Fig 2: ALK inhibition suppresses the HR efficiency.(a, b) Representative images of BRCA1 (a) and CtIP (b) with EdU/DAPI staining in PARP inhibitor (PARPi)-resistant ovarian cancer cells treated with 0.25 µM PARPi (talazoparib) or 0.5 µM ALK inhibitor (ALKi; lorlatinib), either alone or in combination, for 48 hours. Insets, 3.3× magnification. Bar, 20 µm. Data are representative of N = 3 independent experiments with similar results. (c) Quantification of EdU-positive cells with BRCA1 or CtIP foci in PARPi-resistant ovarian cancer cells treated with 0.25 µM PARPi (talazoparib) or 0.5 µM ALKi (lorlatinib), either alone or in combination, for 48 hours. Error bars represent mean ± SD of N = 3 independent experiments, one-way ANOVA analysis, n.s., not significant, Control (Con,?) vs ALK inhibitor (ALKi,?): *P = 0.0235 in BRCA1, ***P < 0.0001 in CtIP; PARP inhibitor + ALK inhibitor (PARPi+ALKi, ¦) vs PARP inhibitor (PARPi, ?): **P = 0.0091 in BRCA1, ***P < 0.0001 in CtIP. (d) Representative images (left) and quantification (right) of EdU positive cells with RAD51 foci in PARPi-sensitive (TNBC) treated with 0.25 µM PARPi (talazoparib) or 0.5 µM ALKi (lorlatinib), either alone or in combination, for 48 h. Insets, 3.3× magnification. Bar, 20 µm. Error bars represent mean ± SD of N = 3 independent experiments, one-way ANOVA analysis, n.s., not significant. (e) Top: Schematic of the DR-GFP reporter assay. Functional HR can repair DNA double-strand breaks and results in the expression of GFP. Bottom: HR efficiency in U2OS DR-GFP cells treated with 0.25 µM PARPi (talazoparib) or 0.5 µM ALKi (lorlatinib), either alone or in combination, for 48 hours. Flow cytometry analysis was used to measure the percentage of GFP + cells. Error bars represent mean ± SEM of N = 3 independent experiments, two-tailed unpaired t test, Control (Con,?) vs ALK inhibitor (ALKi, ¦): *P = 0.0337; PARP inhibitor + ALK inhibitor (PARPi+ALKi, ¦) vs PARP inhibitor (PARPi, ?): *P = 0.0215. (f, g) Gating strategy to determine the percentage of GFP-positive cells in U2OS DR-GFP cells under (f) basal conditions or (g) after transfection with I-SceI expression plasmid. Source data

Fig 3: ALK inhibition demonstrates synergistic effects with PARP inhibitors in vitro.a, IC50 of the PARP inhibitor, talazoparib (left panel) and cisplatin (right panel). Ovarian cancer cells were treated with talazoparib or cisplatin for 6 d and subjected to MTT assay to determine cell viability. Error bars represent mean ± s.e.m. of n = 3 independent experiments, two-tailed, unpaired Student’s t-test. b, Left: quantification of phosphorylation signals of ALK in PARP inhibitor-sensitive and PARP inhibitor-resistant ovarian cancer cells. Cell lysates were analyzed using the Human Phospho-RTK Array Kit following the manufacturer’s instructions. Quantified phosphorylation signal was derived from two antibody spots of ALK. Right: correlations between IC50 of PARP inhibitor (talazoparib) and phosphorylation signals of ALK in PARP inhibitor-/platinum-sensitive and PARP inhibitor-/platinum-resistant ovarian cancer cells (Pearson’s correlation coefficient; P = 0.0084). Independent experiments (n = 3) of MTT assay for calculating the IC50 of PARP inhibitors. c, Correlation between clinical responses to PARP inhibitor (olaparib) and expression of p-ALK (high p-ALK (n = 8 patients) versus low p-ALK (n = 6 patients) in patients with ovarian cancer (two-sided Fisher’s exact test; P = 0.0097). d, Left: representative images and correlation between clinical responses to cisplatin/carboplatin and expression of p-ALK (high p-ALK (n = 9 patients) versus low p-ALK (n = 55 patients) in patients with ovarian cancer (two-sided Fisher’s exact test; P = 0.033). Scale bar, 20 µm. Right: Kaplan–Meier overall survival curves of patients with ovarian cancer, stratified by p-ALK expression levels (high p-ALK (n = 9 patients) versus low p-ALK (n = 55 patients; P = 0.049). e, Cell viability of ALK-knockdown PARP inhibitor-resistant cells treated with the indicated concentration of talazoparib for 6 d. Cell survival is calculated as the percentage relative to the control treatment in each group. OV, ovarian. Data represent n = 3 independent experiments, two-tailed, unpaired Student’s t-test. f, Chou–Talalay analysis of PARP inhibitor-resistant ovarian or TNBC cells treated with varying concentrations of PARP inhibitors (talazoparib or olaparib) and ALK inhibitors (ceritinib or lorlatinib) for 6 d. Synergism showed as CI < 1 at an optimal effect level (Fa > 0.75). The mean percentage of growth inhibition derived from n = 3 independent experiments of the MTT assay was used to calculate the CI value. g, Representative images of clonogenic assay results in PARP inhibitor (PARPi)-resistant ovarian and TNBC cells in the presence of the indicated inhibitor for 12 d. ALKi, ALK inhibitor; LOR, lorlatinib; TALA, talazoparib; Comb, combination of lorlatinib and talazoparib. The mean percentage of growth inhibition derived from n = 3 independent experiments of the clonogenic assay was used to calculate the CI value. Synergistic inhibition of cell proliferation is defined as CI < 1.Source data

