What is ArteraAI and what do its tests do?

ArteraAI is a set of artificial intelligence (AI) tools made by Artera, a company in California, USA. The tools look at a digital picture of a stained tissue slide, the same kind of slide a pathologist already studies under a microscope after a biopsy or surgery. The software combines what it sees in the slide with a few clinical details and gives an estimate of how likely the cancer is to spread in the coming years.

There are two tools with US Food and Drug Administration (FDA) marketing authorisation. ArteraAI Prostate uses slides from a prostate needle biopsy. ArteraAI Breast uses slides from breast cancer surgery. Both give risk estimates for distant metastasis, which means cancer spreading to other parts of the body.

These tools do not diagnose cancer. They are used after cancer has already been confirmed, to add information about risk. FDA says both are meant to help doctors along with other clinical and pathology factors, not on their own. Your oncologist decides whether such a test is useful for you.

Key facts about ArteraAI's FDA status

What FDA's own database and decision summaries show:

  • ArteraAI Prostate: De Novo request DEN240068, granted by FDA on 31 July 2025; the company announced it on 13 August 2025
  • ArteraAI Prostate is a Class II device for men aged 55 or older with non-metastatic prostate cancer, using slides from untreated needle biopsies
  • It gives 10-year risk of distant metastasis and of death from prostate cancer, in high, intermediate or low groups
  • ArteraAI Breast: 510(k) K254115, found substantially equivalent by FDA on 4 May 2026, with ArteraAI Prostate as the comparison (predicate) device
  • ArteraAI Breast is for adults with early-stage HR-positive, HER2-negative breast cancer with no or limited lymph node spread (N0 or N1)
  • It gives 5-year and 10-year risk of distant metastasis and sorts patients into low or high risk
  • Both are prescription-only software and must be used with a complete standard evaluation

Key takeaways: ArteraAI

  • ArteraAI reads a digital image of an ordinary stained biopsy or surgery slide; no extra tissue or new procedure is needed
  • FDA authorised the prostate tool in July 2025 and cleared the breast tool in May 2026, both as aids to risk estimation
  • Prostate evidence comes from large past NRG/RTOG radiotherapy trials, but the FDA studies were retrospective
  • The tools estimate risk for groups of patients; they cannot say for certain what will happen to one person
  • We found no Indian regulatory approval or Indian partner for ArteraAI on the company's pages as of November 2026

How does ArteraAI read a biopsy slide?

After a biopsy, the tissue is fixed, embedded in wax and cut into thin slices. These are stained with two dyes called haematoxylin and eosin, often shortened to H&E. This is the routine stain used in almost every pathology laboratory. A scanner then turns the glass slide into a very detailed digital image, called a whole slide image.

ArteraAI's software uses deep learning, a type of AI that learns patterns from many examples. The company trained it on slides and long-term outcomes from thousands of patients. When it sees a new slide, it looks at the shape and arrangement of the cells and adds clinical details. For breast cancer, FDA lists age, tumour size and lymph node status as the extra inputs.

The result is a risk score and a risk group. FDA's labels say the slide images must come from specific approved scanners, and that results depend on good slide and scan quality. A poor slide or the wrong scanner can affect the answer.

What exactly did FDA authorise for ArteraAI Prostate?

FDA granted ArteraAI Prostate through the De Novo pathway. This route is used for a new kind of moderate-risk device that has no similar device already on the market. FDA received the request on 25 November 2024 and granted it on 31 July 2025. It created a new device category, under regulation 864.3755, for pathology software that analyses digital images for prostate cancer prognosis.

According to the FDA decision summary, the software is for men aged 55 or older with non-metastatic prostate cancer who are candidates for treatment aimed at cure. That includes surgery, radiation with or without hormone or other systemic therapy, or active surveillance. It uses H&E slides from prostate needle biopsies taken before any treatment.

