Single-cell DNA-seq depicts clonal evolution of multiple driver alterations in osimertinib-resistant patients.

BACKGROUND: The development of targeted agents, such as osimertinib for EGFR-mutated non-small-cell lung cancer (NSCLC), has drastically improved patient outcome, but tumor resistance eventually always occurs. In osimertinib-resistant NSCLC, the emergence of a second molecular driver alteration (such as ALK, RET, FGFR3 fusions or BRAF, KRAS mutations) has been described. Whether those alterations and the activating EGFR mutations occur within a single cancer cell or in distinct cell populations is largely debated. PATIENTS AND METHODS: Tumor sequencing was used to identify the acquired resistance mechanisms to osimertinib in the MATCH-R trial (NCT0251782). We implemented single-cell next-generation sequencing to investigate tumor heterogeneity on patient's frozen tissues in which multiple alterations have been identified. Patient-derived models, cell lines, and patient-derived xenografts were exposed to specific inhibitors to investigate combination treatment strategies. RESULTS: Among the 45 patients included in MATCH-R who progressed on osimertinib, 9 developed a second targetable alteration (n = 2 FGFR3-TACC3, n = 1 KIF5B-RET, n = 1 STRN-ALK fusions; n = 2 BRAFV600E, n = 1 KRASG12V, n = 1 KRASG12R, n = 1 KRASG12D mutations). Single-cell analysis revealed that the two driver alterations coexist within one single cancer cell in the four patients whose frozen samples were fully contributive. A high degree of heterogeneity within samples and sequential acquisitions of molecular events were highlighted. A combination treatment concomitantly targeting the two driver alterations was required on the corresponding patient-derived models to restore cell sensitivity, which was consistent with clinical data showing efficacy of brigatinib in the patient with ALK fusion after progression to osimertinib and crizotinib administered sequentially. CONCLUSIONS: Distinct molecular driver alterations at osimertinib resistance coexist with initial EGFR mutations in single cancer cells. The clonal evolution of cancer cell populations emphasized their heterogeneity leading to osimertinib relapse. Combining two targeted treatments is effective to achieve clinical benefit.

Predictors of chemotherapy efficacy in non-small-cell lung cancer: a challenging landscape.

BACKGROUND: Conventional cytotoxic chemotherapy (CCC) is the backbone of non-small-cell lung cancer (NSCLC) treatment since decades and still represents a key element of the therapeutic armamentarium. Contrary to molecularly targeted therapies and immune therapies, for which predictive biomarkers of activity have been actively looked for and developed in parallel to the drug development process ('companion biomarkers'), no patient selection biomarker is currently available for CCC, precluding customizing treatment. MATERIALS AND METHODS: We reviewed preclinical and clinical studies that assessed potential predictive biomarkers of CCC used in NSCLC (platinum, antimetabolites, topoisomerase inhibitors, and spindle poisons). Biomarker evaluation method, analytical validity, and robustness are described and challenged for each biomarker. RESULTS: The best-validated predictive biomarkers for efficacy are currently ERCC1, RRM1, and TS for platinum agents, gemcitabine and pemetrexed, respectively. Other potential biomarkers include hENT1 for gemcitabine, class III ß-tubulin for spindle poisons, TOP2A expression and CEP17 duplication (mostly studied for predicting anthracyclines efficacy) whose applicability concerning etoposide would deserve further evaluation. However, none of these biomarkers has till now been validated prospectively in an appropriately designed and powered randomised trial, and none of them is currently ready for implementation in routine clinical practice. CONCLUSION: The search for predictive biomarkers to CCC has been proven challenging. If a plethora of biomarkers have been evaluated either in the preclinical or in the clinical setting, none of them is ready for clinical implementation yet. Considering that most mechanisms of resistance or sensitivity to CCC are multifactorial, a combinatorial approach might be relevant and further efforts are required.

Negative prognostic value of high levels of intracellular poly(ADP-ribose) in non-small cell lung cancer.

