(670b) A New Noninvasive Method for Measuring the Size Distribution of Lung Metastases Vs Time in a Mouse Breast Cancer Model As an Input into a Mathematical Model of Tumor Size Evolution

Cancer is a major problem for patients and physicians alike. One such cancer that is particularly harmful for human health is breast cancer. Breast cancer is one of the leading causes of death among women. Triple negative breast cancer (TNBC) is the most aggressive form of breast cancer. TNBC is also one of the few cancers that has several lines that can cause cancer in mice from a mouse disease. In most other cancer studies a human cancer must be given to a mouse to model the disease. This has the advantage of studying cancer in a more realistic animal model and not eliminating the mouse’s immune system as would be needed to study human cancer in mice. All these reasons are why TNBC is an attractive cancer model for studies.

TNBC is most lethal once it metastasizes. In TNBC a very common metastasis site is the lungs. Once the cancer enters the lungs, breathing becomes challenging. This metastasis pattern in mice mimicking that in humans allows for analyzing and modeling cancer metastasis in these TNBC mouse models to gain a better understanding in cancer patterns and progression.

Here we used a mouse model, 4T1, of TNBC to study metastasis from the mammary gland to the lungs. Our interest (for our theoretical modeling effort) is to obtain tumor numbers and sizes as functions of time in the mouse lungs after tumor initiation in its mamary fat pad. Existing non-invasive techniques do not come close to providing such results. We describe a new non-invasive computed tomography (CT)-based technique to accomplish this. CT, which is basid on tissue density, cannot normally distinguish tumor from normal tissue, other than by shape analysis by an experienced radiologist. Each mouse was given a CT scan on day 0 before inoculation with cancer cells so that its lungs can serve as their own control. After, cancer cell lines were injected into the mammary fat pad of the mouse and allowed to form primary tumors. CT scans were regularly taken of the lungs to detect remote tumor growth from metastasis. Mice were also placed into two different groups, one group whose immune system did not attack the tumor since the tumor cells were syngeneic, and one group that was treated with anti-PD1 an iummune checkpoint inhibitor, to see if such treatment efffected tumor shrinkage and metastasis.

We hypothesize that as cancerous growths progress the amount of tissue in the lungs increases and the amount of air decreases We analyzed time-gated CT images of the lungs sorteed to reflect the same phase of the breathing cycle. All analysis was performed on images of lungs that were fully empty. For each scan, we separated images of the tissue and air portions of the lungs and calculated the size distribution histograms of the volume segments that comprise it. By comparison these size distribution results of post-inoculation and pre-inoculation scans of the same mouse we develop an algorithm for calculating the size distribution of the additional tissue domains that appear as a result of post-inoculation changes to the lungs. We do this for several sequential scans until mouse death and calculate how these distributions change with time. A small number of mice were given a contrast agent consisting of gold bound to sugar that is known to accumulate more in certain nonlung tumors to see if it alone can distinguish tumor from tissue.

We present the data of mouse lungs as TNBC metastasis size/number histograms as a function of time after tumor cell inoculation, both with and without anti-PD1 treatment. These data are precisely what are needed for our group’s mathematical models of how tumor size/number histograms eveolve in time, with the anit-PD1-free mice allowing us to acces parameters for matastasis formation and metastasis growth, and the anti-PD1-treated mice accesing that for the treatment. This is future work. They can be used in future experiments to be predicted by this computational model of TNBC metastasis. Future, experiments will look at mice, e.g., C57 black, for whom these cells are not syngeneic so as to assess their immune response as well as mice undergoing other treatments. Our findings may have implications for future physicians understanding metastasis patterns, the effects of immunotherapy and for incorporating these understandings into treatment plans.