To investigate the effect of inhibiting CA-IX on the pHe of breast cancer patient-derived xenografts (PDX) models.

Numerous prior studies using MRI or MRSI have shown that extracellular pH in tumors is generally acidic. This is important because an acidic pH environment is a potent stimulator of tumor invasion and can reduce anti-tumor immunity¹. It is canonically believed that this acid pH is a result of hyperactive fermentative (glycolytic) metabolism coupled with poor perfusion. However, other studies in 3-D cell culture have also shown membrane-bound carbonic anhydrase (CA-IX) plays an important role in maintaining an acidic extracellular pH (pHe). For example, induction of CA-IX expression led to a reduced pHe and increased pHi of HCT-116 colon cancer spheroids². Further, inhibition of CA-IX in vivo sensitizes HT-129 tumor to radiotherapy³, although the in vivo effects on tumor pH were not known. In this work, we begin an investigation into the role of CA-IX on in vivo pHe, by inhibiting CA-IX ⁴ in breast cancer PDX models, and monitoring effect using ³¹P MRS of 3-aminopropyl phosphonate (3-APP) as pH biomarker.

All the experiments were performed using a 7T horizontal magnet (ASR 310, Agilent Technologies, Inc.) with 205/120/HDS gradients and 310 mm bore, using a 12mm dual-tuned surface coil (Doty Scientific). A 3-APP pH calibration curve was obtained using a set of 200uL phantoms containing 100mM ATP and 100mM 3-APP; pH range was from 5 to 8. Six mice (NOD/SCID), with implanted patient derived breast cancer xenografts, were imaged before the injection of the cpd25 and again at 2.5 hours and 24 hours after injection of the drug. Mice were anesthetized with 2% isoflurane in O₂ and restrained in a specific holder during data acquisition. Anatomical T1 and T2 images of the tumors were obtained with a fast gradient echo sequence with TR/TE 1000/37ms and 2000/76ms, respectively, 7 slices, 1mm slice thickness, FOV=35x35mm. Non-localized single-pulse sequence with a 16o hard pulse, TR=135ms, 4K averages, and 2K points was used at 125MHz to acquired a ³¹P spectra and measure pH in the tumors using the calibration curve obtained with the phantoms. For each imaging time point 200uL of 3-APP 500mM were injected IP 30 minutes before imaging. The cpd25 was injected IP at 100mg/kg concentration.The calibration curve using phantoms shown in Figure 1 was similar to those reported in the literature with pK_a , δ_min and δ_max of 6.7 , 21.29 and 25.05, respectively. Figure 2 shows the pH values for all 6 tumors before cpd25 injection, 2.5h, and 24 hours after injection. Before the injection of the cpd25, half of the tumors presented with high pH, above 7.2, while the other half had low pH, below 6.6. All tumors except mouse 6 increased the pH at 24 hours or as fast as 2.5 hours (mouse 2) after the injection of cpd25. The increase in pH is most likely due to the inhibition of the CA-IX. In Mouse 6, the pH decreased which is unexpected, we hypothesize this could be related to the high pH which was the highest pH of all tumors before the injection of cpd25 (pH=7.29). Tumors have been resected and are being prepared for histology and immunohistochemistry to determine if there are differences in histomorphology of CA-IX expression that could account for the pHe differences. Figure 3 shows ³¹P spectra from three time points of mouse 4. Note that, in this case, the 3-APP peaks (left peaks on the spectra) are split, which is likely due to the presence of 2 different compartments in the tumor with different pH ⁵. For quantification, the peaks were categorized by the intensity into low peak and high peak as shown in table 1. At 24 hours, the high peak is the higher pH value. This is likely due to more tumor cells that had increased pH after 24 hours of the cpd25 injection.In this work we have measured pH in patient derived, triple negative breast cancer xenografts for the first time. Before the injection of the cpd25 50% of the tumors presented high pH above 7.2, while the other 50% had low pH below 6.6. The pH increased in 5 tumors out of 6 in the first 24 hours after the injection of cpd25. These variations may be due to differences in tumor microenvironments, the presence of necrosis, or other factors that will be checked by histology. Studies are now being conducted with engineered cell lines with overexpressed and knocked down CA-IX expression.