BACKGROUND: Cisplatin-resistant non-small cell lung cancer (NSCLC) cells are often characterized by alterations in vitamin B-related metabolic processes, including the overexpression and hyperactivation of poly(ADP-ribose) polymerase 1 (PARP1) and the downregulation of pyridoxal kinase (PDXK), correlating with elevated apoptosis resistance. Low PDXK expression is an established negative prognostic factor in NSCLC. PATIENTS AND METHODS: We determined by immunohistochemistry the expression of PARP1 and the level of its product, poly(ADP-ribose) (PAR), in two independent cohorts of patients with resected NSCLC. RESULTS: Intratumoral high levels (above median) of PAR (but not PARP1 protein levels) had a negative prognostic impact in both the training (92 stage I subjects) and validation (133 stage I and II subjects) cohorts, as determined by univariate and multivariate analyses. The simultaneous assessment of PAR and PDXK protein levels improved risk stratification. CONCLUSION: NSCLC patients with high intratumoral PARP1 activity (i.e. elevated PAR levels above median) and low PDXK expression (below median) had a dismal prognosis, while patients with low PARP1 activity and high PDXK expression had a favorable outcome. Altogether, these results underscore the clinical potential and possible therapeutic relevance of these biomarkers.

A high-throughput screen identifies PARP1/2 inhibitors as a potential therapy for ERCC1-deficient non-small cell lung cancer.

Excision repair cross-complementation group 1 (ERCC1) is a DNA repair enzyme that is frequently defective in non-small cell lung cancer (NSCLC). Although low ERCC1 expression correlates with platinum sensitivity, the clinical effectiveness of platinum therapy is limited, highlighting the need for alternative treatment strategies. To discover new mechanism-based therapeutic strategies for ERCC1-defective tumours, we performed high-throughput drug screens in an isogenic NSCLC model of ERCC1 deficiency and dissected the mechanism underlying ERCC1-selective effects by studying molecular biomarkers of tumour cell response. The high-throughput screens identified multiple clinical poly (ADP-ribose) polymerase 1 and 2 (PARP1/2) inhibitors, such as olaparib (AZD-2281), niraparib (MK-4827) and BMN 673, as being selective for ERCC1 deficiency. We observed that ERCC1-deficient cells displayed a significant delay in double-strand break repair associated with a profound and prolonged G2/M arrest following PARP1/2 inhibitor treatment. Importantly, we found that ERCC1 isoform 202, which has recently been shown to mediate platinum sensitivity, also modulated PARP1/2 sensitivity. A PARP1/2 inhibitor-synthetic lethal siRNA screen revealed that ERCC1 deficiency was epistatic with homologous recombination deficiency. However, ERCC1-deficient cells did not display a defect in RAD51 foci formation, suggesting that ERCC1 might be required to process PARP1/2 inhibitor-induced DNA lesions before DNA strand invasion. PARP1 silencing restored PARP1/2 inhibitor resistance in ERCC1-deficient cells but had no effect in ERCC1-proficient cells, supporting the hypothesis that PARP1 might be required for the ERCC1 selectivity of PARP1/2 inhibitors. This study suggests that PARP1/2 inhibitors as a monotherapy could represent a novel therapeutic strategy for NSCLC patients with ERCC1-deficient tumours.

Cisplatin benefit is predicted by immunohistochemical analysis of DNA repair proteins in squamous cell carcinoma but not adenocarcinoma: theranostic modeling by NSCLC constituent histological subclasses.

BACKGROUND: Most non-small-cell lung cancer (NSCLC) patients receive cisplatin-based chemotherapy though clinical response is restricted to a subset of patients. DNA repair protein levels are possible surrogates for cisplatin-induced DNA adduct (and subsequent cell death) repair efficiency and thus molecular determinants of therapeutic efficacy. The International Adjuvant Lung Trial (IALT)-Bio study previously suggested ERCC1 and MSH2 as predictive of cisplatin-based therapeutic benefit. PATIENTS AND METHODS: DNA repair protein expression (XPF, BRCA1, ERCC1, MSH2, p53, PARP1, and ATM) was assessed by immunohistochemistry on a large subset of patients (N = 769) from the IALT trial. Tissue Microarray slides were digitally scanned and signal quantified by user-defined macros. Statistical analyses (univariate and multivariate) of 5-year disease-free survival (DFS) and 5-year overall survival used binary cut-offs (H score low/high expression). RESULTS: In patients with squamous cell carcinoma (SCC), ATM, p53, PARP1, ERCC1, and MSH2 displayed significant (borderline) predictive values, mainly on DFS with chemotherapy efficacy limited to low marker levels. Adenocarcinoma (ADC) results were not significant. BRCA1 and XPF were not significant for predictive modeling in either SCC or ADCs. CONCLUSION: Here predictive utility of DNA repair enzymes co-segregates with SCC histology, focusing their predictive value to this histological subclass of NSCLC. Distinct mechanisms of chemotherapeutic response or resistance might exist among histological subclasses of solid tumors.

