Bio


Stefan Thottunkal is a physician in training, Stanford researcher, and policymaker whose work sits at the intersection of artificial intelligence, precision medicine, translational science, and public health innovation. He completed the M.S. in Community Health and Prevention Research at Stanford University as an IIE Quad Fellow, one of the world’s most selective international research fellowships, where his thesis centered on computational pharmacogenomics and the use of data-driven LLM methods to advance precision prescribing.

His research focuses on translating innovation into clinically meaningful and implementation-ready health solutions, with particular interests in pharmacogenomics, chronic disease, and AI-enabled decision support. He is especially interested in how machine learning and large language models can be used not simply as technical advances, but as robust clinical tools that improve prescribing, strengthen care delivery, and incorporate human centered design principles to effectively integrate precision medicine in routine clinical practice.

At Stanford, he contributes to the Han Lab’s research on advancing precision oncology in advanced non-small cell lung cancer, while helping lead NOURISH, a pioneering Stanford Medicine initiative reimagining cardiometabolic care through culturally tailored nutrition science, behavioral insight, and digital innovation. NOURISH advances a model of lifestyle medicine that preserves cultural relevance while applying rigorous scientific methods to improve metabolic health. By integrating culinary medicine with emerging technologies, the initiative is exploring how AI-enabled tools, personalized digital education, and interactive nutrition support systems can make evidence-based dietary guidance more adaptive, engaging, and scalable across diverse populations. His work in this space reflects a broader interest in how technology can help transform nutrition care from generic advice into a more personalized, culturally tailored, and behaviorally attuned form of preventive medicine.

In parallel with his research career, Stefan brings close to half a decade of experience advising the Australian Federal Government on major health and social policy initiatives. His international experience also includes mentoring hackathon teams in India and medical device development in Nigeria, where he contributed to dialysis device innovation and clinical trials design in resource-constrained settings. Together, these experiences reflect his broader commitment to advancing equitable, evidence-based, and culturally tailored global health innovation.

All Publications


  • Evaluating Tumor Burden as a Predictive Biomarker for Epidermal Growth Factor Receptor Targeted Kinase Inhibitor Therapy in Advanced Non-Small Cell Lung Cancer. JCO precision oncology Terashima, R., Fan, J., Gunturkun, F., Nieda, G., Fan, X., Rodriguez, E. M., Tan, A. X., Thottunkal, S., Shaw, M., Su, C. C., Khan, A., Ding, V. Y., Luo, I., Satoyoshi, M., Bhat, A., Gu, B., Henry, S. M., Ellis-Caleo, T. J., Odden, M., Kurian, A. W., Neal, J. W., Wakelee, H. A., Wu, J. T., Han, S. S. 2026; 10 (4): e2500884

    Abstract

    As treatment options for advanced non-small cell lung cancer (NSCLC) evolve, biomarkers are needed to guide therapy selection while balancing efficacy and toxicity. Although tumor burden is a promising candidate, its prognostic role in guiding epidermal growth factor receptor (EGFR)-targeted kinase inhibitor (TKI) therapies remains understudied in real-world settings.We identified patients with de novo stage IV EGFR-mutant NSCLC treated with first-line EGFR-TKI at Stanford Health Care (2000-2021). Tumor burden metrics were manually annotated from 592 baseline radiology reports, encompassing size, number, and location (1,807 lesions). Multivariable Cox regression evaluated associations between tumor burden metric and overall survival (OS), adjusting for confounders, in the overall cohort and an osimertinib subgroup. A weighted composite tumor burden score was constructed using statistically significant metrics to stratify risk.Of 312 patients, bone metastasis (hazard ratio (HR)adjusted, 1.64 [95% CI, 1.23 to 2.19]) and the number of metastatic organs (HRadjusted, 1.21 [95% CI, 1.10 to 1.32]) were independently associated with worse OS and used to construct the composite score. Patients with low tumor burden (composite-score ≤ median 1.06) experienced better OS than those with high tumor burden, with a 3-year OS of 59.8% versus 41.5% (P = .001). Consistent findings were observed in the osimertinib subgroup, with a 3-year OS of 62.2% versus 44.6% (P = .03) for low versus high tumor burden.Tumor burden may serve as a prognostic biomarker in advanced NSCLC receiving EGFR-TKIs. These findings raise the hypothesis that durable survival in low-burden patients may be achievable with monotherapy, potentially sparing unnecessary toxicity from combination regimens. This warrants prospective validation comparing monotherapy versus combination strategies.

