New Research Finds iCAD's ProFound AI Risk for Digital Breast Tomosynthesis is 2.4x More Accurate than Traditional Lifetime Risk Models

Rhea-AI Summary

iCAD has announced positive results from clinical research supporting its ProFound AI Risk tool for Digital Breast Tomosynthesis (DBT), indicating that 14% of women with a negative screen had a significantly elevated risk of developing breast cancer. The study, involving over 154,200 women, achieved an area under the curve (AUC) of 0.82, underscoring its predictive accuracy compared to traditional models. This advancement aims to refine breast cancer screening strategies, potentially improving early detection and outcomes.

Positive

• ProFound AI Risk tool showed an AUC of 0.82 in predicting breast cancer risk, enhancing detection accuracy.
• 14% of women with negative screens identified as high risk, suggesting significant potential for early intervention.
• Higher sensitivity for cancer detection (59%) compared to traditional models like Tyrer-Cuzick.

Negative

• None.

Using U.S. guidelines, ProFound AI Risk found 14% of women studied who had a negative screen had almost 20 times higher risk of developing breast cancer in the next year than the general risk population

NASHUA, N.H., May 16, 2022 (GLOBE NEWSWIRE) -- iCAD, Inc. (NASDAQ: ICAD), a global medical technology leader providing innovative cancer detection and therapy solutions, today announced that promising clinical research supporting ProFound AI^® Risk for Digital Breast Tomosynthesis (DBT) – the world’s first clinical decision support tool that provides an accurate short-term breast cancer risk estimation based on age, breast density and mammographic features – was recently published in the peer-reviewed journal, Science Translational Medicine.ⁱ

In the study, which involved 154,200 women screened at four participating U.S. screening sitesⁱⁱ using DBT from 2014-2019, researchers at the Karolinska Institutet in Stockholm, Sweden found ProFound AI Risk accurately determined women who were at a higher risk of developing breast cancer, with an area under the receiver operating characteristics curve (AUC) of 0.82^i, (95% CI 0.79-0.85). AUC is a standard performance measurement for examination procedures that incorporates sensitivity and specificity into a single metric of overall performance. This data supports iCAD’s internal research, which previously found ProFound AI Risk for DBT offers an AUC of 0.80 (95% CI 0.76, 0.83).ⁱⁱⁱ

“iCAD’s Breast AI Suite offers a complete portfolio of unrivaled breast cancer detection, density assessment, and short-term risk evaluation AI solutions, and we are pleased to see this compelling data further validate the clinical value of ProFound AI Risk, the latest addition to our breast AI portfolio,” said Stacey Stevens, President and CEO of iCAD, Inc. “Physicians have traditionally estimated breast cancer risk by examining the patient’s known risk factors, such as family history, but about 85% of breast cancers occur in women who have no family history of breast cancer.^iv Additionally, traditional long-term risk models may not be as accurate at estimating a woman’s risk of developing breast cancer, as their average AUC is around 0.60.^v ProFound AI Risk offers a more individualized approach, as it includes a woman’s mammography images and focuses on one-to-two-years in the future, which provides critical information that can help clinicians personalize breast cancer screening regimens for patients based on their individual risk of developing cancer before or at their next screening.”

Established lifestyle-familial risk models, such as Tyrer-Cuzick and Gail, are currently used in the U.S. to identify women with a greater than 20% lifetime risk of developing breast cancer who could be offered breast magnetic resonance imaging (MRI) as a supplemental screening modality for breast cancer detection.ⁱ However, these long-term risk models can result in a high number of false positives due to low-to-moderate discrimination performance.^vi ProFound AI Risk complements traditional risk models and is easy for clinicians and medical facilities to adopt, as it only requires the images from a 2D or 3D mammogram, with no questionnaires, portals, or staff required to implement.

Using U.S. guidelines, the researchers determined that 14% of the women studied were at high risk after a negative or benign screening, with an almost 20-fold higher risk of developing breast cancer, compared to the general risk population. In this high-risk group, 76% of stage II or later cancers, 59% of stage 0, and 58% of stage 1 cancers were observed.

Researchers estimated if the 12% of women at highest risk had been offered supplemental screening based on the ProFound AI Risk for DBT model, up to 59% of the cancers could have potentially been detected, compared to 24% of the cancers using Tyrer-Cuzick. This corresponds to 2.4 times higher sensitivity than Tyrer-Cuzick.

“Our research showed that women who ProFound AI Risk determined to be at high risk were more likely to present with later stage tumors than early-stage cancers,” according to lead author of the study, Mikael Eriksson, PhD, Karolinska Institutet. “It is known that breast cancer survival is four and 12-fold worse for stage II and III cancers compared to stage 0 and I cancers in the first four years after diagnosis. Furthermore, the treatment cost for stage II and III cancers is more than twice that of stage 0 and I cancers in the first 24 months after diagnosis. ProFound AI Risk offers the potential to aid radiologists in refining personalizing screening recommendations and discussing risk with women, which could in turn influence their screening regimen compliance and potentially lead to earlier detection, reduced treatment costs and improved outcomes.”