ERCC1 influence on the incidence of brain metastases in patients with non-squamous NSCLC treated with adjuvant cisplatin-based chemotherapy.

BACKGROUND: We recently demonstrated that excision repair cross-complementing group 1 protein (ERCC1) is predictive of adjuvant cisplatin-based chemotherapy benefit in resected non-small-cell lung cancer (NSCLC). Non-squamous cell carcinomas (non-SCCs) carry an increased risk of brain metastases (BMs). We hypothesised that there might be an increased incidence of BMs in ERCC1-negative non-SCCs when treated with adjuvant cisplatin-based chemotherapy. PATIENTS AND METHODS: Incidence of BMs and histoclinical parameters were analysed in a population of 761 patients enrolled in the International Adjuvant Lung Cancer Trial. A subgroup analysis was carried out in patients with ERCC1-negative non-SCCs. RESULTS: Of 761 patients, 98 developed BMs alone or in association with other metastatic sites, with a 5-year incidence rate of 18.0% (14.7%-21.8%). In the multivariate analysis, the clinical parameters associated with the occurrence of BMs were the nodal status (P = 0.02) and histological type [give hazard ratio (HR) for non-squamous to quantify introduction assertion, P = 0.002]. Chemotherapy had no effect on BMs [HR = 1.4 (0.90-2.1), P = 0.14]. In patients with non-SCC histology (n = 335), adjuvant chemotherapy was associated with an increased risk of BMs [HR = 2.1 (1.01-4.3), P = 0.04] for ERCC1-negative tumours, whereas there was no evidence of an effect on BMs for ERCC1-positive tumours [HR = 1.1 (0.38-3.0), P = 0.90]. Nevertheless, these two effects are not different (P = 0.30 for interaction) possibly due to a lack of power in this subgroup. CONCLUSIONS: This study suggests that adjuvant cisplatin-based chemotherapy is associated with an increased risk of BMs in patients with resected chemosensitive non-SCCs. If confirmed, these data could provide a rationale for new follow-up and/or prophylactic strategies.

[ERCC1 and lung cancer]. / ERCC1 et cancer pulmonaire.

ERCC1 plays a critical role in the repair of lesions in DNA induced by cisplatin. It has been demonstrated that, in patients with resected NSCLC, those with ERCC1-negative tumours had greater benefit from adjuvant chemotherapy than patients with ERCC1-positive tumours. On the other hand, patients with ERCC1-positive tumours have a better prognosis. ERCC1 can be evaluated either by the level of protein expression on immunohistochemical analysis or by quantification of its mRNA expression. At present ERCC1 appears to be a promising biomarker for predicting the benefit of cisplatin based chemotherapy and its modulation could allow sensitisation of tumour cells to this drug. Its use in daily clinical practice requires a further stage of prospective validation studies which are already in progress in patients both with completely resected NSCL or metastatic disease.

Synergistic proapoptotic effects of the two tyrosine kinase inhibitors pazopanib and lapatinib on multiple carcinoma cell lines.

