Gursel, Demirkan

Abstract Hepatocellular carcinoma (HCC) is a leading cause of cancer mortality globally, with particularly high burdens among people living with HIV (PLWH) in low‐resource settings like Nigeria. Effective early detection remains a major challenge due to limited access to imaging‐based surveillance and the low sensitivity of current biomarkers such as alpha‐fetoprotein (AFP). We conducted an epigenome‐wide association study (EWAS) of circulating cell‐free DNA (ccfDNA) methylation in a Nigerian cohort of HIV‐positive individuals ( n = 245), spanning HCC, cirrhosis (FibroScan ≥12.3 kPa), fibrosis (FibroScan ≥7.6 and <12.3 kPa), and HCC‐free without fibrosis (FibroScan <7.6 kPa) groups. Using random forest modeling, we developed and evaluated a ccfDNA methylation classifier (ccfDNAmRF) for HCC risk prediction. We identified 73 CpG sites significantly associated with HCC (false discovery rate <0.01). The ccfDNAmRF model demonstrated strong discriminatory power, achieving 100% sensitivity and 80%–91% specificity for distinguishing HCC from cirrhosis, fibrosis, and HCC‐free groups (area under the curve [AUC]: 92%–97%). Combining ccfDNA methylation risk scores with AFP further improved classification accuracy (AUC up to 98.5%). Notably, ccfDNA methylation patterns displayed clear dose–response relationships across the disease spectrum, supporting their utility for early‐stage detection and risk stratification. Our findings highlight the promise of ccfDNA methylation biomarkers as a non‐invasive, blood‐based screening tool for improving early identification of HCC cases among PLWH.

Importance Oral tamoxifen citrate benefits women with ductal carcinoma in situ (DCIS), but concern about toxic effects has limited acceptance. Previous pilot studies have suggested transdermal 4-hydroxytamoxifen gel has equivalent antiproliferative efficacy to oral tamoxifen, with low systemic exposure. Objective To demonstrate that 4-hydroxytamoxifen gel applied to the breast skin is noninferior to oral tamoxifen in its antiproliferative effect in DCIS lesions. Design, Setting, and Participants This randomized, double-blind, phase 2 preoperative window trial was performed at multicenter breast surgery referral practices from May 31, 2017, to January 27, 2021. Among 408 women with estrogen receptor–positive DCIS who were approached, 120 consented and 100 initiated study treatment. The most common reasons for nonparticipation were surgical delay, disinterest in research, and concerns about toxic effects. Data were analyzed from January 26, 2021, to October 5, 2022. Intervention Random assignment to oral tamoxifen citrate, 20 mg/d, and gel placebo or 4-hydroxytamoxifen gel, 2 mg/d per breast, and oral placebo, for 4 to 10 weeks, followed by DCIS resection. Main Outcomes and Measures The primary end point was absolute change in DCIS Ki-67 labeling index (Ki67-LI). Secondary end points included 12-gene DCIS Score, breast tissue tamoxifen metabolite concentrations, tamoxifen-responsive plasma protein levels, and patient-reported symptoms. Noninferiority of Ki67-LI reduction by 4-hydroxytamoxifen gel was tested using analysis of covariance; within- and between-arm comparisons were performed with paired t tests for mean values or the Wilcoxon rank sum test for medians. Results Of 90 participants completing treatment (mean [SD] age, 55 [11] years; 8 [8.9%] Asian, 16 [17.8%] Black, 8 [8.9%] Latina, and 53 [58.9%] White), 15 lacked residual DCIS in the surgical sample, leaving 75 evaluable for the primary end point analysis (40 in the oral tamoxifen group and 35 in the 4-hydroxytamoxifen gel group). Posttreatment Ki67-LI was 3.3% higher (80% CI, 2.1%-4.6%) in the 4-hydroxytamoxifen gel group compared with the oral tamoxifen group, exceeding the noninferiority margin (2.6%). The DCIS Score decreased more with oral tamoxifen treatment (−16 [95% CI, −22 to −9.4]) than with 4-hydroxytamoxifen gel (−1.8 [95% CI, −5.8 to 2.3]). The