Dynamic contrast-enhanced and diffusion MRI show rapid and dramatic changes in tumor microenvironment in response to inhibition of HIF-1alpha using PX-478

Abstract

PX-478 is a new agent known to inhibit the hypoxia-responsive transcription factor, HIF-1alpha, in experimental tumors. The current study was undertaken in preparation for clinical trials to determine which noninvasive imaging endpoint(s) is sensitive to this drug's actions. Dynamic contrast-enhanced (DCE) and diffusion-weighted (DW) magnetic resonance imaging (MRI) were used to monitor acute effects on tumor hemodynamics and cellularity, respectively. Mice bearing human xenografts were treated either with PX-478 or vehicle, and imaged over time. DW imaging was performed at three b values to generate apparent diffusion coefficient of water (ADCw) maps. For DCE-MRI, a macromolecular contrast reagent, BSA-Gd-DTPA, was used to determine vascular permeability and vascular volume fractions. PX-478 induced a dramatic reduction in tumor blood vessel permeability within 2 hours after treatment, which returned to baseline by 48 hours. The anti-VEGF antibody, Avastin, reduced both the permeability and vascular volume. PX-478 had no effect on the perfusion behavior of a drug-resistant tumor system, A-549. Tumor cellularity, estimated from ADCw, was significantly decreased 24 and 36 hours after treatment. This is the earliest significant response of ADC to therapy yet reported. Based on these preclinical findings, both of these imaging endpoints will be included in the clinical trial of PX-478.

Figure 1

DW images at a b value of 25 (up) and corresponding diffusion maps (bottom) of an HT-29 tumor-bearing mouse before, 24 hours, and 48 hours after PX-478 injection. Each image represents an axial slice of the mouse with the tumor area encircled and indicated by an arrow.

Figure 2

Top: Full time course of average tumor ADCw following PX-478 administration (control mice, full line; treated mice, dotted line). A significant increase in average tumor ADCw is observed at 24 and 36 hours posttreatment. Bottom: Summed ADCw histograms of control (filled bars) and treated tumors (open bars) at each time point. A right shift in tumor ADCw is observed at 24 and 48 hours posttreatment.

Figure 3

(A) Permeability maps of tumors 2, 12, 24, and 48 hours after either vehicle (control) or drug (PX-478) injection. Each image represents an axial slice of the mouse with the tumor area encircled. A substantial reduction in tumor vascular permeability is observed as soon as 2 hours after PX-478 injection and until 24 hours, in comparison with the control situation. This is no longer observed by 48 hours after treatment. (B) Vascular volume fraction (VV) maps of tumors 2, 12, 24, and 48 hours after either vehicle (control) or drug (PX-478) injection. Each image represents an axial slice of the mouse with the tumor area encircled. Some individual positive or negative changes can be observed, but these were not significant between groups.

Figure 4

Full time course of average vascular permeability (A) and vascular volume fraction (B) following administration of PX-478 (control mice, full line; treated mice, dotted line). Blood vessel permeability was estimated from the slope of the enhancement curves, and tumor vascular volume (VV) fraction was estimated from the ordinate. A significant reduction in permeability is observed 2, 12, and 24 hours after treatment with PX-478, whereas no changes are observed in the VV fraction.

Figure 5

Summed permeability histograms of control (open, n = 4) and treated tumors (plain, n = 4) at each time point. Note that the median (dotted line) of the treated tumors is lower than the median value of the controls. It is progressively shifted to the median of the controls over time, and is back at control values 48 hours posttreatment.

Figure 6

(A) Relative change in HT-29 tumor vascular permeability and vascular volume fraction 1 hour after treatment with anti-VEGF antibody (Avastin). A significant reduction in permeability as well as in VV fraction is observed with this positive control. (B) Relative change in A-549 tumor (resistant to the antitumor activity of PX-478, negative control) vascular permeability, and vascular volume fraction 2 hours after treatment with PX-478. No significant change is observed in DCE parameters.

Figure 7

HIF-1α levels and antitumor activity of PX-478 in HT-29 human colon cancer and A-549 non small cell lung cancer xenografts in SCID mice. Male SCID mice were injected subcutaneously with (A) 10⁷ HT-29 human colon cancer cells or (B) A-549 non small cell lung cancer cells. The HT-29 tumors were allowed to grow to 400 mm³ and the A-549 tumors to 360 mm³, and treatment was begun with (○) vehicle alone or (■) PX-478 at 80 mg/kg, i.p., daily for 5 days for HT-29 xenografts and 100 mg/kg, i.p., daily for 5 days for A-549 xenografts. Treatment times are shown by arrows. The upper panels show typical immunohistochemical staining for HIF-1α in the untreated tumor xenografts at the start of the study. The lower panels show tumor xenograft growth curves. There were eight mice in each group and bars are SE.