Enzymatic targeting of the stroma ablates physical barriers to treatment of pancreatic ductal adenocarcinoma

Abstract

Pancreatic ductal adenocarcinomas (PDAs) are characterized by a robust fibroinflammatory response. We show here that this desmoplastic reaction generates inordinately high interstitial fluid pressures (IFPs), exceeding those previously measured or theorized for solid tumors, and induces vascular collapse, while presenting substantial barriers to perfusion, diffusion, and convection of small molecule therapeutics. We identify hyaluronan, or hyaluronic acid (HA), as the primary matrix determinant of these barriers and show that systemic administration of an enzymatic agent can ablate stromal HA from autochthonous murine PDA, normalize IFP, and re-expand the microvasculature. In combination with the standard chemotherapeutic, gemcitabine, the treatment permanently remodels the tumor microenvironment and consistently achieves objective tumor responses, resulting in a near doubling of overall survival.

Figure 1. Evolution of the desmoplastic reaction in murine (mPDA) and human PDA (hPDA)

(A-E) Masson’s trichrome histochemistry shows robust collagen deposition (blue) at all stages of disease. (F-J) Movat’s pentachrome histochemistry reveals collagen (yellow), GAGs and mucins (blue), and their co-localization (turquoise/green). (K-O) Histochemistry with HA-binding protein (HABP) reveals intense HA content beginning with preinvasive disease (PanIN). (P-T) Activated PSC express α-smooth muscle actin (α-SMA) and are abundant in preinvasive (Q), invasive (R) and metastatic mPDA (S) and hPDA (T), but not in normal pancreata (P). ac, acini; is, islet; d, duct; v, venule; h, hepatic parenchyma; *, metastatic lesions. Scale bars, 25 μm. See also Figure S1.

Figure 2. Elevated IFP compromises vascular function in PDA

(A) Experimental apparatus to measure IFP. (B and C) IFP probe positioned in PDA at the head of the pancreas (hp). In (B) the probe ends in a tumor obscured by the overlying liver. (D) IFP measurements in normal pancreata (N), untreated KPC and KC tumors and tumors 2 and 24 hours post-PEGPH20 treatment. *p<0.01 for difference between normal pancreata and PEGPH20 treatment groups; **p<0.05 for difference from all other groups with 1-way ANOVA and Newman-Keuls post-hoc multiple comparison test. Note that IFP levels in normal pancreata and PEGPH20-treated KC (mean = 16 mmHg, n = 4) and KPC (mean = 22 mmHg, n = 5) tumors 24 hours post-treatment were not statistically different. Box and whisker plots: boxes display the lower (25^th) and upper (75^th) quartiles with a line at the median; whiskers extend from the minimum to the maximum observation. (E-G) Distribution of CD31⁺ vessel diameter in normal pancreata (E), untreated KPC tumors (F), and KPC tumors 24 hours post-PEGPH20 treatment (G). Insets show data for vessels of diameter >10 μm. Data represent evaluations from four independent sections from each of four separate animals for each condition and are significantly different (p<0.0001) in the pairwise comparisons with KPC. (H-J) Multiphoton excitation of fluorescently-conjugated lectin (red) and second harmonic generation imaging of collagen (green) within intact normal pancreas (H), untreated KPC tumors (I), and KPC tumors 24 hours post-PEGPH20 treatment (J). Asterisks in (H) and (J) highlight examples of large functional vessels. Arrows in (I) and (J) indicate perfused (functional) vessels. Note when a rare functional vessel in untreated KPC tumors is identified, it is extremely small relative to those in normal pancreata and PEGPH20 treated tumors. Scale bar, 10 μm for (H-J). (K-M) Direct fluorescence analysis of Alexa-647-conjugated lectin (red) and doxorubicin (green) in intact normal pancreas (K), untreated KPC tumors (L), and KPC tumors 24 hours after PEGPH20 treatment (M). Arrow in (K) shows example of widely patent, thin-walled vessel and arrowheads indicate ductal epithelium. Large arrows in (L) and (M) highlight regions magnified in respective inset boxes. The left inset box in (L) shows signal in the far-red Alexa-647 channel (lectin) alone while the right inset box shows merged green (doxorubicin) and red (lectin) channels. Small arrows in inset box of panel (M) identify examples of specific nuclear staining from DNA-bound doxorubicin, as well as the presence of free doxorubicin in the tumor. Asterisks in (M) highlight examples of large, functional lectin-positive vessels loaded with doxorubicin. ac, acini; is, islet; d, duct. Scale bar, 50 μm for (K-M). See also Figure S2.