The report gives a 10-year risk group for distant metastasis and for death from prostate cancer. The authorisation also included a plan that lets the company add more FDA-cleared scanners later without a new submission. The device is not meant for people whose cancer has already spread.

ArteraAI Prostate FDA study numbers at a glance

From the pivotal study in the FDA decision summary:

  • Design: retrospective study of 886 patients from 3 US sites, diagnosed since 2005, with median follow-up of 8.2 years
  • Training data: 10,009 patients from several randomised trials and clinical studies
  • 10-year distant metastasis: 28.1% in the high group, 6.6% in the intermediate group and 3.3% in the low group
  • 10-year death from prostate cancer: 10.2% high, 1.1% intermediate and 0.6% low
  • African American patients were a small group (72 people), so their estimates were less precise

What did the NRG/RTOG trial studies of ArteraAI show?

Much of the ArteraAI prostate research uses data from five large randomised trials run by NRG Oncology and the Radiation Therapy Oncology Group (RTOG) in North America: RTOG 9202, 9408, 9413, 9910 and 0126. In all of them, men received radiotherapy, with or without hormone therapy of different lengths.

A 2022 study in npj Digital Medicine, by Esteva and colleagues, built the AI model from biopsy slides and clinical data. Slides were available for 5,654 of the 7,764 men in these trials, with a median follow-up of 11.4 years. In a held-out part of the data, the AI model predicted outcomes better than the standard NCCN risk groups, with a relative improvement of 9.2% to 14.6% across different outcomes. For 5-year distant metastasis, the score used to measure accuracy (AUC) was 0.84 for the AI model and 0.74 for NCCN risk groups.

Several authors worked for Artera, and Artera helped fund the work. All the men had radiotherapy in older trials, so results may not apply in the same way to every treatment or every country.

Can ArteraAI predict who benefits from hormone therapy?

A 2023 study in NEJM Evidence, by Spratt and colleagues, asked a different question. Could AI pick out which men are likely to benefit from adding short-term hormone therapy, called androgen deprivation therapy (ADT), to radiotherapy? ADT can cause side effects such as hot flushes, lower sex drive and loss of muscle.

The model was built using data from 5,727 patients in five phase 3 trials, then locked and tested on RTOG 9408. That trial randomly gave 1,594 men radiotherapy with or without 4 months of ADT, with 14.9 years of median follow-up. In the 543 men (34%) the model called positive, 15-year distant metastasis was 4.0% with ADT and 14.4% without it. In the 1,051 model-negative men, the figures were 6.9% and 7.4%, a difference that was not significant.

The authors note that this was not a new prospective trial, and the test of hormone therapy benefit is not part of the FDA authorisation, which covers risk estimates only. The study was funded partly by the US National Cancer Institute and partly by Artera.

ArteraAI Breast: what FDA cleared in May 2026

FDA cleared ArteraAI Breast on 4 May 2026 through the 510(k) route. In this route a company shows that a new device is substantially equivalent to one already on the market. Here the comparison device was ArteraAI Prostate itself. FDA received the submission on 19 December 2025.

The FDA decision summary says the software looks at H&E slides from breast cancer surgery, taken before any drug treatment. It adds the patient's age, tumour size and lymph node status. It is for adults with early-stage, hormone receptor-positive, HER2-negative invasive breast cancer, with no lymph nodes involved or only limited spread (N0 or N1), and no known spread to other organs after surgery. They must be people who would be considered for standard treatment after surgery.

It gives 5-year and 10-year risk of distant metastasis and places a person in a low or high risk group. FDA's cleared use is risk estimation to help doctors, along with other factors, not choosing a specific treatment.

