Diagnostics · Contraception
adyn
A $369 at-home kit that reads DNA from saliva and six hormones from a finger prick, then puts the result in front of a clinician who helps pick a birth control method. On sale since May 2022.
Day three of the cycle
A box arrives. Inside is a tube to spit into and a lancet with a card that soaks up a few drops of blood. The blood has to be taken on day three of the cycle. That is not a flourish: day three is where the clinical reference ranges for these hormones are defined, so fixing the day is what makes the numbers readable.
Saliva goes for DNA. The card goes for six hormones: anti-Mullerian hormone, estradiol, sex hormone binding globulin, FSH, LH and DHEA sulfate. Two to six weeks later a report comes back, and a provider trained in contraceptive counseling spends 25 minutes on video going through it.
Finger sticks scare people, so staff run video sessions to talk customers through the lancet. They still get it wrong. “And we watched someone spit into the wrong end of a saliva tube,” the founder said in 2025. “And that’s the easy part of the two sample collections.”
The problem is not imaginary
Hormonal contraception is among the most used medicines in the country and the least personalised. Roughly 200 products sit on the US market, and side effects are the usual reason people switch or stop. A Danish cohort of 1,061,997 women, published in JAMA Psychiatry in 2016, found combined oral contraceptive users had a rate ratio of 1.23 for a first antidepressant prescription, rising to 1.8 among girls aged 15 to 19. That is a population association, not a prediction about one person, but it stopped clinicians calling mood effects imaginary.
Clots are the sharper end. The 2024 US Medical Eligibility Criteria puts thrombophilia, naming factor V Leiden and prothrombin gene mutation, in category 4 for combined hormonal contraception, the most restrictive class CDC has. A carrier who does not know she is one is the patient adyn says it exists to find.
A geneticist who was told it was impossible
Elizabeth Ruzzo grew up in Seattle, studied at the University of Washington, took a PhD in human genetics and genomics at Duke, and did a postdoc at UCLA where machine learning on whole genome sequences from autism families turned up new risk genes.
Somewhere in there she went on the pill and became suicidal. Her doctor said contraception does not do that. She stopped taking it and recovered. “I was gaslit by my medical professional who told me that no, birth control does not do that,” she said in 2025. She founded adyn in 2019 as a solo technical founder.
Launch, then a long flat stretch
A $2.5 million seed round co-led by Lux Capital and M13 was announced in April 2021, with Anne Wojcicki of 23andMe among the angels. The test went on sale on May 23, 2022 at $369, everywhere in the US except New York, New Jersey and Rhode Island. In March 2023 the company said it had sold into 43 of the 47 states where it operates, that 94% of customers started or switched a method, and that it had flagged genetic clot risk in 7.5% of an initial population. Since then the record is quiet: no further round, no publication, no registered study. On September 23, 2026 the Y Combinator directory listed adyn as active with seven employees.
What is proven, and what is still claimed
| Evidence | What the record shows | Source type |
|---|---|---|
| FDA status | No 510(k), PMA, classification or device listing record in openFDA, searched Sep 23, 2026. Trade coverage of the launch noted the test “has not been reviewed by the Food and Drug Administration.” FDA states that “In general, direct-to-consumer tests for non-medical, general wellness, or low risk medical purposes are not reviewed by the FDA before they are offered.” FDA’s rule to phase in oversight of laboratory developed tests was vacated on March 31, 2025. | Public record |
| Clot genetics | CPIC lists one gene paired with hormonal contraceptives for systemic use: F5, the factor V Leiden gene, at CPIC level C, ClinPGx annotation level 2B, a “Moderate level of evidence for the association.” CPIC states that “CPIC level C and D gene/drug pairs are not considered to have adequate evidence or actionability to have prescribing recommendations.” No guideline covers the pair; F2 has no pair with contraceptives at all. | Public record |
| Guidance on screening | The 2024 US MEC thrombophilia entry: “Routine screening in the general population before contraceptive initiation is not recommended.” A 2016 CDC genomics post is softer, listing factor V Leiden carriers on oral contraceptives as a case where thrombophilia testing is appropriate. | Public record |
| Which variants adyn uses | The test “directly assays two well-established genetic risk markers known to lead to a higher risk of clotting.” They are not named on the site. In 2025 the founder names factor V Leiden and says two of about 15 clot loci met the company’s thresholds. | Company-stated |
| Depression prediction | Done with a polygenic risk score. No published study shows such a score predicts who becomes depressed on a particular contraceptive formulation. | Not found |
| Contraceptive pharmacogenomics | A 2019 study of 350 implant users, genotyped across 120 variants, concluded: “Only BMI and duration of implant use remained significantly associated with steady-state etonogestrel concentrations.” A later genome wide study of 880 users by the same group reported that “we did not identify genetic variants in metabolic pathways significantly associated with alterations in serum etonogestrel concentrations.” | Public record |
| Laboratory partner | Samples run at CLIA certified labs, described in 2022 coverage as CLIA/CAP accredited. The laboratories are not named in any source found. | Not named |
| Trials and papers | No entry for adyn or The Birth Control Test in the trial registry or PubMed as of September 23, 2026. NIH awards returned by a name search belong to Adynxx, Inc., an unrelated company. | Not found |
| Size of the evidence base | One page of the site cites “evidence from 60+ peer-reviewed publications.” Another cites “over 50 peer-reviewed research studies.” No list is public. | Figures differ |
Read plainly: the clot half rests on the best established pharmacogenomic association in contraception, and even that sits where the field’s own consortium declines to recommend a prescribing action. The depression half rests on a polygenic score with no published evidence it predicts drug specific response. Between them sits a real clinical service: a long counseling visit, a structured history, and a check for variants the CDC treats as category 4 once known. Whether the test prevents side effects, the marketing claim, has not been measured against a control group.
