The Pharmacy Wants to Be the Factory | The Lattice Brief (10/11/26)

This week’s Lattice Brief looks at three bets on printing medicines in the pharmacy: PharmaTher’s CraftMake roll-out to US compounding pharmacies, CurifyLabs’ dose-accuracy work, and France’s first license for drug-printing cartridges. Also inside: FDA clears a patient-specific 3D-printed radiation bolus, a one-year randomized trial of printed vs zirconia crowns, and our updated 3D Printed Drugs Guide.

The big thing.

The pharmacy wants to be the factory

Three companies made the same bet in two weeks. PharmaTher says it expects the first orders for its CraftMake pharmaceutical 3D printer this quarter, with up to 50 installations in US compounding pharmacies within a year and a New Drug Application for a rare pediatric therapy by late 2028. A doctoral thesis from Åbo Akademi University and spin-off CurifyLabs reported on dose accuracy for pharmacist-dispensed gels. And France’s medicines regulator licensed MB Therapeutics to make printing cartridges as a pharmaceutical manufacturer.

The bet: 3D printing can scale in compound medicine.

The clinical case is real but thin. A 2026 review counted 10 clinical studies of printed medicines made at the point of care. Most were single-centre and short. The best known comes from Santiago de Compostela, where four children with maple syrup urine disease took printed chewable isoleucine for six months. According to the study, indexed in PubMed, their blood levels sat closer to target and varied less than on hand-filled capsules (DOI). Children and rare diseases are where tailored doses matter most, so that is where this market starts.

The business model of choice is razor and blade. Place the printer once. Sell validated formulations, cartridges and software every week.

On the regulatory front, the rules are arriving unevenly. The UK’s point-of-care framework has been in force since July 2025. FDA’s July proposal would let a distributed manufacturing network register as one establishment, but it does not cover compounding, which is where US printed doses live today.

Again, the printer is not the hard part. Batch release, adverse-event reporting and recalls across 50 sites are. Taking responsibilities for a bad batch is required to own the market.

What do you think? Is 3D printed drugs incremental improvement or transformational to the future of medicine?

We have updated our 3D Printed Drugs Guide with everything that changed since August 2025: processes, players, clinical evidence and regulation. For the people building it, listen to FABRX’s Alvaro Goyanes on Lattice Podcast Episode 112, or watch our panel with FABRX, Craft Health, CurifyLabs and NIST on demand.

From 3DHEALS

Regulatory watch

  • FDA clears Myrava’s patient-specific 3D printed bolus (3D Printing Industry, Oct 2). A 510(k) for a printed layer that shapes radiation dose at the skin. It is the former Ricoh 3D for Healthcare’s first clearance beyond anatomical models, made from treatment-plan files at FDA-registered point-of-care sites. Hospital printing is moving from models to devices that change treatment.
  • FDA to prioritize guidance on AI and surgical robots next year (MedTech Dive, Oct 2). CDRH’s fiscal 2027 A-list puts final guidance on AI-enabled devices and robotic surgery submissions at the top. 3D printing is not on the list.

Clinical & research

  • Printed resin crowns vs zirconia: a one-year randomized trial (Journal of Dentistry, via PubMed). In 56 patients at Yonsei, fracture-free survival was 78.6% for printed resin crowns vs 92.9% for zirconia, a difference that was not statistically significant. Fit and patient satisfaction were similar. Printed crowns wore about six times more, and 7 of 8 fractures were on the back-most teeth. Fine for some cases; not yet a zirconia replacement for molars.
  • Printed vs thermoformed finger orthoses (3D Printing in Medicine, via VoxelMatters). In a pilot with 20 occupational therapy students, printed trigger-finger orthoses scored higher on fit, comfort and durability and cut hands-on labor. They took more total time and material cost. Small and not randomized.
  • MLATE V3: open-source AI for printable scaffolds (Biofabrication, via PubMed). A Polish Academy of Sciences team released an open dataset of 2,646 samples and a free tool that predicts printability and cell response before printing, then ranks bioink formulations. Fewer failed prints, if it generalizes.

Also this week

Dental and devices

Research

Industry and physical AI

Community

Get The Lattice Brief every week.

Subscribe to The Lattice Brief

Comments