    View details for DOI 10.1200/PO-25-00884

    View details for PubMedID 41955549

  • Clinician Experiences at the Frontier of Pharmacogenomics and Future Directions. Journal of personalized medicine Thottunkal, S., Spahn, C., Wang, B., Rohatgi, N., Hong, J., Khandelwal, A., Palaniappan, L. 2025; 15 (7)

    Abstract

    Pharmacogenomics (PGx) has emerged as a powerful tool to personalize drug selection and dosing based on a patient's genetic profile. However, there are a range of challenges that impede uptake in current clinical practice. For example, clinicians often express frustration with commercially available PGx panel tests, which fail to consistently include all key actionable PGx genes (according to the Clinical Pharmacogenetics Implementation Consortium (CPIC), Food and Drug Administration (FDA) PGx guidelines, or The Dutch Pharmacogenetics Working Group (DPWG) guidelines) and instead are too long with clinically unimportant information (unvalidated genotypes). Additionally, the lack of EMR integration, clinician education and awareness of the benefits of PGx impedes uptake. This paper examines key challenges identified in clinical practice and proposes future directions, focusing on limiting PGx reports to essential data, providing point-of-prescription alerts, and establishing reimbursement pathways that encourage adoption. Future directions include leveraging large language models, integrating point-of-prescription alerts and phenoconversion calculators into the electronic medical record, increasing the genomic diversity of PGx study populations, and streamlining coverage by payers.

    View details for DOI 10.3390/jpm15070294

    View details for PubMedID 40710411

  • What influences the implementation of health checks in the prevention and early detection of chronic diseases among Aboriginal and Torres Strait Islander people in Australian primary health care? Findings from an evidence mapping review. Health research policy and systems Yadav, U. N., Thottunkal, S., Agostino, J., Sinka, V., Wyber, R., Hammond, B., Butler, D. C., Belfrage, M., Freeman, K., Passey, M., Walke, E., Smith, M., Jones, B., Lovett, R., Douglas, K. A. 2025; 23 (1): 70

    Abstract

    Chronic disease is the leading cause of morbidity and mortality among Aboriginal and Torres Strait Islander peoples in Australia. A comprehensive health assessment is available as an annual health check (HC) to Aboriginal and Torres Strait Islander peoples through the Medicare Benefits Schedule in primary health care settings. This review aims to systematically identify contextual and mechanistic factors that contribute to the success or failure of implementing effective HCs in the prevention and early detection of chronic diseases among Aboriginal and Torres Strait Islander people in Australian primary health care (PHC).We systematically searched for peer-reviewed and grey literature, including policy reports, theses, and guidelines, between 1 November 1999 and 30 June 2023, using a combination of keywords and subject headings related to "health checks", "chronic disease", and "Aboriginal and Torres Islander peoples" in seven databases. The extracted data were summarized using a content analysis approach, applying strength-based approaches.In total, 16 peer-reviewed articles and five grey literature that met the inclusion criteria were used for evidence synthesis that identified several contextual and mechanistic factors that influenced the implementation of HCs. Barriers included resource constraints driven by complexities in administrative, workforce and policy domains that significantly impeded the implementation of HCs. Within PHC, physical space constraints, competing demands and a focus on acute care over preventive measures hindered HC implementation. In addition, inconsistent identification of Aboriginal and Torres Strait Islander status, negative attitudes of PHC staff towards HC efficacy and patients' fear of stigma or confidentiality breaches were barriers. Patients reported HCs as failing to address holistic health needs. To improve HC implementation, enablers included strong clinical leadership, recruitment of culturally competent non-Indigenous and Aboriginal and Torres Strait Islander staff, Indigenous partnership and community engagement and incentives for participation. Effective electronic records, transport provision and flexible scheduling also increased accessibility.Our findings suggest that future implementation research must adopt a more comprehensive and holistic approach across different models of PHC, with clearly identified contextual and mechanistic factors linked to people-reported and service outcomes, to guide the implementation and evaluation of HCs. While undertaking future research, it is crucial to implement policy and practice reforms as identified in this review to create a culturally safe service at the PHC level required to drive the uptake of quality HCs that aligns with community priorities and aspirations for the prevention and early detection of chronic diseases.

