Targeting PARP-1 with Alpha-Particles Is Potently Cytotoxic to Human Neuroblastoma in Preclinical Models

Abstract

Alpha-emitters can be pharmacologically delivered for irradiation of single cancer cells, but cellular lethality could be further enhanced by targeting alpha-emitters directly to the nucleus. PARP-1 is a druggable protein in the nucleus that is overexpressed in neuroblastoma compared with normal tissues and is associated with decreased survival in high-risk patients. To exploit this, we have functionalized a PARP inhibitor (PARPi) with an alpha-emitter astatine-211. This approach offers enhanced cytotoxicity from conventional PARPis by not requiring enzymatic inhibition of PARP-1 to elicit DNA damage; instead, the alpha-particle directly induces multiple double-strand DNA breaks across the particle track. Here, we explored the efficacy of [²¹¹At]MM4 in multiple cancers and found neuroblastoma to be highly sensitive in vitro and in vivo Furthermore, alpha-particles delivered to neuroblastoma show antitumor effects and durable responses in a neuroblastoma xenograft model, especially when administered in a fractionated regimen. This work provides the preclinical proof of concept for an alpha-emitting drug conjugate that directly targets cancer chromatin as a therapeutic approach for neuroblastoma and perhaps other cancers.

Figure 1:

Expression patterns of PARP-1 at the mRNA and protein level in normal pediatric tissues and neuroblastoma. A) PARP-1 mRNA is elevated in neuroblastoma vs. normal pediatric tissues allowing for the selective delivery of PARP-1 targeted radiopharmaceutical therapeutic. B) Two publicly available databases showed elevated PARP-1 mRNA expression was positively correlated with high-risk and worse prognosis in patients with neuroblastoma, highlighting the need and opportunity for developing a PARP-1 targeted radiopharmaceutical therapeutic. C) A single punch mini array from 23 clinical cases of neuroblastoma confirmed that PARP-1 expression is variable and that its overexpression can be targeted for delivery of alpha-emitting radiopharmaceutical therapeutic.

Figure 2:

In vitro studies evaluating [²¹¹At]MM4 cytotoxicity and DNA damage in neuroblastoma cell lines. A) Waterfall plot of EC₅₀ values for a panel of 19 neuroblastoma cell lines that were used to test the in vitro cytotoxicity of [²¹¹At]MM4 vs. controls: free astatine-211 ([²¹¹At]NaAt^x), and [²¹¹At]NaAt^x with non-radioactive analogue PARP inhibitor KX1 (ANOVA p-value < 0.0001 for [²¹¹At]MM4 vs. controls in all cell lines). B) The comparison of EC₅₀ values in molar units for [²¹¹At]MM4 vs. KX1 (ANOVA p- value < 0.0001 for all cell lines). C) Neuroblastoma cell line radiosensitivity correlation between [²¹¹At]MM4 and [²¹¹At]NaAt^x (linear regression R² = 0.708, p-value <0.0001 for non-zero slope). D) Neuroblastoma cell line radiosensitivity correlation between [²¹¹At]MM4 and [²¹¹At]MM4 normalized with [²¹¹At]NaAt^x (linear regression R² = 0.400, p-value <0.0033 for non-zero slope). E) Immunofluorescence of γH2AX and PARP-1 after 24 h treatment with [²¹¹At]MM4. PARP-1 was increased (t-test, p-value < 0.001) in SK-N-BE(2)-C cells and both cell lines showed increased γH2AX (t-test, p-value < 0.001). F) NLF cells treated with [²¹¹At]MM4 were analyzed by flow cytometry for ATM and H2AX phosphorylation. There was a 98% increase in ATM and H2AX phosphorylation in treated cells from control indicating double strand DNA breaks were the major form of DNA damage. G) Cell cycle analysis of NLF cells treated cells showed accumulation at the G2/M boundary.

Figure 3:

In vivo biodistribution and ex vivo autoradiography of [²¹¹At]MM4 in IMR-05 tumor bearing mice. A) Radioactivity measured in organs and tumor (dashed-line box) at time points of 2 min, 1 and 2 h post injection. B) Average relative uptake in organs compared to tumor at 2 h showed higher activity in tumor compared to all organs except liver and stomach (p-value < 0.05). C) Ex vivo autoradiographs of tumor vs. muscle showed a tumor:muscle ratio of 10 (p-value < 0.05).

Figure 4:

In vivo efficacy studies evaluating single and fractionated dosing regimens of [²¹¹At]MM4 in an IMR-05 xenograft mouse model. Tumor growth and Kaplan-Meier curves for IMR-05 tumor bearing mice treated with A) single dose of 555 or 1,110 kBq of [²¹¹At]MM4 (control vs. 555 kBq and 1110 kBq mixed linear model p-value < 0.0001; control vs. 555 kBq and 1110 kBq survival mantel cox test p-value < 0.0001, 555 KBq vs. 1110 kBq not significant), and B) single high dose of 1480 kBq vs. a fractionated dose of 370 kBq twice weekly for a cumulative dose of 1480 kBq (control vs. fractionated mixed linear model p-value < 0.0001, fractionated vs. high dose not significant; survival mantel cox test high dose vs. control p-value < 0.0001, fractionated vs. control p-value = 0.03). C) Toluadine blue stains on control vs. fractionated therapy treated tumor sections. D) Control vs. treated tumor from the in vivo efficacy study evaluating fractionated vs. single high dose therapy. Immunofluorescence on tumor sections were performed for PARP-1 and Ki-67. E) Tumor sections fluorescently stained for PARP-1 taken from IMR-05 tumor bearing mice treated with 370 kBq of [²¹¹At]MM4 at 24, 48, and 72 h after treatment (t-test, p-value < 0.001 (24 h), 0.05 (72 h). * denote statistical significance described in the figure legend.

Figure 5:

The positive feedback of DNA damage induced by [²¹¹At]MM4 increases PARP-1 expression and enhances drug targeting with fractionated dosing regimens.