What to watch
- Whether adyn publishes. Its Close the Gap programme has had IRB approval since 2022 and nothing has appeared in PubMed.
- Any prospective comparison of side effect rates between tested and untested users, which would settle the claim.
- Whether CPIC or ClinPGx move F5 above level C, or add a second gene.
- Insurance coverage, called a priority since 2022, with only HSA and FSA accepted.
In their words
“The problem right now is that doctors can’t easily predict who is going to have what kind of side effects.”
Elizabeth Ruzzo, founder and CEO, to Axios, 2021 · Independent
“For the blood clot variants, there were, I think, 15 that had been implicated in the literature, sites in the genome associated with blood clot, right? But only two of them met our criteria in terms of sample size, study design, and statistical significance for us to be reporting on them.”
Elizabeth Ruzzo, The Bioinformatics CRO Podcast, 2025 · Interview
“A warning personalization would be something like, you know, you carry the Factor V Leiden mutation, which means you need to avoid these specific birth controls.”
Elizabeth Ruzzo, The Bioinformatics CRO Podcast, 2025 · Interview
“There’s a lot of tests in the market out there that are more what I would consider infotainment, right?”
Elizabeth Ruzzo, The Bioinformatics CRO Podcast, 2025 · Interview
“We need these findings to be replicated independently.”
Elizabeth Ruzzo on the 2019 CYP3A7*1C result, to BioSpace, 2021 · Independent
“Despite what we found, that genetics can affect the birth control hormones, we still know that the implant and the IUS are the best birth control methods by far.”
Aaron Lazorwitz, MD, University of Colorado, to The Guardian, 2019 · Independent
“Although genetic makeup might have an impact on how well contraception might work for some women, this is only one study and it is very early to tell whether this [genetic testing] technology will be helpful in the future.”
Sharon Cameron, Faculty of Sexual and Reproductive Healthcare, to The Guardian, 2019 · Independent
Related companies
Sources
- Public recordU.S. Medical Eligibility Criteria for Contraceptive Use, 2024
- Public recordCPIC gene and drug pair table, ATC G03A and F5, F2, CYP3A7
- Public recordPrioritization: assignment of CPIC levels
- Public recordEvidence-Based Framework for Pharmacogenomics Knowledge
- Public recordInfluence of Genetic Variants on Etonogestrel Concentrations
- Public recordPharmacogenomics of the Etonogestrel Contraceptive Implant
- Public recordAssociation of Hormonal Contraception With Depression
- Public recordIn vitro diagnostics: direct-to-consumer tests
- Public recordopenFDA device database queries for adyn
- Public recordAppropriateness and Cost of Thrombophilia Panel Testing
- Public recordCourt Vacates FDA’s Laboratory Developed Tests Final Rule
- IndependentGenetics may reduce efficacy of hormonal contraception
- IndependentPersonalized birth control tests and side effects
- IndependentCreating A Safer Birth Control: adyn Raises $2.5M Seed
- IndependentAdyn Test Brings Precision Medicine to Birth Control
- IndependentAt-Home Test Helps With Birth Control Selection
- InterviewThe Bioinformatics CRO Podcast, episode 61
- Interviewadyn: A New Era in Personalized Birth Control
- CompanyProduct, science, overview and FAQ pages
- Companyadyn reports blood clot risk and sales results
- Companyadyn directory entry: status and headcount
Profile researched and written by Healthcare Discovery. Last updated September 29, 2026.