    View details for DOI 10.1186/s12961-025-01325-9

    View details for PubMedID 40426191

    View details for PubMedCentralID PMC12107981

  • Clinician Experiences at the Frontier of Pharmacogenomics and Future Directions Journal of Personalized Medicine Thottunkal, S., Spahn, C., Wang, B., Rohatgi, N., Hong, J., Khandelwal, A., Palaniappan, L. 2025; 15 (7)

    View details for DOI 10.3390/jpm15070294

  • Outbreak response capacity of the Global Outbreak Alert and Response Network across WHO's South-East Asia and Western Pacific regions. Western Pacific surveillance and response journal : WPSAR Parry, A. E., Campbell, S., Thottunkal, S., Mandal, P. P., Salmon, S. 2024; 15 (5 Spec edition): 1-7

    Abstract

    The COVID-19 pandemic challenged the Global Outbreak Alert and Response Network's (GOARN) mechanism used to rapidly deploy technical support for international responses and highlighted areas that require strengthened capacity within the Network. GOARN's partners in the World Health Organization's (WHO) South-East Asia and Western Pacific regions were engaged to explore their levels of preparedness, readiness and ability to respond to international public health emergencies.Consultative discussions were held and a survey was conducted with GOARN's partners from the two WHO regions. Discussion topics included partners' capacity to support and participate in a GOARN deployment, training, research and collaboration. Descriptive and content analyses were conducted.Barriers to engaging in GOARN's international outbreak response efforts included limited numbers of personnel trained to respond to outbreaks; institutional, financial and administrative hurdles; and limited collaboration opportunities. Partners identified innovative solutions that could strengthen their engagement with deployment, such as financial subsidies, mentorship for less experienced staff, and the ability to provide remote support.GOARN plays an important role in enabling WHO to fulfil its international alert and response duties during disease outbreaks and humanitarian crises that have the potential to spark disease outbreaks. Yet without systematic improvement to strengthen national outbreak capacity and regional connectedness, support for international outbreak responses may remain limited. Thus, it is necessary to integrate novel approaches to support international deployments, as identified in this study.

    View details for DOI 10.5365/wpsar.2024.15.5.1109

    View details for PubMedID 39171201

    View details for PubMedCentralID PMC11335431

  • Prevalence of extra-pulmonary tuberculosis in Africa: A systematic review and meta-analysis. Tropical medicine & international health : TM & IH Hailu, S., Hurst, C., Cyphers, G., Thottunkal, S., Harley, D., Viney, K., Irwin, A., Dean, J., Nourse, C. 2024; 29 (4): 257-265