ArteraAI Breast study numbers at a glance

From the clinical study in the FDA 510(k) decision summary:

  • Design: retrospective study of 1,271 patients at 3 US sites, diagnosed between 2006 and 2019
  • Low risk group (827 people): 5-year distant metastasis 0.9% and 10-year 2.8%
  • High risk group (444 people): 5-year distant metastasis 8.7% and 10-year 16.6%
  • Whole group: 3.6% at 5 years and 7.6% at 10 years
  • Cleared scanners: Philips Ultra Fast Scanner and Leica Aperio GT450 DX
  • Some groups, such as African American patients (63 people), were small with few events

Does NCCN mention the ArteraAI test?

The National Comprehensive Cancer Network (NCCN) is a group of US cancer centres that writes widely used treatment guidelines. In March 2024, Artera announced that the NCCN Guidelines for Prostate Cancer released on 27 February 2024 included the ArteraAI Prostate Test.

The company reports that NCCN listed it with a category 2A recommendation and level IB evidence, and that it was the only test included for its ability to predict benefit from therapy. The company describes category 2A as uniform NCCN agreement that the test is an appropriate option. We could not open the NCCN guideline text itself, because it needs a login, so these details come from the company's announcement.

NCCN guidelines are updated often, and wording can change between versions. They are written for the US setting. An oncologist in India may follow NCCN, other international guidelines or Indian practice, depending on the hospital and the patient. Ask your doctor what the current guideline says about AI pathology tests for your situation.

FDA-authorised ArteraAI software and the ArteraAI laboratory test: not the same thing

This point confuses many people. Artera offers two different things with similar names. The first is the FDA-authorised software devices described above, ArteraAI Prostate and ArteraAI Breast. The company says the prostate software can be used in qualified US pathology laboratories, and FDA lists the scanners that must be used.

The second is the ArteraAI Prostate Test and ArteraAI Breast Cancer Test, which Artera offers from its own laboratory in Jacksonville, Florida. The company's patient and breast test pages describe these as laboratory-developed tests and state that they have not been cleared or approved by FDA. They are ordered by a licensed clinician, and the company says results usually come within 1 to 2 days after the specimen reaches the laboratory.

So when you read that ArteraAI is FDA-authorised, it refers to the software devices and their specific uses. A report from the company laboratory may cover more than the FDA-authorised use, for example therapy benefit. Your oncologist can explain which version a report came from.

Limits of the ArteraAI evidence

Important gaps to keep in mind when reading about ArteraAI:

  • The FDA pivotal studies were retrospective, looking back at past patients, not new prospective trials
  • Validation was mainly in the United States; we found no published validation in Indian patients
  • Trial patients were treated years ago, and treatments and imaging have changed since then
  • Risk groups describe what happened to groups of patients and cannot predict one person's future
  • Several authors of the key studies worked for or advised Artera, and Artera helped fund the research
  • Results depend on slide quality and on using the approved scanners

Is ArteraAI available in India?

As of November 2026, we did not find any Indian regulatory approval for ArteraAI, or any Indian testing partner, on the company's pages. The FDA authorisation applies to the United States. In April 2026, Artera announced CE marking under the European Union's in vitro diagnostic rules for its prostate biopsy and breast cancer assays, covering the 27 EU countries and Iceland, Liechtenstein and Norway. That announcement did not mention India or other Asian countries.

The company's breast test page names a partner in the United Kingdom. Its US laboratory test is ordered by licensed clinicians, and the company states that patients cannot order it themselves.

Because the test works on a digital image of an existing slide, some families wonder whether slides could be sent abroad. Whether that is possible, what it would cost and whether the result would change anything in your plan are questions for your oncologist and pathologist. Doctors already use risk information such as Gleason grade, PSA, stage and receptor results to plan treatment.

How does ArteraAI fit with Gleason score, staging and other tests?

ArteraAI does not replace the usual tests. For prostate cancer, doctors already group risk using the Gleason grade from the biopsy, the PSA blood test and the stage. These form the NCCN risk groups. The Esteva study compared the AI model against these risk groups, and the FDA label says the software is used along with other clinical and pathology factors.