median 4-hydroxytamoxifen concentrations deep in the breast were nonsignificantly higher in the oral tamoxifen group (5.7 [IQR, 4.0-7.9] vs 3.8 [IQR, 1.3-7.9] ng/g), whereas endoxifen was abundant in the oral tamoxifen group and minimal in the 4-hydroxytamoxifen gel group (median, 13.0 [IQR, 8.9-20.6] vs 0.3 [IQR, 0-0.3] ng/g; P &amp;lt; .001). Oral tamoxifen caused expected adverse changes in plasma protein levels and vasomotor symptoms, with minimal changes in the transdermal group. Conclusions and Relevance In this randomized clinical trial, antiproliferative noninferiority of 4-hydroxytamoxifen gel to oral tamoxifen was not confirmed, potentially owing to endoxifen exposure differences. New transdermal approaches must deliver higher drug quantities and/or include the most potent metabolites. Trial Registration ClinicalTrials.gov Identifier: NCT02993159

Abstract Background: Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related deaths worldwide. In countries with high human immunodeficiency virus (HIV) prevalence, such as Nigeria, HIV-associated HCC causes a great health burden due to its early onset, late diagnosis, and poorer prognosis. HIV infection is involved in inflammation of the liver, which may determine an increased risk of hepatocyte neoplastic transformation. Inflammation-related DNA methylation signatures obtained in liquid biopsy, such as circulating cell-free DNA (ccfDNA) extracted from serum/plasma are promising minimally-invasive biomarkers that may inform HCC among people living with HIV. Methods: A total of 289 Nigerian participants with information on ccfDNA and other covariates were included. Participants were classified into three groups by their HCC/HIV status: 1) HCC+/HIV+ (n=28); 2) HCC-/HIV+ (n=185); and 3) HCC+/HIV- (n=76). We constructed ccfDNA methylation inflammation scores using 49 CpGs previously linked to circulating C-reactive protein (CRP) concentrations, and higher scores represented elevated CRP concentrations with lower methylation levels. To compare the ccfDNA methylation inflammation score across the groups, we performed multivariable logistic regression analyses adjusting for age, sex, alcohol consumption, smoking status, body mass index, study sites, and technical variables. Results: Participants with HCC+/HIV+ presented the highest ccfDNA methylation inflammation scores (mean=-0.03, standard deviation [SD] =0.01). Participants with HCC-/HIV+ presented the lowest scores (mean=-0.05, SD=0.01). In the multivariable logistic regressions, one SD increase of inflammation score was associated with 2.5 times higher odds of having HCC among HIV-infected participants (HCC+/HIV+ vs. HCC+/HIV-; OR=2.56, 95% CI=1.39-4.73, P=0.002). No evidence was found for the association between the inflammation score and HIV status among HCC patients (HCC+/HIV+ vs. HCC+/HIV-; OR=1.00, 95% CI=0.59-1.70, P=0.991). In the secondary analysis comparing HCC+ vs. HCC- adjusting for HIV status, one SD increase of inflammation score was associated with 3.2 times higher odds of having HCC (OR=3.22, 95% CI=1.39-7.46, P=0.006). Conclusions: We observed that ccfDNA methylation inflammation score is associated with HCC status among people with HIV. Our findings suggest that ccfDNA methylation-based inflammatory profiles may serve as a potential biomarker for early detection and risk stratification of HCC in resource-constrained countries with a high prevalence of HIV. Citation Format: Kyeezu Kim, Yinan Zheng, Claudia Hawkins, Edith Okeke, Olufunmilayo Lesi, Yishu Qu, Lewis R. Roberts, Demirkan Gursel, Fatimah B. Abdulkareem, Alani Akanmu, Godwin Imade, Jian-Jun Wei, Masha Kocherginsky, Kwang-Youn Kim, Wasiu L. Adeyemo, Firas Wehbe, Chad J. Achenbach, Atiene Sagay, Folasade T. Ogunsola, Robert L. Murphy, Lifang Hou. DNA methylation-based inflammation score is associated with hepatocellular carcinoma among people living with HIV. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6013.

