Loss of 53BP1 causes PARP inhibitor resistance in Brca1-mutated mouse mammary tumors

Abstract

Inhibition of PARP is a promising therapeutic strategy for homologous recombination-deficient tumors, such as BRCA1-associated cancers. We previously reported that BRCA1-deficient mouse mammary tumors may acquire resistance to the clinical PARP inhibitor (PARPi) olaparib through activation of the P-glycoprotein drug efflux transporter. Here, we show that tumor-specific genetic inactivation of P-glycoprotein increases the long-term response of BRCA1-deficient mouse mammary tumors to olaparib, but these tumors eventually developed PARPi resistance. In a fraction of cases, this resistance is caused by partial restoration of homologous recombination due to somatic loss of 53BP1. Importantly, PARPi resistance was minimized by long-term treatment with the novel PARP inhibitor AZD2461, which is a poor P-glycoprotein substrate. Together, our data suggest that restoration of homologous recombination is an important mechanism for PARPi resistance in BRCA1-deficient mammary tumors and that the risk of relapse of BRCA1-deficient tumors can be effectively minimized by using optimized PARP inhibitors.

Significance: In this study, we show that loss of 53BP1 causes resistance to PARP inhibition in mouse mammary tumors that are deficient in BRCA1. We hypothesize that low expression or absence of 53BP1 also reduces the response of patients with BRCA1-deficient tumors to PARP inhibitors.

Figure 1.

Acquired resistance of Pgp-deficient Brca1^Δ/Δ;p53^Δ/Δ (KB1PM) mouse mammary tumors to the PARPi olaparib. A, Kaplan–Meier curve showing survival of mice bearing Pgp-proficient KB1P or Pgp-deficient KB1PM tumors, either untreated or treated with 50 mg/kg olaparib given intraperitoneally for 28 consecutive days. Treatment was resumed when a relapsing tumor reached a size of 100% (the tumor size at the start of the treatment). Individual tumor responses are shown in Supplementary Fig. S1. The Gehan–Breslow–Wilcoxon P value is indicated. B, response to daily treatment with 50 mg/kg olaparib given intraperitoneally of olaparib-resistant tumors from 3 donor tumors (KB1PM1, 3, and 4) and drug-naïve control tumors from the corresponding donors. C, levels of PAR detected in whole-tumor extracts from olaparib-resistant and control tumors derived from KB1PM1, KB1PM3, and KB1PM4. The tumors were harvested without treatment, 30 minutes after one dose of 50 mg/kg olaparib given intraperitoneally, or 2 hours after the last dose of 7 days of daily treatment. n.d., not detectable (lower than 2*SD above background). Data are presented as mean + SD of 3 mice per donor per treatment.

Figure 2.

Response of olaparib-resistant tumors to DNA-damaging agents. A, classification of the response of olaparib-resistant (Res) and control tumors (Ctr) from 5 individual donors (KB1PM1, 3, 4, 5, and 8) to olaparib (50 mg/kg, daily for 28 days), topotecan (2 mg/kg, days 0–4 and 14–18), cisplatin (6 mg/kg, day 0), and doxorubicin (5 mg/kg, days 0, 7, and 14). Untreated tumors would be classified as “poor responders.” B –D, for the same group of mice, the relapse-free survival is shown in response to topotecan, cisplatin, or doxorubicin. The poor and intermediate responders of A have a relapse-free survival of 0 days, as the tumor did not shrink below 50% of the original size. Day 0 is the start of the treatment. The Gehan–Breslow–Wilcoxon P values are indicated.

Figure 3.

Loss of 53BP1 protein in olaparib-resistant tumors. A, 53BP1 in control and olaparib-resistant KB1PM tumors. Depicted are 2 individual tumors with loss of 53BP1 and one of the olaparib-resistant tumors that still expresses 53BP1. The 53BP1-positive cells in the resistant tumors of KB1PM5 and KB1PM8 are either stroma or a duct from the wild-type host mammary gland. B, olaparib-resistant tumor KB1PM3 shows focal loss of 53BP1 expression. C, cDNA sequencing revealed a duplication of exons 25 and 26 in Trp53bp1, resulting in a frame shift and premature stop codon in the olaparib-resistant tumor cells of KB1PM5. The alternating codons are underlined and the premature stop codon is indicated in bold. D, identification of a heterozygous truncating mutation Q626* in exon 12 Trp53bp1 in the 53BP1-negative areas of olaparib-resistant (ol-res) tumor KB1PM8. cDNA sequencing identified only the mutated allele. Primers that were used for KB1PM5 and KB1PM8 are listed in Supplementary Table S2. FFPE, formalin-fixed, paraffin-embedded. Scale bar, 100 μm.

