Cyrcadia Health has developed such an adjunct solution for breast cancer screening called the Cyrcadia Breast Monitor (formerly known as the Circadian Biometric Recorder (CBR^TM )

Scientific Background

The global impact of breast cancer is huge and is increasing. According to the Global Cancer Observatory (GCO) and the International Association of Cancer Registries (IACR), breast cancer is the second most common cancer in the world and the most frequent cancer among women [1].  Therefore, for better survival rate and less aggressive treatments, it is important to detect any abnormal changes in the breast tissue early.

In breast tissue, local circadian clocks function to modulate cellular activity.  In abnormal breast tissue, changes in gene expression patterns, intrinsic factors such as genetic defects or ageing, and extrinsic factors such as irregular shift work disrupt the cell cycle, promoting angiogenesis [2]. Angiogenesis is the biological process that stimulates growth of new blood vessels from pre-existing vessels. An impact of these changes is the disruption of the circadian clocks in breast tissue cells [2,3]. Changes in the body’s central circadian clock modify otherwise normal circadian rhythms (a 24-hour cycle in the biochemical, physiological, or behavioral processes of living entities) [3].

Also, some of these new blood vessels that form due to the process of angiogenesis lack smooth muscle, and hence, do not respond to normal physiological control mechanisms like vasodilation and vasoconstriction [4,5]. This lack of vascular receptivity leads to a more constant blood flow in the abnormal area which alters the skin temperature circadian rhythm [4].

Cyrcadia Health has developed the Cyrcadia Breast Monitor which is a non-invasive, non-compressive and non-radiogenic wearable system that captures these thermo-circadian rhythm alterations over a period of time (four-dimensional system) to detect any early breast tissue abnormalities.

Technology

Cyrcadia Breast Monitor  uses a pair of wearable conformable patches, one for each breast. Each patch has eight embedded digital temperature sensors that transmit temperature data to a Bluetooth enabled Data Recording Device (DRD). In order to be able to be used in all patient population, the patches are available in six different sizes.

To ensure that the system is able to handle large volumes of incoming patient data and report the results back to the user directly, Cyrcadia Breast Monitor utilizes a cloud-based back-end architecture. Cybersecurity and patient privacy are critical issues for devices connected to the internet, and so, the architecture was developed using Microsoft Azure which has passed the strict global data security and privacy regulations.

The DRD stores the measurement data until a local Bluetooth connection can be established with a smartphone (or similar Internet enabled device) which has the Cyrcadia Mobile Application (App) installed. The App receives the recorded data and transfers it to the Local Data Manager for storage and for further transmission to the Core Lab where Cyrcadia’s proprietary patented analytics framework resides. In the Core Lab, the data is analyzed, and the results are sent back to the Local Data Manager and also stored in the Core Lab Central Database. The Local Data Manager reports the results back to the App and the prescribing physician who can then discuss the results with the patient.

Clinical Trials

Initial Clinical Studies

The initial phase I, II and population studies were conducted at three centers – Clem Plam Breast Clinic, La Plata, Argentina, Ohio State University (OSU) and Green Memorial Hospital (GMH), Ohio. In a preliminary analysis using data from the patients enrolled at OSU, it was observed that the system had a sensitivity of 91% (21 out of 23 cancers were detected), whereas the sensitivity of mammography was 83%. It was concluded that “the term ‘false positive’ should be used with caution since the system was positive for three cancers missed by mammography which were either from premenopausal or peri-menopausal women with dense parenchyma, with tumor sizes as small as 0.5 cm as well as micro-calcifications. Patient data which are positive on the system in the absence of mammographic or physical evidence of cancer does not preclude the presence of cancer at its earliest stages. These patients may be considered at “high risk” for the disease which may become clinically evident at a future date.”

Another analysis was conducted with data from all three hospitals. After removing outliers (data affected by loose sensors and out-of-range values), the final dataset had 93 benign and 108 malignant samples. These samples were used for building Cyrcadia’s proprietary patented analytics framework. An accuracy of 78%, sensitivity of 83.56%, and specificity of 71.53% were obtained using this analytics framework. Accuracy is defined as the percentage of benign and malignant samples identified correctly. Sensitivity is defined as percentage of malignant samples correctly identified as malignant by the system, and specificity is the percentage of benign samples correctly identified as benign.

Pilot Studies for Cyrcadia Breast Monitor Validation

A pilot study of 173 patients was designed to validate the results of the initial studies. Enrollment happened at El Camino Hospital, Mountain View, CA and Ohio State University’s James Cancer Hospital, Columbus, Ohio. 66 patients have been enrolled so far, and 47 patients had valid data recordings. An interim analysis was conducted on these samples. An accuracy of 81.82%, sensitivity of 100%, and specificity of 75% were observed.

Updated analytics 

In the first-generation system used in the initial clinical studies, temperature data was acquired at a five-minute sampling rate for 24 hours. The updated second-generation system had the capability of sampling every 10 seconds. In order to verify if the higher sampling rate of 1 minute resulted in getting similar or better accuracies, several analytical experiments were conducted on the data from the pilot studies. An analytical model that can work on 2 hours at 1-minute sampling rate (minimum 6-hour wear time) data was developed. This model resulted in a higher accuracy of 84%, sensitivity of 90.16%, and specificity of 84%.

Performance in dense breasts

Among the 47 women enrolled so far in the pilot studies, 28 were identified as having dense breast tissue.  Of the 28 women with dense breasts, seven were confirmed through biopsy and corresponding histological analysis to have malignant lesions.  Cyrcadia’s technology accurately identified malignancy in 6 of the 7 cases (85.7%). However, mammography had categorized only 3 out of these 7 cases as BI-RADS 5 (42.9%).  This is important considering women with dense breast tissue represents up to 40% of the population.   

References

• 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018; 68(6): 394-424.
• 2. Blakeman V et al.  Circadian Clocks and Breast Cancer.  Breast Cancer Res.  2016; 18(1): 89.
• 3. Keith LG, Oleszczuk JJ, Laguens M. Circadian rhythm chaos: a new breast cancer marker. Int J Fertil Womens Med 2001; 46(5):238-247.
• 4. Farrar WB, Patricia R, Sexton RN, Marsh W, Olsen J. Scientific exhibit presented at the 76th Annual clinical congress of the American college of surgeons. San Francisco, California; An evaluation of a new objective method for breast cancer screening. October 8–11, 1990.
• 5. Filho AL, Lopes JM, Schmitt FC. Angiogenesis and Breast. J Clin Oncol 2010.