Abstract Background: Adjuvant oral tamoxifen (TAM) benefits women with DCIS, but toxicity concerns have limited its acceptance. Transdermal therapy with 4-hydroxy tamoxifen (4-OHT) gel applied to the breast skin is a possible solution. Previous pilot data suggest equivalent anti-proliferative efficacy of TAM and 4-OHT gel, but minimal systemic exposure with transdermal therapy. We report a prospective double blinded randomized phase 2 trial comparing TAM to 4-OHT gel in women with DCIS. Methods: 107 women with estrogen receptor positive (≥10%) DCIS were randomized to TAM (20 mg/day + placebo gel) or 4-OHT gel (2mg 4-OHT gel/breast, bilaterally + oral placebo), for 4-10 weeks prior to surgery. The primary endpoint was reduction in DCIS Ki67 labeling index (LI). Secondary endpoints included the 12-gene DCIS Score assay (Exact Sciences), breast tissue and plasma concentrations of 4-OHT and endoxifen, TAM-responsive circulating proteins, and patient reported symptoms (Breast Eight Symptom Scale). We estimated that 80 evaluable participants would provide 80.5% power to establish non-inferiority of 4-OHT, defined as relative Ki67-LI decline &gt;35% and absolute decline &gt;2.6%, with one-sided 𝛼=0.10. Non-inferiority of 4-OHT gel for Ki67-LI reduction was tested using an ANCOVA model. Statistical comparisons within- and between-arms were calculated with paired t-test and Welch Two Sample t-test, respectively. Results: 72 of 87 women adhered to the protocol, and were evaluable for the primary endpoint (39 TAM and 33 4-OHT gel). Mean treatment duration was 47 days for TAM and 44 days for 4-OHT gel (p=0.2). The median absolute decline in Ki67 labeling index was significant in the oral TAM (-3.7%, p&lt; 0.001) but not in 4-OHT gel arm (-1.3%, p=0.2) (p=0.002). Ki67 results following menopausal stratification also favored the TAM arm: (-1.3%; p=0.06 in 37 premenopausal women and -3.7%; p=0.02 in 35 postmenopausal women). Similarly, DCIS score showed a significantly greater reduction in the TAM (-14, p&lt; 0.001) but not in the 4-OHT gel arm (-4, p=0.1). Tissue 4-OHT concentrations were non-significantly higher in the TAM arm and were similar between superficial and deep sampling locations (superficial 6.1 and 4.2 ng/g for TAM and 4-OHT gel, respectively, p= 0.55; deep 5.7 and 3.8 ng/g, respectively, p= 0.06), whereas plasma 4-OHT concentration was markedly lower in the gel group (2 ng/mL and 0.24 ng/mL for TAM and 4-OHT gel, respectively, P &lt; 0.001). Endoxifen was abundant in plasma (11 ng/mL) and deep tissue (13 ng/g) of the TAM arm, but present in trace amounts in the 4-OHT gel arm (undetectable in plasma and 0.31 ng/g in tissue; p &lt; 0.001). Circulating TAM responsive markers (insulin like growth factor 1, sex hormone binding globulin, von Willebrand factor, and protein S total) and vasomotor symptoms were significantly and unfavorably modulated by TAM, but not by 4-OHT gel therapy. Conclusions: The non-inferiority of transdermal 4-OHT gel to Tam in terms of anti-proliferative effect in DCIS lesions was not demonstrated at the doses used for this study. DCIS Score analysis gave similar results. Tissue 4-OHT concentration in 4-OHT gel and Tam-treated subjects was roughly similar. However, endoxifen exposure was higher with oral TAM therapy and may partially explain the observed differences in major endpoints. In future studies, use of higher 4-OHT gel doses, longer duration of treatment, or different formulation may overcome these. Citation Format: Oukseub Lee, Xinlei Mi, Yanfei Xu, Luis Blanco, Azza M. Akasha, Kelly Benante, Shanshan Zhang, Carissa LaBoy, Thomas Helland, Melissa Pilewskie, Amy Degnim, Zahraa Al-Hilli, Amanda L. Amin, E Shelley Hwang, Joseph M. Guenther, Simon Steinar Hustad, Demirkan B. Gursel, Masha Kocherginsky, Gunnar Mellgren, Eileen Dimond, Marjorie Perloff, Brandy M. Heckman-Stoddard, Seema Khan. PD15-12 A pre-surgical window trial of oral tamoxifen versus transdermal 4-hydroxytamoxifen gel in women with estrogen receptor positive duct carcinoma in situ (DCIS) [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD15-12.

