Effects of Anti-Angiogenesis on Glioblastoma Growth and Migration: Model to Clinical Predictions
(a) Multilayer Structure of GBM in the presence of angiogenesis (Full model). (b) Multilayer structure of GBM of the AA model. The last rows of (a) and (b) plot a vertical slice, along the y-axis of the corresponding 4 cell-type concentrations at the final time step. (c) cartoon depicting the cycle of tumor growth and brain invasion that expands necrosis in the absence of angiogenesis. Gray and black areas represent hypoxia and necrosis, respectively. Red balls are cells and blue balls are cells. (d)–(s) are MRIs of 2 GBMs with expanding necrosis on bevacizumab; (d)–(g) and (l)–(o) are Gadolinium-enhanced T1-weighted images (T1/Gad); (h)–(k) and (p)–(s) are FLAIR images (arrows). The first row demonstrates first recurrence before bevacizumab (arrows). The second row shows the maximal effects of bevacizumab. The third row shows expanding necrosis without significant new enhancement (arrows in f and n) but with enlargement of the FLAIR signal (arrows in j and r). The fourth row shows tumor progression on bevacizumab. Clinical details related to (d)–(s) can be found in S1 Text. (t) is a Kaplan-Meier analysis of the progression free survival times (PFS) of patients with (Necrosis +, ) and without (Necrosis -, ) expanding areas of necrosis (Log-Rank ). (u)-(w) show the results of the simulations of the effects of, respectively, AT + PD, AT, and PD on the Full (orange), AA (blue, ie starting from time 0), and treatment (green) models. The latter consists of the Full model until time step = 2500 (first black arrow) when AA is applied and then lifted at time step = 3500 (second black arrow). Time units are arbitrary.