Fig 4: Toxicity assay results and body weight measurements of mice treated with PARP inhibitors (PARPi) and ALK inhibitors.(a) Panels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN) and creatinine in mice bearing PARPi-resistant tumors. Mice were treated with oral talazoparib (0.33 mg/kg) and lorlatinib (5 mg/kg), either alone or in combination, five times per week. (b, c) Body weight of mice bearing PARPi-resistant tumors. Mice were treated with oral talazoparib (0.33 mg/kg) and lorlatinib (5 mg/kg), either alone or in combination, five times per week (b), or with oral olaparib (50 mg/kg) and ceritinib (7.5 mg/kg), either alone or in combination, five times per week (c). N = 5 mice were used in each treatment group. Error bars represent the mean ± SD. Statistical analysis was carried out using the two-way ANOVA analysis with no statistical significance specified as n.s. Source data

Fig 5: The combination of ALK and PARP inhibitors effectively suppresses tumor growth in vivo.a–d, Tumor volume and Kaplan–Meier survival curves of mice bearing subcutaneous injected SKOV3 ovarian tumors (a and b) and orthotopic PARP inhibitor-resistant (acquired resistance) SUM149 tumors (nos. 6 and 15; c and d). Mice were treated with oral talazoparib (0.33 mg kg-1) and lorlatinib (5 mg kg-1), either alone or in combination, five times per week (a, c and d), or with oral olaparib (50 mg kg-1) and ceritinib (7.5 mg kg-1), either alone or in combination, five times per week (b). Tumor volume data were reported as mean ± s.e.m. Statistical analysis was carried out using two-tailed, unpaired Student’s t-test (n = 5 mice in each treatment group). e,f, Tumor volume curves (e) and representative images of IHC staining (f) of PARP inhibitor-sensitive SUM149 (parental) tumors from mice with the indicated antibodies. Mice were treated with oral talazoparib (0.33 mg kg-1) five times per week. Scale bar, 20 µm. Tumor volume data were reported as mean ± s.e.m. (n = 5 mice in each treatment group). Statistical analysis was carried out using the two-tailed, unpaired Student’s t-test: **P = 0.0052. g, Representative IHC images stained with indicated antibodies in tumor tissues from mice bearing orthotopic PARP inhibitor-resistant (acquired resistance) SUM149 cells. Data represent images of n = 3 mice. Scale bar, 20 µm. h, Model of PARP inhibitor-resistant mechanism mediated by the ALK–p-Tyr19-CDK9 axis: (1) inactivation of PARP leads to PARP1 trapping on the DNA and increases unrepaired DNA lesions; (2) in PARP inhibitor-resistant cells, p-ALK interacts with and tyrosine phosphorylates CDK9 at Tyr19 to increase the protein stability and kinase activity of CDK9; p-Tyr19-CDK9 regulates formation and nuclear localization of P-TEFb and transcriptionally activates HR-repair genes by phospho-Ser2-RNA Pol II, which in turn contributes to HR proficiency and PARP inhibitor resistance; (3) p-ALK is inhibited by treatment with ALK inhibitors, and the ubiquitination and proteasomal degradation of CDK9 are increased by binding of the E3 ligase Skp2, which in turn blocks the transcription of HR-repair genes and increases PARP inhibitor sensitivity and cell death; and (4) ALK inhibitors induce synthetically lethal PARP inhibitors via induction of HR deficiency.Source data