For breast cancer, the usual picture includes tumour size, grade, lymph node status, and hormone receptor and HER2 results from the pathology report. ArteraAI Breast uses age, tumour size and node status as inputs, plus the slide image. FDA requires it to be used with a complete standard evaluation.

So the AI result is one more piece of information. A doctor still weighs it with the scans, the full pathology report, your general health and what matters to you. If an AI risk result and your other results seem to point in different directions, that is a good question to raise at your next visit.

What happens next for ArteraAI research?

Several next steps are already described in published sources. FDA's prostate authorisation includes a plan that allows Artera to add other FDA-cleared scanners without a new 510(k) submission, which could widen the number of laboratories able to use it. The breast clearance currently lists two scanners.

Outside the US, the company's April 2026 announcement said it was in discussions with European pathology laboratories and healthcare partners to set up local access. Whether these tools reach Asian countries, including India, is not stated in any source we found.

On the science side, the authors of the hormone therapy study noted that their model was not tested in a new prospective trial, and that their study was underpowered to look closely at some subgroups. Independent studies in more diverse populations, including South Asian patients, and prospective studies that show whether using the tool improves patient outcomes would answer the biggest open questions. Until then, the evidence is promising but limited to the settings studied.

Questions to ask your doctor about ArteraAI

  • Would an AI pathology test such as ArteraAI add anything to my Gleason grade, PSA, stage or receptor results?
  • Is my type and stage of cancer within the group the test was studied in?
  • If the test is not available in India, what other risk information do we already have?
  • Could my slides be sent abroad, and what would it cost and how long would it take?
  • Would a low or high result change my treatment plan, and how?
  • Which guideline does this hospital follow for prostate or breast cancer risk tests?

Sources

1. US Food and Drug Administration. De Novo classification database: ArteraAI Prostate (DEN240068). FDA, 2025. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/denovo.cfm?ID=DEN240068 2. US Food and Drug Administration. De Novo decision summary: ArteraAI Prostate (DEN240068). FDA, 2025. https://www.accessdata.fda.gov/cdrh_docs/reviews/DEN240068.pdf 3. Innolitics. ArteraAI Prostate (DEN240068), text of the FDA decision summary. Innolitics FDA database, 2025. https://fda.innolitics.com/device/DEN240068 4. US Food and Drug Administration. 510(k) premarket notification database: ArteraAI Breast (K254115). FDA, 2026. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K254115 5. US Food and Drug Administration. 510(k) decision summary: ArteraAI Breast (K254115). FDA, 2026. https://www.accessdata.fda.gov/cdrh_docs/reviews/K254115.pdf 6. Esteva A, et al. Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials. npj Digital Medicine, 2022. https://pmc.ncbi.nlm.nih.gov/articles/PMC9177850/ 7. Spratt DE, et al. Artificial Intelligence Predictive Model for Hormone Therapy Use in Prostate Cancer. NEJM Evidence, 2023. https://escholarship.org/uc/item/68n7n862 8. Artera. Artera receives U.S. FDA De Novo marketing authorization for AI digital pathology software. Artera news, 2025. https://artera.ai/news/artera-receives-u-s-fda-de-novo-marketing-authorization-for-ai-digital-pathology-software-revolutionizing-prostate-cancer-care 9. Artera. ArteraAI announced as the first and only predictive test for therapy personalization in the 2024 NCCN Guidelines for Prostate Cancer. Artera news, 2024. https://artera.ai/news/arteraai-announced-as-the-first-and-only-predictive-test-for-therapy-personalization-in-the-2024-nccn-guidelines-for-prostate-cancer 10. Artera. Artera secures CE marking for its AI prostate and breast cancer assays. Artera news, 2026. https://artera.ai/news/artera-secures-ce-marking-for-its-ai-prostate-and-breast-cancer-assays-unlocking-global-expansion 11. Artera. ArteraAI Breast Cancer Test. Artera, 2026. https://artera.ai/breast-cancer-test 12. Artera. ArteraAI Prostate Test for patients. Artera, 2026. https://artera.ai/for-patients

Frequently asked questions

What is ArteraAI and how does it work?