Figure 3. Gemcitabine+PEGPH20 combination therapy fundamentally changes the tumor vasculature

(A) Treatment schedule for administration of PEGPH20 (P) or placebo and gemcitabine (G). Mice undergoing gemcitabine monotherapy also received placebo (vehicle) injections. Each 4 week course represents one cycle of therapy. (B) Tumor IFP at survival endpoint after treatment with gemcitabine (G; n=9) or Gem+PEGPH20 (GP; n=9), *p<0.0001. Box and whisker plots: boxes display the lower (25^th) and upper (75^th) quartiles with a line at the median; whiskers extend from the minimum to the maximum observation. (C) Gross pathology of transected PDA at endpoint after gemcitabine monotherapy. (D) Gross pathology of transected PDA at endpoint after combination therapy. (E,F) Distribution of significantly different (p<0.0001) CD31⁺ vessel diameter in Gem (E) and Gem+PEGPH20 (F) treated KPC tumors at survival endpoint. Inset graphs show significantly different (p<0.0001) distribution also for subsets of vessels with diameters between 10-20 μm and >20 μm. Data are from four independent sections from five separate animals for each treatment arm. See also Figure S3.

Figure 4. Gemcitabine+PEGPH20 induces objective responses through reduced proliferation and increased apoptosis

(A) High-resolution ultrasound images of PDA before and after the indicated treatment. Scale bar, 1 mm. t, tumor. (B) Quantitative analysis of tumor volume in Gem (G; n=5) and Gem+PEGPH20 (GP; n=6) treated mice after one cycle of therapy (*p=0.009). Box and whisker plots: boxes display the lower (25^th) and upper (75^th) quartiles with a line at the median; whiskers extend from the minimum to the maximum observation. (C) Quantitative analysis of proliferation assessed as % of Ki-67⁺ cells per image field in untreated (U) and Gem (G) or Gem+PEGPH20 (GP) treated tumors after either two weekly doses of therapy (2 Dose; *p<0.035) or at endpoint (End; *p=0.0009). (D) Quantitative analysis of apoptosis assessed as % cleaved caspase-3 (CC3)⁺ cells per image field in untreated (U) and Gem (G) or Gem+PEGPH20 (GP) treated tumors after two doses of therapy (*p=0.035) or at endpoint (*p=0.01). Data in (C and D) are plotted as mean ± SEM. See also Figure S4.

Figure 5. Combination therapy with Gemcitabine+PEGPH20 remodels the tumor stroma

(A and B) IHC for the PSC activation marker α-SMA in PDA at survival endpoint after the indicated treatment. Scale bar, 50 μm. (C) Quantitative analysis of PSC activation in untreated tumors (U) or tumors treated with Gem (G), PEGPH20 (P) or Gem+PEGPH20 (GP) at survival endpoint (*p=0.005). (D and E) Representative micrographs from intravital SHG imaging of collagen (green) combined with multiphoton excitation of endogenous fluorescence from NADH (red) in PDA at survival endpoint just prior to euthanasia. Scale bar, 50 μm. (F) Mean SHG intensity in PDAs at endpoint as a measure of fibrillar collagen concentration (*p=0.028). (G) Percent area occupied by fibrillar collagen in tumors at endpoint (*p=0.028). Data in (C), (F) and (G) are plotted as mean ± SEM. See also Figure S5.

Figure 6. Gemcitabine+PEGPH20 combination therapy decreases metastatic tumor burden and improves survival

(A and B) HA expression in lung metastases from untreated (A) and PEGPH20-treated (B) animals. (C) Quantitative analyses of proliferation assessed as % Ki-67⁺ cells in metastases to liver (filled circles) and lung (open squares) after gemcitabine monotherapy (G) or Gem+PEGPH20 (GP) combination therapy vs. (p=0.0012). (D) Quantitative analysis of apoptosis assessed as % CC3⁺ cells in liver (filled circles) and lung (open squares) metastases (p=0.0001). (E) Kaplan-Meier survival curves from time of enrollment in control (Con; n=16), Gem (n=16), PEGPH20 (n=15), and Gem+PEGPH20 treated KPC animals (n=14). (Black bar indicates maximum duration of therapy.) Median overall survival of Gem (55.5 days) and Gem+PEGPH20 (91.5) treated mice are significantly different (*p=0.004). Treatment with PEGPH20 alone showed a trend toward increased survival (median = 63 days) that did not reach statistical significance (p=0.1). (F) Metastatic burden in Gem+PEGPH20 (GP) treated mice is significantly decreased compared with Gem treatment alone (G) (*p=0.014). See also Figure S6 and Table S1.

Figure 7. Altering physicomechanics and remodeling the stroma in PDA to therapeutic advantage

(A) Intratumoral mechanics in PDA impede diffusion and convection of small molecules. (B) Enzymatic degradation of stromal HA decreases IFP and relieves physical constraints on small molecule perfusion which can reconstitute in the absence of additional therapy. (C) Combined enzymatic and cytotoxic therapy permanently remodels the tumor microenvironment to favor the delivery and distribution of small molecules. Blue spheres represent chemotherapy molecules, vessels are shown in red, carcinoma cells in light blue, activated PSC in brown, collagen in green, and HA in yellow. P_i, interstitial fluid pressure; P_v, intravascular fluid pressure. See text for details.