Figure 4.

DNA damage foci in irradiated cells. A, detection of ionizing radiation–induced RAD51, 53BP1, and γ-H2AX foci in BRCA1-proficient KP3.33 cells (5), a cell line derived from a KB1PM5 control tumor (control 1) and a 53BP1-negative cell line derived from the olaparib-resistant KB1PM5 tumor (ol-res 1). For characterization of the cell lines, see also Supplementary Fig. S5. Images show the maximum projection, covering the whole cell in the z-direction. B, quantification of RAD51 focus formation of 3 KB1PM5 control cell lines and 3 KB1PM5 olaparib-resistant cell lines in 2 independent experiments. The 3 control cell lines are combined in one bar, and each olaparib-resistant cell line is shown separately for both experiments. C, quantification of RAD51 IRIFs in a control KP tumor and in matched olaparib-sensitive and -resistant KB1PM tumors. See also Supplementary Fig. S6. IR, ionizing radiation.

Figure 5.

The effect of 53BP1 loss on olaparib sensitivity. A, Western blot analysis showing 53BP1 levels in KB1P-B11 and KB1P-G3 cells that express a nontargeting hairpin (NT) or a hairpin against Trp53bp1. B, clonogenic assay with olaparib. The IC₅₀ is indicated between brackets. C, Western blot analysis showing the reconstitution of 53BP1 in 53BP1-deficient KB1P-3.12 cells. KP cells are used as positive control for the BRCA1 Western blot analysis. D, clonogenic assay of 53BP1-negative KB1P-3.12 cells, h53BP1-reconstituted KB1P-3.12 cells, and BRCA-proficient KP cells. E, overall survival of mice with a 53BP1-positive (shNT) or -negative (sh53BP1) tumor treated with one regimen of olaparib daily for 28 days and the untreated control mice. 53BP1 expression in these tumors is shown in Supplementary Fig. S7C. F, relapse-free survival of mice with a 53BP1-positive (shNT) or negative (sh53BP1) tumor treated with one dose of cisplatin. The Gehan–Breslow–Wilcoxon P values are indicated.

Figure 6.

Non–Pgp-mediated resistance to the next generation PARPi AZD2461. A, overall survival of mice with an olaparib-resistant KB1P tumor, with 80-fold increase in Mdr1b expression, that were treated with the vehicle of AZD2461 (0.5% v/w HPMC), olaparib, olaparib in combination with the Pgp inhibitor tariquidar, or with AZD2461. B, tumor growth of the Pgp-deficient, olaparib-resistant tumors from KB1PM1, KB1PM3, and KB1PM4, either untreated or treated with AZD2461. C, Kaplan–Meier curve showing survival of mice with a Pgp-proficient KB1P tumor, either untreated or after treatment with olaparib or AZD2461. Individual tumor responses are shown in Supplementary Figs. S1C and S9C. The Gehan–Breslow–Wilcoxon P value is indicated. D, 53BP1 in control and AZD2461-resistant KB1P tumors. Depicted are 3 individual tumors with loss of 53BP1 and one of the AZD2461-resistant tumors that still express 53BP1. Scale bar, 100 μm. E, identification of a 94-bp deletion in exon 21 of Trp53bp1 in AZD2461-resistant tumor KB1P2, leading to a frame shift and early stop codon in exon 21. F, identification of a 34-bp deletion at the splice acceptor site of exon 25 of Trp53bp1.

Figure 7.

Chronic antitumor efficacy of the novel PARPi AZD2461. A, overall survival of mice receiving 100 days of olaparib (see also ref. 9) or AZD2461 treatment. After relapse of a tumor to a size of 100%, treatment was given for another 100 days. Individual tumor responses are shown in Supplementary Fig. S10A–S10C. B, relapse-free survival of mice with a KB1P tumor that received AZD2461 for 28 consecutive days or for 100 consecutive days. The Gehan–Breslow–Wilcoxon P values are indicated.