Window-of-Opportunity Study of Valproic Acid in Breast Cancer Testing a Gene Expression Biomarker

Purpose

The anticancer activity of valproic acid (VPA) is attributed to the inhibition of histone deacetylase. We previously published the genomically derived sensitivity signature for VPA (GDSS-VPA), a gene expression biomarker that predicts breast cancer sensitivity to VPA in vitro and in vivo. We conducted a window-of-opportunity study that examined the tolerability of VPA and the ability of the GDSS-VPA to predict biologic changes in breast tumors after treatment with VPA.

Patients and Methods

Eligible women had untreated breast cancer with breast tumors larger than 1.5 cm. After a biopsy, women were given VPA for 7 to 12 days, increasing from 30 mg/kg/d orally divided into two doses per day to a maximum of 50 mg/kg/d. After VPA treatment, serum VPA level was measured and then breast surgery or biopsy was performed. Tumor proliferation was assessed by using Ki-67 immunohistochemistry. Histone acetylation of peripheral blood mononuclear cells was assessed by Western blot. Dynamic contrast-enhanced magnetic resonance imaging scans were performed before and after VPA treatment.

Results

Thirty women were evaluable. The median age was 54 years (range, 31-73 years). Fifty-two percent of women tolerated VPA at 50 mg/kg/d, but 10% missed more than two doses as a result of adverse events. Grade 3 adverse events included vomiting and diarrhea (one patient) and fatigue (one patient). The end serum VPA level correlated with a change in histone acetylation of peripheral blood mononuclear cells (ρ = 0.451; P = .024). Fifty percent of women (three of six) with triple-negative breast cancer had a Ki-67 reduction of at least 10% compared with 17% of other women. Women whose tumors had higher GDSS-VPA were more likely to have a Ki-67 decrease of at least 10% (area under the curve, 0.66).

Conclusion

VPA was well tolerated and there was a significant correlation between serum VPA levels and histone acetylation. VPA treatment caused a decrease in proliferation of breast tumors. The genomic biomarker correlated with decreased proliferation. Inhibition of histone deacetylase is a valid strategy for drug development in triple-negative breast cancer using gene expression biomarkers.

Abstract B84: Physiologic pancreatic cancer imaging using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI)

Background: Accurately identifying and monitoring pancreatic ductal adenocarcinoma (PDA) remains a significant challenge. Whereas conventional imaging (CT and MRI) report only anatomic information, dynamic contrast-enhanced MRI (DCE-MRI) has the ability to show physiologic differences in tissue perfusion based on the kinetics of IV-contrast. We coupled DCE-MRI with data-modeling methods to obtain reproducible, high-resolution physiologic images in a series of patients with and without PDA in order to more accurately identify PDA.

Methods: Patients with borderline-resectable PDA (BR-PDA) or at high-risk for PDA (HR), but without cancer were imaged with DCE-MRI. T1-weighted images were collected during IV contrast injections [Figure 1]. Both tumor and the pancreatic body or non-tumor pancreas were identified and using dual-compartment pharmacokinetic modelling, DCE-MRI parameters were quantified and compared between groups.

Results: Between 6/2014 and 2/2015, 12 DCE-MRIs were completed: 9 HR and 3 BR-PDA. Significant differences in the DCE-MRI parameters blood flow (F), first-pass extraction (E), mean capillary transit-time (tc), transfer constant (Ktrans), volume of extravascular-extracellular space (Ve) and blood volume (Vb) were found in normal HR pancreases vs. the tumors in BR-PDA patients [Table 1A]. When comparing non-tumoral pancreases in both cohorts, only F, Ktrans, and Vb differed [Table 1B].

Conclusions: DCE-MRI of the pancreas is feasible and provides imaging biomarkers of vascular structure and function in both tumoral and non-tumoral pancreases which may allow for earlier and more accurate identification of PDA in these high risk populations.

Citation Format: Laura E. Fischer, Matthias Schabel, Bryan Foster, Charles R. Thomas, Jr., William Rooney, Brett C. Sheppard, Erin W. Gilbert.{Authors}. Physiologic pancreatic cancer imaging using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Advances in Science and Clinical Care; 2016 May 12-15; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2016;76(24 Suppl):Abstract nr B84.