Abstract Purpose: Hepatocellular Carcinoma (HCC) is the leading cause of cancer-related mortality in Nigeria. HIV and viral hepatitis B (HBV) and C (HCV) co-infection are common in Nigeria that significantly accelerates liver disease progression including HCC. Aging-related DNA methylation signatures obtained in liquid biopsy, such as circulating cell-free DNA (ccfDNA) extracted from serum/plasma are promising minimally-invasive biomarkers that may inform HIV-associated HCC. We examined the epigenetic age acceleration (EpiAgeAccel) in ccfDNA in HCC patients with HIV. Methods: The study included three groups of participants: a) HIV positive with HCC (n=7); b) HIV positive and cancer-free (n=45); and c) HIV negative with HCC (n=33). Epigenetic age was estimated by Horvath's calculator using genome-wide ccfDNA methylation data profiled by Illumina EPIC array. EpiAgeAccel was computed as the residuals of a linear model of epigenetic age on chronological age, namely the unexplained portion of epigenetic age by chronological age. We used multiple linear regression to compare EpiAgeAccel between HIV/HCC groups, adjusting for sex, age, education, alcohol intake, and HBV/HCV infection. Results: Among HIV positive participants, there was a higher percentage of men (57% vs. 22%, p&lt;0.001), HBV infection (29% vs. 11%, p=0.004), and HCV infection (57% vs. 2%, p&lt;0.001) in the HCC group compared to the cancer-free group. EpiAgeAccel was 4.8 years higher in HIV positive patients with HCC compared to cancer-free HIV positives (p=0.02). Among HCC patients, EpiAgeAccel was 2.1 years higher in HIV positives compared to HIV negative but not statistically significant. Conclusion: Epigenetic age in ccfDNA is accelerated in HIV-positive HCC patients. EpiAgeAccel measured in ccfDNA may be developed into a surrogate biomarker for minimally invasive HCC detection among HIV-infected patients in low- and middle-income countries. Citation Format: Yinan Zheng, Claudia Hawkins, Edith Okeke, Mary Duguru, Emuobor Odeghe, Olufunmilayo Lesi, Yishu Qu, Kyeezu Kim, Brian Joyce, Jun Wang, Lewis Roberts, Demirkan Gursel, Abdulkareem Fatimah, Alani Akanmu, Godwin Imade, Jian-Jun Wei, Masha Kocherginsky, Kwang-Youn Kim, Oluwakemi Odukoya, Wasiu Adeyemo, Firas Wehbe, Chad Achenbach, Atiene Sagay, Folasade Ogunsola, Robert Murphy, Lifang Hou. Blood-Based Circulating Cell-Free DNA Epigenetic Age Is Accelerated Among HIV-Infected Patients with Hepatocellular Carcinoma in Nigeria [abstract]. In: Proceedings of the 9th Annual Symposium on Global Cancer Research; Global Cancer Research and Control: Looking Back and Charting a Path Forward; 2021 Mar 10-11. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2021;30(7 Suppl):Abstract nr 83.

e16137 Background: Hepatocellular Carcinoma (HCC) is the leading cause of cancer-related mortality in Nigeria. HIV and viral hepatitis B (HBV) and C (HCV) co-infection are common in Nigeria that significantly accelerates liver disease progression including HCC. Aging-related DNA methylation signatures obtained in liquid biopsy, such as circulating cell-free DNA (ccfDNA) extracted from serum/plasma are promising minimally-invasive biomarkers that may inform HIV-associated HCC. We examined the epigenetic age acceleration (EpiAgeAccel) in ccfDNA in HCC patients with HIV. Methods: The study included three groups of participants: a) HIV positive with HCC (n=7); b) HIV positive and cancer-free (n=45); and c) HIV negative with HCC (n=33). Epigenetic age was estimated by Horvath’s calculator using genome-wide ccfDNA methylation data profiled by Illumina EPIC array. EpiAgeAccel was computed as the residuals of a linear model of epigenetic age on chronological age, namely the unexplained portion of epigenetic age by chronological age. We used multiple linear regression to compare EpiAgeAccel between HIV/HCC groups, adjusting for sex, age, education, alcohol intake, and HBV/HCV infection. Results: Among HIV positive participants, there was a higher percentage of men (57% vs. 22%, p<0.001), HBV infection (29% vs. 11%, p=0.004), and HCV infection (57% vs. 2%, p<0.001) in the HCC group compared to the cancer-free group. EpiAgeAccel was 4.8 years higher in HIV positive patients with HCC compared to cancer-free HIV positives (p=0.02). Among HCC patients, EpiAgeAccel was 2.1 years higher in HIV positives compared to HIV negative but not statistically significant. Conclusions: Epigenetic age in ccfDNA is accelerated in HIV-positive HCC patients. EpiAgeAccel measured in ccfDNA may be developed into a surrogate biomarker for minimally invasive HCC detection among HIV-infected patients in low- and middle-income countries.