# Epidemiology of the Inverse Comorbidity of Dementia and Cancer: Statistical Methods and Biological Mechanisms

> **NIH NIH R03** · UNIVERSITY OF MARYLAND BALTIMORE · 2021 · $309,000

## Abstract

Over 44 million adults worldwide are affected by dementia. Alzheimer’s disease (AD) and its related dementias
(ADRD) are the most common causes of dementia, contributing to 60%-80% of cases in older adults. A
defining feature of AD/ADRD is the accumulation of amyloid-beta protein in extracellular plaques within the
brain and subsequent neuronal loss. Consequently, the dominant hypothesis for AD pathogenesis, the
“amyloid hypothesis,” postulates that amyloid-beta accumulation is the primary event leading to AD. Most
candidate treatments have targeted amyloid-beta; however, the unsuccessful trials of these treatments
challenge the amyloid hypothesis and motivate the need for new intervention targets. Recent epidemiologic
data also support alternative hypotheses for AD pathogenesis. Specifically, observational studies have
consistently shown that dementia and AD are inversely related to cancer. This “inverse comorbidity” may be
due to pathogenesis at opposite ends of a shared biological mechanism. Two proposed mechanisms to explain
inverse comorbidity between dementia and cancer are energy metabolism and inflammation. The scientific
premise for these mechanisms is supported by findings that neuronal glycolysis is downregulated in AD and
upregulated in cancer cells; also, higher markers of systemic inflammation are observed in persons with AD,
whereas inflammatory complexes may promote or inhibit tumor growth depending on context. However,
epidemiological studies of dementia and cancer have not rigorously investigated whether energy metabolism
and inflammation explain the inverse comorbidity. Moreover, studies of inverse comorbidity are vulnerable to
biases from selective survival because persons with cancer are more likely to die before dementia onset than
persons without cancer. Limitations of current statistical methods are a key barrier to accurately quantifying the
strength of inverse association between dementia and cancer and to identifying markers of biological
mechanisms that may explain inverse comorbidity. Therefore, new statistical methods are needed. The specific
aims of this proposal are to 1) quantify the association between incident dementia and cancer over time and 2)
test the relation of metabolomic and inflammatory markers jointly with incident dementia and cancer. To
achieve these goals, we will adapt and refine novel structural models for bivariate time-to-event data and apply
them to harmonized archived data on over 10,000 adults aged at least 65 years enrolled in three prospective
cohort studies (Health, Aging and Body Composition; Osteoporotic Fractures in Men Study; and Study of
Osteoporotic Fractures). We hypothesize that even after addressing selective survival, dementia and cancer
have an inverse association that is explained, in part, by markers of energy metabolism and inflammation.
Novel statistical methods will provide a rigorous framework to test inverse comorbidity and will be made
available to the scientific communi...

## Key facts

- **NIH application ID:** 10105042
- **Project number:** 1R03AG070178-01
- **Recipient organization:** UNIVERSITY OF MARYLAND BALTIMORE
- **Principal Investigator:** Michelle Denise Shardell
- **Activity code:** R03 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $309,000
- **Award type:** 1
- **Project period:** 2021-03-01 → 2024-02-29

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10105042

## Citation

> US National Institutes of Health, RePORTER application 10105042, Epidemiology of the Inverse Comorbidity of Dementia and Cancer: Statistical Methods and Biological Mechanisms (1R03AG070178-01). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/10105042. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*