Highly accelerated dynamic contrast-enhanced MRI with temporal constrained reconstruction

For DCE MRI applications, the images of adjacent time frames are often similar, especially when motion is minimal, in which case temporal TV is a reasonable regularization term. Temporal constraint reconstruction (TCR) has been developed to reconstruct dynamic images from undersampled k-t space data based on such prior information. However, the convergence speed of the algorithm highly depends on the initialization method. In this study, we study initialization using a composite high resolution image based on a jigsaw sampling pattern during pre-contrast frames. The proposed initialization method converges much faster than a conventional initialization method using low resolution images, especially at high reduction factors. In vivo breast imaging experiments were carried out to evaluate the performance of the proposed method. Experiments show the new initialization method allows TCR to achieve a high reduction factor up to 40 without compromising much of the spatial or temporal resolution. The reconstruction errors are much lower than those using the low resolution initialization when the same number of measurements is used.

Effect of short course, high-dose valproic acid on proliferation in breast adenocarcinomas

e13522

Background: Our in vitro and in vivo studies showed that single agent valproic acid (VPA), a histone deacetylase (HDAC) inhibitor, is effective in a subset of breast cancers. Whether VPA is effective as a single agent and what the optimal dosing would be are unknown. Methods: In an ongoing IRB-approved, window-of-opportunity study testing a genomic predictor of sensitivity to VPA (NCT01007695), we performed biopsies and DCE-MRI on women before and after seven days of escalating doses of VPA on untreated women with breast adenocarcinoma with an intact breast tumor. Within patient dose-escalation was done with a goal dose of 50 mg/kg. Proliferation was assessed with MIB-1 immunohistochemistry. Only correlative secondary endpoints are presented here. Results: Sixteen women have completed treatment and are evaluable. Total and free VPA levels on the last day of treatment averaged 136 µg/ml (range 15-242) and 30 µg/ml (range 1-90), respectively. Proliferation rate by MIB-1 decreased by and average of -7.25% (range +4% to -45%, p=0.04). The change in proliferation did not correlated with VPA levels, although only one of four women with VPA levels less than 100 µg/ml had a decrease in proliferation, compared to five of twelve women with VPA levels greater than 100 µg/ml. Four women had MIB-1 decreases of 10% or greater. Two women have had post-treatment biopsies with fibrosis similar to that seen with chemotherapy treatment effect. DCE-MRI parameters, including ktrans, vp, and kep, did not change significantly with VPA treatment or correlate with VPA levels. Conclusions: VPA as a single agent can reduce proliferation in a subset of women with breast cancers. Higher VPA levels are needed than those achieved at standard dose levels, but VPA levels by themselves do not predict effect. Other monitoring parameters, such as peripheral blood histone acetylation changes, or other predictors of benefit are needed. Clinical trial information: NCT01007695.

Abstract 1679: Combined efficacy of Cediranib and Quinacrine in glioma is enhanced by hypoxia and is associated with autophagic vacuole accumulation

Despite robust vascularity of malignant gliomas, anti-angiogenic therapy largely fails to induce durable responses. We have previously reported that efficacy with Cediranib (Ced), a VEGF/PDGF receptor tyrosine kinase inhibitor, is synergistically enhanced via combination with late-stage autophagy inhibitor quinacrine (Quin), in intracranial 4C8 mouse glioma, resulting in decreased tumor vascularization and growth, increased tumor necrosis and improved mouse survival (#1905, AACR 2012). Our present study was aimed at investigating the role of autophagy (a cellular catabolic pathway that promotes tumor cell survival under hypoxic/nutrient stress), in this synergistic efficacy. MTS assays revealed dose-dependent reductions in cell viability for Ced and Quin under normal (Nrm) and hypoxic (0.5%O2, Hyp) conditions: IC50s(μM) under Nrm and Hyp were 2.7±0.1 and 2.4±0.2 for Ced, and 3.2±0.2 and 2.4±0.04 for Quin, respectively. Greater than additive combined efficacy for Ced+Quin occurred only under Hyp (cell viability reductions for 1μM C + 2.5μM Q: 78±7%(Hyp) vs. 31±3%(Nrm), p<0.05). Western blotting for cleaved caspase 3 also indicated a marked increase in apoptosis with Ced+Quin/Hyp versus all other groups. Western blotting for autophagic vacuole (AV) associated LC3-II protein, which increases with increased autophagic flux and/or late stage autophagic inhibition, not only indicated increases with Quin, but also with Ced, suggesting that Ced may increase autophagic flux. Hypoxia potentiated LC3-II increases in the presence of either Quin or Ced, with the largest increases occurring with Ced+Quin/Hyp, suggesting that hypoxia-induced autophagic flux stimulation combined with late stage autophagic inhibition can trigger AV accumulation and cell death. Bafilomycin A1 (Baf), another late-stage autophagy inhibitor, also decreased cell viability. However, the efficacy of combined Baf and Ced was only additive and lacked substantial potentiation by Hyp (cell viability reductions for 1μM Ced/5ηM Baf: 54±8% (Hyp) vs 37±11% (Nrm)). LC3-II accumulation with Baf was substantially lower than with Quin, either with Hyp or Nrm, suggesting that in addition to inhibiting late-stage autophagy, Quin may also stimulate autophagic flux, consistent with its reported inhibition of PI3K/Akt/mTOR signaling. Thus, a key role for AV accumulation is consistent with its increase with Quin versus Baf in tandem with the improved efficacy of Quin over Baf, in combination with Ced. Our results suggest that the unique cytotoxic efficacy of Ced+Quin we previously reported in vivo, could be associated with increased AV accumulation within hypoxic tumor cells, induced by Ced and possibly Quin, in combination with late-stage autophagic inhibition by Quin. These findings provide a rationale for a careful evaluation of a Ced+Quin combination therapy in patients with malignant gliomas.

