The Do-It-Yourself IVF Cycle

From Democratizing IVF: Using Science, Medicine, And Engineering (And Venture Capital) To Help Build Families

I’ve met some pretty extraordinary inventors since I launched the IVF venture project a few years ago. Let’s take some of their ideas, add some technology in development and some off-the-shelf healthcare services, and play around with a wild idea: a do-it-yourself IVF cycle.

Getting started is easy: a combination of mail-in AMH ovarian reserve testing, home urine hormone testing and a home mini-centrifuged semen analysis to verify that there are indeed moving sperm in the ejaculate.

Enter the results into an online database, which, having analyzed tens or hundreds of thousands of successful and unsuccessful IVF stimulations, comes up with an appropriate stimulation protocol, which is then transmitted to an online pharmacy which, after a quick doctor check, dispenses medications. Stimulation instructions follow, with a monitoring schedule. A home self-ultrasound is performed by placing a transducer either on the abdominal wall or in the vagina — the probe does a survey of the pelvis and constructs three-dimensional images of the ovaries and the uterus. The information is transmitted to the database (bluetooth? wifi? connected to a smart phone?), along with the hormone results from that morning’s home urine test. Further medication and monitoring instructions follow, algorithm-generated, until the eggs are ready for trigger and retrieval. The database shows a list of free-standing egg retrieval stations, and the patient clicks a box and schedules the egg retrieval, at a convenient time and place.

The retrieval may occur at an IVF clinic, or an ob/gyn office, or perhaps a surgical center (probably not at a local CVS or Walgreen’s, but you never know.) The patient meets the doctor right before the procedure. The eggs are aspirated transvaginally under ultrasound guidance, and the follicular fluid is pumped directly into a machine where the eggs are isolated, washed, stripped of the surrounding cells by a specialized filter, vitrified (flash frozen) and placed is a specialized container, which the patient leaves with. The container can be kept at home or delivered to a central storage facility until needed.

Up to this point the cycle was pretty much controlled by the patient. It would be advisable to have a pre-cycle checkup and exam, including an assessment of the health of the fallopian tubes and the shape of the uterine cavity. These can be done as part of the precycle exam or scheduled independently with a radiology practice. I also assume that the central stimulation database service would have a way to flag outlier results and reflex recommend an online consult with a staff doctor to assess them; the tiny percentage of “non-routine” results means that the backup doctor could oversee thousands of cycle at a time.

Phase three of the cycle starts with delivery or transfer of the vitrified eggs to an IVF lab, along with a detailed history, access to a blockchain based database with the stimulation information and the characteristics of the eggs. With permission, the IVF clinic can access genetic testing information, including karyotype (chromosome counts) and mutation screening to assess the risk for inherited disease. The database will list options and make recommendations, again based on the machine-learned aggregation of tens or hundreds of thousands of prior cycles, the inputs drawn again from the blockchain with access given by permission of the patient.

We’re a long way off from total mechanized embryology, but thaw and non-ICSI (intracytoplasmic sperm injection) fertilization (in other words the type where sperm and eggs are placed together in a Petri dish) can be mechanized, and embryo development assessed with a combination of time-lapse imaging, and continuous media sampling and analysis. If, as several of the researchers I have met with suggest, these data, combined with AI, can replace embryo biopsy and preimplantation genetic testing for aneuploidy or mutation analysis, then the resulting embryos can be ranked for probability of outcome, with one triaged to robotic loading into a transfer catheter and the rest of the viable embryos cryopreserved, all inside the machine, and then given to the patient for storage.

A quantitative home urine pregnancy test follows in nine days or so and repeated two days later if positive. A home ultrasound a week later runs its pictures through the blockchain data to cross check dates(accessed by plugging the ultrasound probe into a smart phone), and checks for a gestational sac and yolk sac. Repeated a week later, the same procedure detects and calculates the fetal heart rate and measures the growth of the fetal pole. At that point the patient goes online and schedules a prenatal visit.

A pretty radical reinvention of the practice of IVF, I admit. Now let’s pick it apart.

The first objection is that IVF patients, used to lots of personal attention and access to doctors and nurses on a daily basis, won’t tolerate a cycle where these interactions are replaced by impersonal protocols.