Researchers also found ProFound AI Risk provided high accuracy in estimating future risk for invasive and in-situ cancers, screen detected and interval cancers, and in women with both dense and non-dense breasts.

“Earlier cancer detection can have a tremendous impact on women, from treatment to outcomes. And because women are often caretakers, improving outcomes in women’s health can also have cascading benefits for children, families, and communities,” added Stevens. “Only iCAD’s Breast AI suite provides clinicians unprecedented insights into each patient’s present and future, which offers the potential to transform the trajectory of a woman’s outcome and life.”

About iCAD, Inc.

Headquartered in Nashua, NH, iCAD^® is a global medical technology leader providing innovative cancer detection and therapy solutions. For more information, visit www.icadmed.com.

Forward-Looking Statements
Certain statements contained in this News Release constitute “forward-looking statements” within the meaning of the Private Securities Litigation Reform Act of 1995, including statements about the expected benefits of ProFound AI^® Risk, the implications of clinical studies, the benefits of the Company’s products, external factors affecting the market for our products, behavior of clients and prospective clients, and future prospects for the Company’s technology platforms and products. Such forward-looking statements involve a number of known and unknown risks, uncertainties and other factors which may cause the actual results, performance, or achievements of the Company to be materially different from any future results, performance, or achievements expressed or implied by such forward-looking statements. Such factors include, but are not limited, to the Company’s ability to achieve business and strategic objectives, the willingness of patients to undergo mammography screening in light of risks of potential exposure to Covid-19, whether mammography screening will be treated as an essential procedure, whether ProFound AI will improve reading efficiency, improve specificity and sensitivity, reduce false positives and otherwise prove to be more beneficial for patients and clinicians, the impact of supply and manufacturing constraints or difficulties on our ability to fulfill our orders, uncertainty of future sales levels, to defend itself in litigation matters, protection of patents and other proprietary rights, product market acceptance, possible technological obsolescence of products, increased competition, government regulation, changes in Medicare or other reimbursement policies, risks relating to our existing and future debt obligations, competitive factors, the effects of a decline in the economy or markets served by the Company; and other risks detailed in the Company’s filings with the Securities and Exchange Commission. The words “believe,” “demonstrate,” “intend,” “expect,” “estimate,” “will,” “continue,” “anticipate,” “likely,” “seek,” and similar expressions identify forward-looking statements. Readers are cautioned not to place undue reliance on those forward-looking statements, which speak only as of the date the statement was made. The Company is under no obligation to provide any updates to any information contained in this release. For additional disclosure regarding these and other risks faced by iCAD, please see the disclosure contained in our public filings with the Securities and Exchange Commission, available on the Investors section of our website at http://www.icadmed.com and on the SEC’s website at http://www.sec.gov.

Media Inquiries:
Jessica Burns, iCAD
+1-201-423-4492
jburns@icadmed.com

Investor Inquiries:
iCAD Investor Relations
ir@icadmed.com

ⁱ Eriksson, M et al. A risk model for digital breast tomosynthesis to predict breast cancer and guide clinical care. Science Translational Medicine. 14 (644). 2022 May 11. Accessed via https://www.science.org/doi/10.1126/scitranslmed.abn3971.
ⁱⁱ Participating U.S. screening sites: Boca Raton Regional Hospital, Boca Raton, FL; Elizabeth Wende Breast Care, Rochester, NY; Larchmont, NJ; Zwanger-Pesiri Radiology, Lindenhurst, NY.
ⁱⁱⁱ iCAD data on file. Variations per vendor and population may occur. ProFound AI Risk is a clinical decision support tool. Information is reviewed by the physician. All care decisions are up to physicians.
^iv Breastcancer.org. U.S. Breast Cancer Statistics. Accessed via https://www.breastcancer.org/symptoms/understand_bc/statistics.
^v Eriksson M, Czene K, Strand F et al. Identification of Women at High Risk of Breast Cancer Who Need Supplemental Screening. Radiology. 2020 Sept 8. Accessed via https://doi.org/10.1148/radiol.2020201620.
^vi M. H. Gail, R. M. Pfeiffer, Breast cancer risk model requirements for counseling, prevention, and screening. J. Natl. Cancer Inst. 110, 994–1002 (2018).

FAQ

What are the results of the recent iCAD study on ProFound AI Risk?

The study found that ProFound AI Risk accurately identifies women at high risk of breast cancer, achieving an AUC of 0.82.

How significant is the risk identified by ProFound AI Risk in the recent iCAD study?

The study indicated that 14% of women screened were at nearly 20 times higher risk of developing breast cancer than the general population.

What percentage of cancers could ProFound AI Risk potentially detect?

Researchers estimated that ProFound AI Risk could detect up to 59% of cancers in the highest risk group, compared to only 24% with the Tyrer-Cuzick model.

What is the impact of ProFound AI Risk on breast cancer screening?

ProFound AI Risk personalizes breast cancer screening regimens, potentially leading to earlier detection and improved patient outcomes.

When was the iCAD study on ProFound AI Risk published?

The study was published on May 11, 2022, in the journal Science Translational Medicine.