Citation Format: Merryl Lobo, Peter Kurre, Matthias Schabel, Yancey Gillespie, Randall Woltjer, Martin Pike. Combined efficacy of Cediranib and Quinacrine in glioma is enhanced by hypoxia and is associated with autophagic vacuole accumulation. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1679. doi:10.1158/1538-7445.AM2013-1679

Quantitative analysis of cytokine-induced vascular toxicity and vascular leak in the mouse brain

A storm of inflammatory cytokines is released during treatment with pro-inflammatory cytokines, such as interleukin-2 (IL-2), closely approximating changes initially observed during sepsis. These signals induce profound changes in neurologic function and cognition. Little is known about the mechanisms involved. We evaluated a number of experimental methods to quantify changes in brain blood vessel integrity in a well-characterized IL-2 treatment mouse model. Measurement of wet versus dry weight and direct measurement of small molecule accumulation (e.g. [(3)H]-H(2)O, sodium fluorescein) were not sensitive or reliable enough to detect small changes in mouse brain vascular permeability. Estimation of brain water content using proton density magnetic resonance imaging (MRI) measurements using a 7T mouse MRI system was sensitive to 1-2% changes in brain water content, but was difficult to reproduce in replicate experiments. Successful techniques included use of immunohistochemistry using specific endothelial markers to identify vasodilation in carefully matched regions of brain parenchyma and dynamic contrast enhanced (DCE) MRI. Both techniques indicated that IL-2 treatment induced vasodilation of the brain blood vessels. DCE MRI further showed a 2-fold increase in the brain blood vessel permeability to gadolinium in IL-2 treated mice compared to controls. Both immunohistochemistry and DCE MRI data suggested that IL-2 induced toxicity in the brain results from vasodilation of the brain blood vessels and increased microvascular permeability, resulting in perivascular edema. These experimental techniques provide us with the tools to further characterize the mechanism responsible for cytokine-induced neuropsychiatric toxicity.

A randomized phase 2 trial of combretatstatin A4 phosphate (CA4P) in combination with paclitaxel and carboplatin to evaluate safety and efficacy in subjects with advanced imageable malignancies

