A Novel Ultrasound Elastography Technique for Evaluating Tumor Response to Neoadjuvant Chemotherapy in Patients with Locally Advanced Breast Cancer
Abstract
Breast cancer is the second most diagnosed cancer in women, estimated to affect 1 in 8 women during their lifetime. About 10% to 20% of new breast cancer cases are diagnosed with locally advanced breast cancer (LABC). LABC tumors are usually larger than 5 cm and/or attached to the skin or chest wall. It has been reported that when such cases are treated with surgery alone, metastasis and mortality rates are high, especially where skin involvement or attachment to the chest wall is extensive. As such, efficient treatment for this kind of breast cancer includes neoadjuvant chemotherapy (NAC) to shrink the tumor and detach it from the chest wall followed by surgery. Several studies have shown that there is a strong correlation between response to NAC and improved treatment outcomes, including survival rate. Unfortunately, 30% to 40% of patients do not respond to chemotherapy, hence losing critical treatment time and resources. Predicting a patient’s response at the early stages of treatment can help physicians make informed decisions about whether to continue the treatment or use an alternative treatment if a poor response is predicted. Such early and accurate response prediction can shorten the wasted time and reduce resources dedicated to patients while they endure significant side effects. Therefore, it is important to identify this group of non-responder patients as early as possible so that they can be prescribed alternative treatments. Current methods for evaluating LABC response to NAC are based on changes in tumor dimensions using physical examinations or standard anatomical imaging. Such changes may take several months to be detectable. Studies have shown that there is a correlation between LABC response to NAC and tumor softening. In other words, in contrast to responder patients where tumor stiffness generally decreases in response to NAC, in non-responder patients the stiffness of the tumor increases or does not change significantly. As such, a reliable and widely available breast elastography technique can have a major impact on the effective treatment of LABC patients. In this study, we first develop a tissue-mechanics-based method for improving the accuracy of ultrasound elastography. This method consists of 3 steps that are applied to the displacement fields generated from conventional motion-tracking methods. These three steps include: smoothing the displacement fields using Laplacian filtering, enforcing tissue incompressibility equation to refine the displacement fields, and finally enforcing tissue compatibility equation to refine the strain fields. The method was promising through validation using in silico, phantom, and in vivo studies. A huge improvement of this method compared to other motion-tracking methods is its ability in generating lateral displacement with high accuracy. This becomes especially important when the displacement and strain fields are used as inputs to an inverse-problem framework for calculating the stiffness characteristics of tissue, for example, Young’s modulus. We then use this enhanced ultrasound elastography technique to assess the response of LABC patients to NAC based on monitoring the stiffness of their tumors throughout the chemotherapy course. Our results show that this method is effective in predicting patients’ responses accurately as early as 1 week after NAC initiation.