Fig. 6 | Scientific Reports

Fig. 6

From: Predicting the response of triple negative breast cancer to neoadjuvant systemic therapy via biology-based modeling and habitat analysis

Fig. 6

Panel (A) shows distributions of ADC and MSI at V1 and V2 for each of the three habitats for an example patient. Panel (B) shows ADC, MSI, and habitat for the center slice of the same patient’s tumor. Panel C has the same plots as Panel A, but for all tumor bearing voxels across all patients. We use the mean values of each parameter distribution in each habitat to assign habitat meanings. The overall mean ADC and MSI values are 1.3 × 10− 3 mm2/s and 0.12 s− 1, respectively. For habitat 1, the mean ADC value at V1 is lower than the overall mean, and the mean ADC value at V2 is higher than the overall mean. Thus, this habitat has high cellularity at V1 and low cellularity at V2. Using the same logic for the habitat 1 MSI, we assign the label of high V1 and V2 MSI. Taken in concert, habitat 1 is interpreted as a region that is well-perfused and capable of delivering treatment to the tumor effectively (as manifested by the high MSI value at both V1 and V2) which can then effectively kill tumor cells (as manifested by the decrease in cellularity from V1 to V2). Habitat 2 has low V1 and V2 ADC, corresponding to high V1 and V2 cellularity. When combined with the high V1 and low V2 MSI, we interpret habitat 2 as experiencing a reduction in perfusion from V1 to V2, meaning it is increasingly difficult to deliver treatment to this region. As a result, fewer tumor cells are killed, and the cellularity remains high at V2. Habitat 3 has high V1 and V2 ADC, meaning cellularity is consistently low in this habitat. With a low MSI across visits, this habitat is interpreted as poorly perfused, meaning both drug and nutrients may not be effectively delivered to this habitat. Thus, the cellularity remains consistently low, as the tumor is not able to grow or die due to lack of nutrient and drug delivery, respectively.

Back to article page