That may be true, but as we enter our fourth decade of online interaction and commerce we may be able to combine the lessons learned from Zocdoc, Amazon, Zappos and the migration of the entire commercial banking industry to a satisfactory model. One after another, healthcare sectors have migrated to different combinations of patient-directed care. I make my own appointments, and I rarely interact with the professional staff at One Medical for my primary care; colonoscopy centers operate efficiently with minimal pre-procedure contact and lots of patient autonomy with minimal teaching for bowel preps. Of course the multi-day fertility drug stimulation procedure is a lot more complicated than mixing laxatives and drinking quarts of GoLYTELY, but that complexity is a result of repetitive steps, none particularly onerous, even the subcutaneous injections — something that young diabetics have been doing for decades.

But even if do-it-yourself IVF is feasible, isn’t a high-touch, hand-holding model preferable to leaving the patient on her own?

Well, uh, yes.

That is, unless high-touch, hand-holding IVF is not the alternative and the actual alternative is not having access to IVF at all.

The point of do-it-yourself IVF is not to create a clever hack. As part of the drive to democratize IVF, it’s an extreme version of the type of alternative delivery that could make IVF available to those who cannot afford it now, or who can’t commit the amount of back-and-forth to the clinic time, or who live too far away from a stand-alone, do everything under one roof clinic. All else being equal, these patients probably would prefer a high-touch, constant contact system. But the fact that only about 1% of our infertile family members, friends, colleagues and neighbors are achieving an IVF solution to their problem tells us that the current model is not enough, that we need alternatives. If not the extreme of do-it-yourself, then maybe a technology-enabled, less personal but still effective choice for the huge population of people for whom the alternative is not what other people are using; for them the alternate is nothing at all.

So how close is this vision to reality? Is this the near future, or is it science fiction? Let’s take it step by step.

  1. Mail-in AMH testing: available today.
  2. Quantitative home urine hormone testing: about halfway there, a few hormone short of a full panel. Give it a year or two.
  3. Home mini-centrifuge for do-it-yourself semen analysis: available today.
  4. Database of IVF stimulation management: the data exists, it needs a good patient-friendly front end, and outcome validation. Being worked on in several places, needs a touch more venture financing. Not rocket science, from what my data colleagues tell me.
  5. Home ultrasound that can interpret the ovarian and uterine parameters: we’re getting there. The technology is being developed for cardiology, extension to IVF, with a dedicated team and financing could make it real by 2025 (my opinion.)
  6. Getting the database to talk to IVF labs and present alternatives for where to go for retrieval: the tech is trivial. Borrow it from Kayak.
  7. The closed-engineered, retrieval to vitrification system, allowing egg retrievals anywhere, keeping the eggs sterile since they never see the room air: this is being pieced together in various forms at several places around the world. Hooking up the tubing from the aspiration needle to the machine is tricky — pressure management, routing the eggs as they come to their own title sections and processing the cells around them is not simple. The automated freezing part is out there already but still needs some work — a series of media changes, temperature and pressure adjustments and movement of the egg into a portable frozen form is part straightforward, part NASA level complex.
  8. Central cryo (frozen) storage: exists already, can be modified for home.
  9. Home collected genetic testing: already exists.
  10. Blockchain-based IVF specimen data collection and storage with access by selective permission: the current joint venture being developed by IBM and the nonprofit ivfOPEN (disc: I am co-chair of ivfOPEN.)
  11. Robotic fertilization: I have seen a robot perform ICSI, but the system was not quite ready for clinical use. Robotic IVF (ie not ICSI) is under development.
  12. Continuous media sampling: I have seen this in an academic lab; others are working on it.
  13. Time lapse embryo assessment: in the field and working well.
  14. Noninvasive embryo genetic assessment: at the preliminary data stage.

In sum, a lot of the technology we need for do-it-yourself IVF is already out there, some of it in use for discrete steps in the IVF process, some of it under development as comprehensive solutions, and some of it waiting on the shelf, having been developed for other medical applications. With the equivalent of one or two immuno-oncology sized series B venture rounds of funding and a room full of bioengineers and embryologists we could get there in five years or so.

But do we want to? Maybe do-it-yourself IVF is too bio-hack-ey and insufficiently Hippocratic. But the millions of people suffering from infertility without access to any form of IVF deserve something in the middle. We have bespoke kitchens in destination restaurants, and we also have hot dogs cooking on metal rollers in plastic cases in the Seven Eleven. The Holiday Inn is about five blocks from the Ritz Carlton in Union Square. Technology, engineering and science together can and will provide more options, more than just a choice between full-fare IVF and nothing at all.

bio fund manager, Columbia prof, ex-reproductive endocrinologist, roadie for @PriyaMayadas

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store