Quantum computing in healthcare and pharma

Healthcare sits at both ends of the quantum story. Pharma has the clearest scientific reason to want a quantum computer, because molecules follow quantum mechanics and classical computers approximate them. Hospitals and insurers hold some of the most sensitive, longest-lived data in any economy, which makes them early targets for “harvest now, decrypt later” attacks. Quantum computing in healthcare and pharma therefore means two very different projects, with very different timelines.

Which part of quantum computing is urgent for healthcare?

The urgent part of quantum computing for healthcare is data protection, not simulation. Patient records, genomes and trial data stay sensitive for decades, and much of that data is protected by RSA or elliptic-curve cryptography a future quantum computer could break. Drug discovery on quantum hardware is still research with no reliable date.

A patient’s diagnosis, genome or mental health history does not lose sensitivity after five years. Records protected today with RSA or elliptic-curve cryptography can be copied now and decrypted once a large enough quantum computer exists. Retention rules make the window long: Colombia’s Resolution 839 of 2017 sets 15 years from the last visit. Mexico’s NOM-004-SSA3-2012 sets a shorter five-year minimum, but a genetic test or a mental health diagnosis stays sensitive for the patient’s whole life, whatever the legal minimum says.

Clinical trial data raises the same issue for pharma. Trial sponsors, contract research organizations and hospital sites exchange patient-level data across borders for years, often through portals and file transfers protected by TLS with RSA or elliptic-curve keys. That traffic is a good candidate for early hybrid post-quantum key exchange.

Medical devices add a second problem. A connected infusion pump or imaging system can stay in service for a decade or more, and its cryptography is hard to change after it ships. In the United States, section 524B of the Food, Drug and Cosmetic Act requires cybersecurity information in premarket submissions for connected devices, and the FDA’s device cybersecurity guidance was updated in June 2025 to cover it. Manufacturers that design for crypto agility now will have an easier time when post-quantum algorithms become expected.

Quantum computing in healthcare and pharma research: what is real so far

Simulation. Here are the public examples we consider solid, and what each actually showed.

Boehringer Ingelheim and Google Quantum AI

Boehringer partnered with Google on quantum computing in 2021. Researchers from both companies, with coauthors from QSimulate, published a 2022 study in PNAS on cytochrome P450, the enzyme family behind most drug metabolism. The paper estimated the quantum and classical resources needed to model these enzymes and where a quantum advantage might begin. It is a map of the road, not a result from a quantum computer.

Moderna and IBM

The two companies used IBM quantum hardware to predict mRNA secondary structure, a step in designing mRNA medicines. The sequences were short, and classical tools still handle much longer ones.

AstraZeneca, IonQ, AWS and NVIDIA

In 2025 the group reported a hybrid quantum and classical simulation of a step in a Suzuki-Miyaura reaction, a common way to build small-molecule drugs. The speedup they reported was against earlier runs of the same quantum workflow, not against the best classical chemistry codes.

Cleveland Clinic and IBM

Since March 20, 2023 Cleveland Clinic has hosted an IBM Quantum System One on its main campus, the first quantum computer dedicated to healthcare research, as part of a 10-year partnership announced in 2021.

The common thread: serious companies are building skills and resource estimates while the hardware matures. None of them claims a quantum computer has found a drug.

How long until quantum simulation matters for pharma?

Quantum simulation will matter for pharma once error-corrected machines with thousands of logical qubits exist, and nobody can give a reliable year for that. Until then, the useful work is resource estimates, small hybrid experiments and training, so a research team is not starting from zero when the hardware arrives.

The chemistry problems with the clearest advantage, such as enzymes with many strongly interacting electrons, need error-corrected machines with thousands of logical qubits. Hardware makers publish roadmaps toward that scale, and Google’s Willow chip, described in Nature in December 2024, showed that adding qubits can reduce logical error rates, an important step. But roadmaps slip, and classical methods keep improving too.

A practical rule: if your research depends on a molecule class that classical methods clearly handle badly, start building internal skills now so you are not starting from zero later. If not, follow the field and revisit once a year.

What should a hospital or pharma company do first?

A hospital should start by inventorying where patient data is encrypted and asking device and records vendors for dated post-quantum plans. A pharma company should do the same for R&D data, then pick one molecular problem that classical methods handle badly and estimate what a quantum approach would need.

For hospitals, clinics and health insurers:

1. Inventory where patient data is encrypted and which systems use RSA or elliptic curves, starting with the electronic health record, backups and data exchanges with labs and payers.
2. Ask device and EHR vendors for their post-quantum plans in writing, with dates.
3. Rank systems by data lifetime. Genomic and mental health data go first.

For pharma and biotech:

1. Pick one well-defined molecular problem where classical methods struggle, and estimate what a quantum approach would need.
2. Train two or three computational chemists in quantum algorithms so they can read the literature critically.
3. Protect R&D data the same way as patient data. Trial results and compound libraries are exactly what an attacker willing to wait would store.

How AndesQubit can help

On the security side, post-quantum cryptography migration starts with the inventory described above. On the research side, quantum use case discovery screens molecular and optimization problems against what classical methods already do, and a hybrid quantum proof of concept tests the one that survives against a classical baseline. We are opening engagements in stages; get in touch if you want to be on the early-access list.