma were also different. At CCH, similarly to what’s commonly reported in the literature, 40% of pts had primary gastric involvement, 20% salivary gland, 12% lung, 8% orbit, 8% thyroid, 4% intestine, 4% skull base tumor, and 4% breast. At UACC, however, 22% of pts had primary skin involvement, 19% gastric, 16% lung, 13% orbit, 8% salivary gland, 6% intestine, 5% breast, 3% thyroid and 8% other. The difference was statistically significant for the frequency of primary skin (p=0.006) but not gastric involvement (p=0.084). 80% of patients with primary skin involvement were Caucasian. The median follow-up of the survivors was 18 months for both groups, with no statistically significant difference in median progression-free (7.5 years) or overall survival (14.6 years). Conclusions: There was higher prevalence of Caucasian race and skin involvement at UACC than at CCH. We speculate that sun-induced photodamage causes chronic inflammation contributing to eventual malignant transformation to primary skin MALToma.