ArteraAI is a set of AI tools from the US company Artera. The software studies a digital image of a routine H&E-stained slide from a prostate biopsy or breast cancer surgery, adds a few clinical details, and estimates the risk of the cancer spreading in later years.

It does not diagnose cancer and needs no extra tissue. FDA says it is meant to help doctors along with other clinical and pathology factors, so your oncologist decides whether it is useful in your case.

Is ArteraAI FDA approved?

FDA granted ArteraAI Prostate through the De Novo pathway on 31 July 2025, and cleared ArteraAI Breast through the 510(k) route on 4 May 2026. Both are prescription-only software for estimating the risk of distant metastasis.

The company's own laboratory versions, the ArteraAI Prostate Test and Breast Cancer Test, are laboratory-developed tests, and the company states they have not been cleared or approved by FDA. Ask your doctor which version a report came from.

What does an ArteraAI prostate report show?

Under the FDA authorisation, ArteraAI Prostate gives a 10-year risk group, high, intermediate or low, for distant metastasis and for death from prostate cancer. It is for men aged 55 or older with non-metastatic cancer, using slides from biopsies taken before treatment.

The group is a risk estimate based on past patients. It cannot say for certain what will happen to one person, so your oncologist reads it together with your Gleason grade, PSA, stage and other results.

How accurate is ArteraAI for prostate cancer?

In the FDA pivotal study of 886 patients, 10-year distant metastasis was 28.1% in the high risk group, 6.6% in the intermediate group and 3.3% in the low group. In a 2022 study using NRG/RTOG trial data, the AI model predicted outcomes better than NCCN risk groups, with an AUC of 0.84 versus 0.74 for 5-year metastasis.

The FDA study was retrospective and done in the US. Several study authors worked for Artera, and we found no validation in Indian patients.

Can ArteraAI tell if I need hormone therapy with radiotherapy?

A 2023 NEJM Evidence study tested this using the RTOG 9408 trial. In men the model called positive, 15-year metastasis was 4.0% with short-term hormone therapy and 14.4% without it. In model-negative men, the figures were 6.9% and 7.4%.

This was a look back at an older trial, not a new prospective study, and the FDA authorisation covers risk estimates only. The decision on hormone therapy rests with your oncologist, who weighs side effects and your full picture.

What is ArteraAI Breast cleared for?

FDA cleared ArteraAI Breast on 4 May 2026 for adults with early-stage, HR-positive, HER2-negative invasive breast cancer with no or limited lymph node spread (N0 or N1) and no known spread after surgery. It uses the surgery slide plus age, tumour size and node status.

It gives 5-year and 10-year risk of distant metastasis and a low or high group. In the FDA study of 1,271 patients, 10-year metastasis was 2.8% in the low group and 16.6% in the high group.

Is ArteraAI test available in India?

As of November 2026, we found no Indian regulatory approval and no Indian testing partner for ArteraAI on the company's pages. FDA authorisation covers the US, and in April 2026 the company announced CE marking for the European Union.

The company's laboratory test is ordered by licensed clinicians, not by patients. If you are curious, ask your oncologist whether it would add anything to your existing results, and whether sending slides abroad is possible and worth the cost.

Is ArteraAI in the NCCN guidelines?

Artera announced that the NCCN Guidelines for Prostate Cancer released on 27 February 2024 included the ArteraAI Prostate Test. The company reports a category 2A recommendation and level IB evidence.

We could not check the NCCN text directly because it needs a login, and guidelines change between versions. NCCN is a US guideline. Ask your oncologist what the current guideline says and which guidelines their hospital follows for your type of cancer.

This page is general information, not medical advice. OncoVOIS explains and helps you prepare; your oncologist decides about your care.