    Abstract

    The burden of extra-pulmonary tuberculosis (EPTB) is not well quantified in TB endemic countries such as those in sub-Saharan Africa. This study aimed to quantify that burden via a systematic review of the prevalence of EPTB in African countries.Studies were retrieved by searching five databases; 105 studies published between 1990 and 2023 were included. The studies described the prevalence of EPTB among the general population (4 studies), TB patients (68) and patients with other conditions, including HIV (15), meningitis (3), renal failure (3) and other comorbidities, some of which are cancer (12). Due to the low number of studies reporting EPTB in patients with conditions other than TB, the meta-analysis was performed on studies reporting on EPTB among TB patients (68 studies). Meta-analysis was performed on the 68 studies (271,073 participants) using a random-effects model to estimate the pooled prevalence of EPTB. Meta-regression was used to explore possible explanations for heterogeneity according to regions and time periods.The pooled prevalence of EPTB among TB patients was 26% (95% CI 23-29%). There was substantial heterogeneity of prevalence for the five African regions. The Eastern region had the highest prevalence of 32% (95% CI 28-37%) and the lowest in Western Africa, 16% (95% CI 10-24%). There was no significant difference in the prevalence of EPTB between the 3 eleven-year time periods.Our systematic review and meta-analysis give insight into the burden of EPTB in Africa. This review could inform clinical and programmatic practices-a higher suspicion index for clinicians and more effort for better services. This could contribute to efforts aiming to end TB, which have historically been focused on PTB.Coordinated efforts that target both EPTB and PTB are needed.

    View details for DOI 10.1111/tmi.13970

    View details for PubMedID 38263374

  • Pain and its interference with daily living in relation to cancer: a comparative population-based study of 16,053 cancer survivors and 106,345 people without cancer BMC CANCER Joshy, G., Khalatbari-Soltani, S., Soga, K., Butow, P., Laidsaar-Powell, R., Koczwara, B., Rankin, N. M., Brown, S., Weber, M., Mazariego, C., Grogan, P., Stubbs, J., Thottunkal, S., Canfell, K., Blyth, F. M., Banks, E. 2023; 23 (1): 774

    Abstract

    Pain is a common, debilitating, and feared symptom, including among cancer survivors. However, large-scale population-based evidence on pain and its impact in cancer survivors is limited. We quantified the prevalence of pain in community-dwelling people with and without cancer, and its relation to physical functioning, psychological distress, and quality of life (QoL).Questionnaire data from participants in the 45 and Up Study (Wave 2, n = 122,398, 2012-2015, mean age = 60.8 years), an Australian population-based cohort study, were linked to cancer registration data to ascertain prior cancer diagnoses. Modified Poisson regression estimated age- and sex-adjusted prevalence ratios (PRs) for bodily pain and pain sufficient to interfere with daily activities (high-impact pain) in people with versus without cancer, for 13 cancer types, overall and according to clinical, personal, and health characteristics. The relation of high-impact pain to physical and mental health outcomes was quantified in people with and without cancer.Overall, 34.9% (5,436/15,570) of cancer survivors and 31.3% (32,471/103,604) of participants without cancer reported bodily pain (PR = 1.07 [95% CI = 1.05-1.10]), and 15.9% (2,468/15,550) versus 13.1% (13,573/103,623), respectively, reported high-impact pain (PR = 1.13 [1.09-1.18]). Pain was greater with more recent cancer diagnosis, more advanced disease, and recent cancer treatment. High-impact pain varied by cancer type; compared to cancer-free participants, PRs were: 2.23 (1.71-2.90) for multiple myeloma; 1.87 (1.53-2.29) for lung cancer; 1.06 (0.98-1.16) for breast cancer; 1.05 (0.94-1.17) for colorectal cancer; 1.04 (0.96-1.13) for prostate cancer; and 1.02 (0.92-1.12) for melanoma. Regardless of cancer diagnosis, high-impact pain was strongly related to impaired physical functioning, psychological distress, and reduced QoL.Pain is common, interfering with daily life in around one-in-eight older community-dwelling participants. Pain was elevated overall in cancer survivors, particularly for certain cancer types, around diagnosis and treatment, and with advanced disease. However, pain was comparable to population levels for many common cancers, including breast, prostate and colorectal cancer, and melanoma.

    View details for DOI 10.1186/s12885-023-11214-5

    View details for Web of Science ID 001064533600001

    View details for PubMedID 37700229

    View details for PubMedCentralID PMC10498633