Improved Thyroid Nodule Assessment and Stratification using Optoacoustic-Ultrasound Imaging with Machine Learning (THYNAS+)

This is an early feasibility study exploring the use of OA for thyroid nodule

evaluation using HbR and HbO2 signal. Much is still unknown about the

interpretation of the images in thyroid nodules,therefore, both qualitative and

quantitative analysis will be performed as primary analysis:



1. Images and data will be reviewed first by Investigator Independent Reviewers

(experienced radiologists) to provide a basis for determining feature score

signals that indicate benign or malignant thyroid nodules as well as aggressive

and non-aggressive cancers.

2. A comparison of the in vivo opto-acoustic features on imaging with those

obtained from standard diagnostic imaging will also be done by the Investigator

Independent Reviewers (experienced radiologists).

3. Once thyroid feature scores are developed, Physician Independent Readers

(experienced radiologists) will read the images collected and score the nodules

using the new thyroid feature scoring system to determine whether a nodule is

considered benign or malignant. Descriptive statistics on the feature scores

will be performed as part of the primary analysis.

4. Lymph Nodes will be assessed for pathology correlation with thyroid cancer

histopathology and cytology The thyroid nodule will be marked with a region of

interest (ROI) in the obtained OA images. The signal (HbR vs HbO2) will be

qualitatively interpreted by three independent readers not involved in standard

clinical practice (experienced radiologists), blinded for

cytology/histopathology results. The index nodule will be classified as *not

suspicious* or *suspicious* for malignancy. In addition, quantitative

descriptive statistics will be performed.

5. The diagnostic accuracy of OA will be determined using cytology and

histopathology as golden standard. As such, sensitivity (Se), Sp, positive

predictive value (PPV), negative predictive value (NPV) and accuracy are

calculated.

6. AEs will be summarized for the Safety population. The number and percentage

of subjects with at least one event will be presented overall and by MedDRA

system organ class (SOC) and preferred term (PT). Subjects with more than one

occurrence of an AE within a given SOC and PT will be counted only once toward

that SOC and PT. AE will be summarized in a similar manner under the maximum

severity experienced. Subjects with more than one occurrence of an AE within a

given SOC and PT will be counted only once toward that SOC and PT under the

maximum severity experienced. These analyses will be repeated for SAE, ADE,

SADE, and unanticipated SADE.



For quantitative analysis, descriptive statistics will be used for the ITD

population and for the Safety population in order to assess subject

characteristics, cytopathological/histopathological thyroid nodule and lymph

node characteristics. Other quantitative study parameters collected include:

- Baseline characteristics (sex, age, race, etc.);

- Thyroid nodule size and location;

- OA/US vs US TI-RADS scores;

- Indication for US and FNA;

- Results of clinical US and FNA cytology;

- Preoperative laboratory results (e.g. TSH, T4, T3) (standard of care);

- Histopathological cancer type (papillary, follicular, medullary carcinoma,

etc.).

Primary analyses of study data includes descriptive statistics of the

aforementioned study parameters. Confidence intervals will be calculated as a

two-sided 95% confidence interval where appropriate, unless otherwise

specified.

Secondary analyses includes TI-RADs scores and OA/US thyroid nodule feature

scores stratified by cytology and histopathology type.

In addition to overall (i.e. pooled site data) analyses, summaries of each

study endpoint and other clinical/demographic/safety data will be provided by

study site in order to evaluate the consistency of the results across study

sites as a secondary analysis.

Additional analyses by clinical and/or demographic subgroups may be performed

on the following characteristics:

- Nodule Size

- Nodule Depth

- Age of Patient

- Type of cancer

Two-sided 95% confidence intervals will be calculated where appropriate.