Fred Hutch researchers look at longevity in older women, metabolic signatures of vaginitis

Few studies have been conducted on women who live into their 80s and 90s; it’s often difficult to recruit large enough numbers of elderly individuals. The Women’s Health Initiative, however, housed at Fred Hutch, is the exception, with 30-plus years’ worth of research data on thousands of elderly women, proving it to be the gift that keeps on giving.

Reiner and his group will use a large sample of older women from the WHI who’ve already gone through deep targeted CHIP sequencing, including 7,000 women with longitudinal CHIP data at two separate time points, to answer these questions.

“We will supplement existing WHI CHIP data by obtaining new baseline CHIP measures to study exceptional longevity prospectively [i.e., moving forward through time],” Reiner said.

The WHI data includes detailed demographics, anthropometrics (height, weight, BMI, etc.), lifestyle behaviors, cardiometabolic risk factors, medical history, environmental exposures, medication inventories, chronic diseases, physical and cognitive function and multi-omics, including genomics, epigenomics and proteomics.

“All told, we should have nearly 9,000 WHI women who survived more than 90 years and around 1,500 women who survived to 100 years during follow-up,” Reiner said. “Our study will be the first and largest to examine the prospective association of CHIP with exceptional longevity and health span, with the largest number of 90-somethings and centenarians of any study to date.”

Key findings from the study will be replicated and extended to men using two other cohorts with CHIP measurements, the phenotyped UK Biobank and the Jackson Heart Study, the largest long-term, single-site study of cardiovascular disease in African Americans.

Reiner believes the study will prove that CHIP is indeed associated with reduced longevity and health span for those 90 and older and plans to identify the biological and mechanistic pathways underlying these associations.

“Mounting evidence suggests chronic, low grade inflammation during aging contributes to CHIP clonal expansion and that CHIP-driven inflammation promotes chronic disease burden,” he said. “Our study will evaluate the potential for CHIP to be used as a novel, blood-based aging biomarker to predict risk of age-related diseases to advance precision medicine in older adults. This may identify novel therapeutic targets to delay or prevent multiple chronic diseases and preserve mobility in older adults as they age.”

Biomarkers for better vaginitis diagnosis

Vaginitis is a common condition that can cause burning, itching, discharge and odor.

“It’s a major reason women seek clinical care,” said Srinivasan, who studies how microbial communities in the vagina influence health and disease, including bacterial vaginosis (BV). Her research explores fundamental questions about the vaginal microbiome, such as how vaginal microbes interact with each other and with their host, and whether biofilm formation contributes to BV recurrence.

“In doing that research, translational questions came up,” she said, including, “Can we make better diagnostics?”

The challenge facing patients and clinicians is that vaginitis can have different infectious causes (if the cause is infectious at all). One woman’s vaginitis may be a case of BV, a disruption of the vaginal microbiome with a shift toward more anaerobic bacteria. Another’s vaginitis may be a yeast infection, while a third woman may have trichomoniasis, a sexually transmitted infection caused by a parasite.

“The symptoms can be similar, but the treatments are different,” Srinivasan said. Yeast infections need antifungals, while BV or trichomoniasis need different antimicrobial treatments. Determining the underlying cause is essential for appropriate treatment.

“We do have diagnostic tools, but errors can occur,” Srinivasan said. One study she cited estimated that over 40% of women who sought vaginitis care in community-based clinics received inappropriate treatment.

Existing approaches involve tradeoffs, Srinivasan said. Tests performed during a clinic visit can provide rapid answers but may not be accurate. Laboratory-based molecular tests can be accurate but may require additional time, equipment and cost.

When a diagnosis is delayed or incorrect, patients may need repeat visits or additional treatment, increasing healthcare use and extending the time to symptom resolution, she said.

Srinivasan, with Fred Hutch microbiome researcher David Fredricks, MD, has studied bacterial communities associated with BV. The composition of these communities can differ substantially among individuals, and no single bacterial species is present in every case of BV. Rather than replacing organism-based molecular testing, the researchers asked whether small molecules produced by microbial metabolism could provide a complementary, function-based signal across varied BV-associated microbial communities.

“We saw greater consistency in the small molecule metabolites,” Srinivasan said. The metabolite signatures they identified were promising biomarkers for BV and supported their potential use in diagnosis. Srinivasan, Fredricks and Fred Hutch metabolomics expert Daniel Raftery, PhD, have continued to identify and evaluate metabolites for BV diagnosis.

But for a vaginitis test to guide appropriate treatment, a vaginitis test must be able to distinguish among the three common infectious causes.

“For it to be clinically useful, it would need to have a single platform that can distinguish among the three causes,” Srinivasan said. Her R01 award will support the development and evaluation of small molecule metabolite-based biomarkers for BV, yeast infections and trichomoniasis. This is the first, essential step needed to lay the groundwork for a future diagnostic platform.

Srinivasan and her collaborators, Fredricks and Raftery, will use machine learning approaches to comb through more than 4,000 metabolite features and identify patterns associated with vaginitis. Fredricks’ team has assembled a unique set of samples to work from: daily vaginal swabs collected from 251 women before vaginitis developed, as well as before and after treatment.

This rich set of samples, collected over time, will enable Srinivasan to look for metabolite patterns that associate with infectious causes of vaginitis and to see if metabolite signatures change over time. She and her collaborators also plan to prospectively collect vaginal fluid samples at a second site. This will help them evaluate whether the biomarker signatures they discover generalize across populations and are suitable for broader use.

“We’ll be able to see if, after treatment, do the small-molecule metabolites return to an earlier profile?” Srinivasan said. “Can we use those changes to assess a sustained response to treatment, and can we use that information to predict recurrence? Those are questions that have not been typically addressed in vaginitis.”

This is called test-of-cure, a follow-up assessment to determine whether treatment produced a sustained clinical response rather than short-term symptom improvement. The longitudinal samples will also allow Srinivasan and her team to assess whether metabolites change before symptoms begin, potentially creating an opportunity for earlier vaginitis detection.

Once they have developed algorithms that can distinguish between different infectious causes of vaginitis, the team will evaluate them in an independent set of samples from 720 women, with and without vaginal discharge, previously collected at 26 clinical sites across the US. Srinivasan hopes that the work will yield a streamlined diagnostic algorithm for vaginitis.

Developing that algorithm is an essential step toward her longer-term goal: an easy-to-use diagnostic panel that, during a clinic visit, can accurately distinguish among infectious causes of vaginitis and help clinicians provide the appropriate treatment.

“The current R01 supports the development of the algorithm, not the diagnostic panel itself, but that’s what we’re working toward, that’s our ‘north star,’” Srinivasan said.