AI-Powered Precision for Neoadjuvant Therapy Planning

Ataraxis™ Breast NEO

Ataraxis Breast NEO is an AI-powered test that helps guide neoadjuvant therapy decision prior to surgery for patients with stage I–III breast cancer across all subtypes. By analyzing morphological features from standard H&E slides together with clinical data, the test delivers an individual’s likelihood of achieving pathologic complete response (pCR) after neoadjuvant therapy.

Overall pCR likelihood: deeper predictive insight for neoadjuvant decision-making

How do we predict likelihood of achieving pCR?

Flowchart showing H&E slide and clinical data input into Ataraxis AI models predicting 50% likelihood of strong response with neoadjuvant therapy or 5% likelihood suggesting surgery or alternative strategies.

Flowchart showing leveraging routinely collected data from H&E slide and clinical data to predict Ataraxis NEO High with 50% likelihood of reaching pCR leading to confident start of neoadjuvant therapy, and Ataraxis NEO Low with 5% likelihood suggesting surgery first or alternative neoadjuvant strategies.

Data you already have

Our platform uses the existing H&E-stained biopsy or resection slide collected during diagnosis, along with standard clinical variables such as:

Tumor size and nodal status
ER, PR, and HER2 status
Age and histologic subtype

Multimodal AI Analysis

The Ataraxis AI model evaluates over 10,000 micro-regions within the slide to capture tumor morphology, immune infiltration, and stromal architecture—features that correlate strongly with treatment response.
‍
These imaging-derived signals are integrated with clinical variables in a deep neural network that outputs a single, interpretable probability score.

Predictive Output

The test provides a continuous probability (%) of achieving a pathologic complete response (pCR) with neoadjuvant therapy.

It reports one predictive score: Overall pCR likelihood.

How to interpret results:

High likelihood of pCR: Patient is likely to respond to standard neoadjuvant therapy → supports confidence in the treatment plan.
Low likelihood of pCR: Suggests consideration of alternative neoadjuvant regimens or an upfront surgery approach.

Proven clinical impact

Ataraxis Breast NEO accurately predicts neoadjuvant therapy response

External validation on 563 patients from four independent, international datasets

By extracting morphological patterns from a single core needle biopsy H&E, Ataraxis Breast NEO was predictive of pathCR with an AUC of 0.72 (0.65-0.80)

ROC curves comparing true positive rate and false positive rate for UChicago, Providence, Jagiellonian, and IMPRESS datasets.

Ataraxis Breast NEO accurately predicts neoadjuvant therapy response in all subtypes

Bar chart showing AUC ROC values for different patient groups: All NAT-Treated patients 0.721, TNBC (n=190) 0.675, HER2+ (n=197) 0.633, HR+HER2- (n=164) 0.689.

Ataraxis Breast NEO can group patients by their chance of achieving pCR

Bar chart showing increasing PCR rates across four groups: Q1 Bottom 25% at 15%, Q2 at 38%, Q3 at 50%, and Q4 Top 25% at 67%.

Make the most informed call for your patients.

How likely is treatment before surgery to eliminate all visible cancer by the time of surgery?

Personalized insight to guide your first treatment choice

Female doctor showing information on a tablet to a woman patient in a consultation.

If you’ve been diagnosed with breast cancer, your doctor may recommend treatment before surgery, called neoadjuvant therapy. It can be an important step, but it’s often hard to know upfront how well it will work.

Ataraxis Breast NEO uses the biopsy slides already collected during your care, plus routine clinical information, to estimate how your cancer is likely to respond. No new procedures are needed.

Our test can be used for early-stage breast cancer across:

All molecular subtypes, regardless of ER, PR, or HER2 status
Premenopausal women and postmenopausal women
Regardless of cancer presence in lymph nodes

I might want this test if...

My doctor is recommending treatment before surgery, and I want to know my chances of a complete response.

I’m choosing between different pre-surgery treatment plans and want guidance tailored to me.

Keeping or clearing cancer from the lymph nodes is a key concern in my treatment plan.

I’m hoping for breast-conserving surgery, and want to know if treatment is likely to shrink the tumor enough.

I'm in a gray area under standard guidelines, and want a more personalized answer.

I want to avoid unnecessary treatment and side effects if the expected benefit is low.

How do we predict likelihood of achieving pCR?

Data you already have

Our platform uses the existing H&E-stained biopsy or resection slide collected during diagnosis, along with standard clinical variables such as:

Tumor size and nodal status
ER, PR, and HER2 status
Age and histologic subtype

Multimodal AI Analysis

Predictive Output

The test provides a continuous probability (%) of achieving a pathologic complete response (pCR) with neoadjuvant therapy.

It reports one predictive score: Overall pCR likelihood.

How to interpret results:

High likelihood of pCR: Patient is likely to respond to standard neoadjuvant therapy → supports confidence in the treatment plan.
Low likelihood of pCR: Suggests consideration of alternative neoadjuvant regimens or an upfront surgery approach.

What does pCR mean and why is it so important?

Achieving pCR means that after neoadjuvant therapy, no cancer is found in the breast or lymph nodes at surgery. Patients who achieve pCR usually have:

A much lower risk of the cancer returning

Some of the best long-term outcomes

Knowing your chance of pCR upfront can help you and your doctor choose the treatment plan that fits you best.

Two circular illustrations comparing tissue before and after treatment; before treatment shows clustered blue active cancer cells among red cells, after treatment shows only red cells with a checkmark, indicating no cancer found and successful treatment.

Wondering if this test could guide your care? Contact us—we’re happy to help.

Site Map

Legal