14060

Background: CA4P is the lead compound in a class of agents termed Vascular Disrupting Agents. Phase I studies of CA4P demonstrated toxicities that do not overlap those of chemotherapy. In vitro studies show synergistic cytotoxicity for CA4P and chemotherapy. This study was designed to assess safety, toxicity and tumor perfusion of CA4P at 2 doses when combined with paclitaxel and carboplatin at fixed doses. Methods: A one stage, randomized trial for subjects with imageable tumors by DCE-MRI for perfusion. Subjects received either 45 or 63 mg/m2 CA4P IV on Days 1, 8, and 15, then 200 mg/m2 paclitaxel and AUC 6 carboplatin on Day 2 of each 21 day cycle. After completion of 6 cycles, subjects could continue with CA4P alone. DCE-MRI was measured prior and 24 hours after the first dose of CA4P. Results: 13 subjects were accrued from 3/05–11/05. 78 cycles were administered and 1 subject continues active therapy on cycle 19. The median number of cycles administered was 6. Both 45 and 63 CA4P with paclitaxel/carboplatin were well tolerated. The most frequent adverse events are listed in Table 1 . Grade 3–4 AEs were similar between the two dose groups. Table 1 : Most Frequent AEs by CA4P Dose DCE-MRI data showed a mean reduction in blood flow for both dose groups (46% vs. 19% respectively) confirming CA4P’s tumor vascular disrupting capabilities. Tumor responses were observed and were similar between both dose groups. Best overall response through cycle 6 was 3 PR, and 6 SD. Two subjects with thyroid cancer were randomized to the low dose group and completed cycle 6. The 1st had SD and progressed after cycle 6. The 2nd achieved PR and progressed after 8 cycles. These subjects also had the greatest reduction in tumor blood flow. Ktrans was reduced by 73 and 79% 24 hours post CA4P. Conclusion: These doses of CA4P in combination with paclitaxel and carboplatin were well tolerated, demonstrated anti-tumor activity and reduction of tumor blood flow.

[Table: see text]

No significant financial relationships to disclose.

Noise propagation in linear and nonlinear inverse scattering

The propagation of noise from the data to the reconstructed speed of sound image by inverse scattering within the framework of the Lippmann-Schwinger integral equation of scattering is discussed. The inversion algorithm that was used consisted in minimizing a Tikhonov functional in the unknown speed of sound. The gradient of the objective functional was computed by the method of the adjoint fields. An analytical expression for the inverse scattering covariance matrix of the image noise was derived. It was shown that the covariance matrix in the linear x-ray computed tomography is a special case of the inverse scattering matrix derived in this paper. The matrix was also analyzed in the limit of the linearized Born approximation, and the results were found to be in qualitative agreement with those recently reported in the literature for Born inversion using filtered backpropagation algorithm. Finally, the applicability of the analysis reported here to the obstacle problem and the physical optics approximation was discussed.

Pharmacokinetics and tissue retention of (Gd-DTPA)-cystamine copolymers, a biodegradable macromolecular magnetic resonance imaging contrast agent

PURPOSE

To investigate the pharmacokinetics, long-term tissue retention of Gd(III) ions, and magnetic resonance imaging (MRI) contrast enhancement of extracellular biodegradable macromolecular Gd(III) complexes, (Gd-DTPA)-cystamine copolymers (GDCC), of different molecular weights.

METHODS

The pharmacokinetics of blood clearance and long-term Gd(III) retention of GDCC were investigated in Sprague-Dawley rats. Pharmacokinetic parameters were calculated by using a two-compartment model. The blood pool contrast enhancement of GDCC was evaluated in Sprague-Dawley rats on a Siemens Trio 3T MR scanner. Gd-(DTPA-BMA) was used as a control.

RESULTS

The alpha phase half-life of Gd-(DTPA-BMA) and GDCC with molecular weights of 18,000 (GDCC-18) and 60,000 Da (GDCC-60) was 0.48 +/- 0.16 min, 1.08 +/- 0.24 min, and 1.74 +/- 0.57 min, and the beta phase half-life was 21.2 +/- 5.5 min, 26.5 +/- 5.9 min, and 53.7 +/- 15.9 min, respectively. GDCC had minimal long-term Gd tissue retention comparable to that of Gd-(DTPA-BMA). GDCC resulted in more significant contrast enhancement in the blood pool than Gd-(DTPA-BMA).

CONCLUSIONS

GDCC provides a prolonged blood pool retention time for effective MRI contrast enhancement and then clears rapidly with minimal accumulation of Gd (III) ions. It is promising for further development as a blood pool MRI contrast agent.

Influence of satellite data uncertainties on the detection of externally forced climate change

Two independent analyses of the same satellite-based radiative emissions data yield tropospheric temperature trends that differ by 0.1 degrees C per decade over 1979 to 2001. The troposphere warms appreciably in one satellite data set, while the other data set shows little overall change. These satellite data uncertainties are important in studies seeking to identify human effects on climate. A model-predicted "fingerprint" of combined anthropogenic and natural effects is statistically detectable only in the satellite data set with a warming troposphere. Our findings show that claimed inconsistencies between model predictions and satellite tropospheric temperature data (and between the latter and surface data) may be an artifact of data uncertainties.