Pazopanib and lapatinib are two tyrosine kinase inhibitors that have been designed to inhibit the VEGF tyrosine kinase receptors 1, 2 and 3 (pazopanib), and the HER1 and HER2 receptors in a dual manner (lapatinib). Pazopanib has also been reported to mediate inhibitory effect on a selected panel of additional tyrosine kinases such as PDGFR and c-kit. Here, we report that pazopanib and lapatinib act synergistically to induce apoptosis of A549 non-small-cell lung cancer cells. Systematic assessment of the kinome revealed that both pazopanib and lapatinib inhibited dozens of different tyrosine kinases and that their combination could suppress the activity of some tyrosine kinases (such as c-Met) that were not or only partially affected by either of the two agents alone. We also found that pazopanib and lapatinib induced selective changes in the transcriptome of A549 cells, some of which were specific for the combination of both agents. Analysis of a panel of unrelated human carcinoma cell lines revealed a signature of 52 genes whose up- or downregulation reflected the combined action of pazopanib and lapatinib. Indeed, pazopanib and lapatinib exerted synergistic cytotoxic effects on several distinct non-small-cell lung cancer cells as well as on unrelated carcinomas. Altogether, these results support the contention that combinations of tyrosine kinase inhibitors should be evaluated for synergistic antitumor effects. Such combinations may lead to a 'collapse' of pro-survival signal transduction pathways that leads to apoptotic cell death.

Suppression of the DNA damage response in acute myeloid leukemia versus myelodysplastic syndrome.

The molecular mechanisms responsible for the evolution from the preleukemic entities of low-risk myelodysplastic syndrome (MDS) to the less favorable forms of high-risk MDS, as well as those enabling transformation to acute myeloid leukemia (AML), are still incompletely understood. Abundant evidence from solid tumors demonstrates that preneoplastic lesions activate signaling pathways of a DNA damage response (DDR), which functions as an 'anticancer barrier' hindering tumorigenesis. Testing the hypothesis that subgroups of MDS and AML differ with respect to DDR, we first assessed markers of DDR (phosphorylation of ATM, Chk-1, Chk-2 and H2AX) in cell lines representing different entities of MDS (P39, MOLM-13) and AML (MV4-11, KG-1) before and after gamma-irradiation. Although gamma-irradiation induced apoptosis and G(2)/M arrest and a concomitant increase in the phosphorylation of ATM, Chk-1 and H2AX in MDS-derived cell lines, this radiation response was attenuated in the AML-derived cell lines. It is noteworthy that KG-1, but not P39 cells exhibit signs of an endogenous activation of the DDR. Similarly, we found that the frequency of P-ATM(+) cells detectable in bone marrow (BM) biopsies increased in samples from patients with AML as compared with high-risk MDS samples and significantly correlated with the percentage of BM blasts. In contrast, the frequency of gamma-H2AX(+) cells was heterogeneous in all subgroups of AML and MDS. Whereas intermediate-1 MDS samples contained as little P-Chk-1 and P-Chk-2 as healthy controls, staining for both checkpoint kinases increased in intermediate-2 and high-risk MDS, yet declined to near-to-background levels in AML samples. Thus the activation of Chk-1 and Chk-2 behaves in accord with the paradigm established for solid tumors, whereas ATM is activated during and beyond transformation. In conclusion, we demonstrate the heterogeneity of the DDR response in MDS and AML and provide evidence for its selective suppression in AML because of the uncoupling between activated ATM and inactive checkpoint kinases.

Telomere shortening is correlated with the DNA damage response and telomeric protein down-regulation in colorectal preneoplastic lesions.

BACKGROUND: A relation between telomere attrition in early carcinogenesis and activation of DNA damage response (DDR) has been proposed. We explored telomere length and its link with DDR in colorectal multistep carcinogenesis. PATIENTS AND METHODS: We studied normal mucosa, low-grade dysplasia (LGD) and high-grade dysplasia (HGD) and invasive carcinoma (IC) in matched human colon specimens by evaluating p-ataxia telangiectasia mutated (ATM), p-checkpoint kinase 2 (Chk2), c-H2AX, TRF1 and TRF2 expressions by immunohistochemistry. FISH was used to assess telomere length. RESULTS: Telomeres shortened significantly from normal (N) to LGD and HGD (P < 0.0001; P = 0.012), then increased in length in IC (P = 0.006). TRF1 and TRF2 expressions were diminished from N to LGD and HGD (P = 0.004, P < 0.0001, ns) and were reexpressed at the invasive stage (P = 0.053 and P = 0.046). Phosphorylated ATM, Chk2 and H2AX appeared already in LGD (respectively, P = 0.001, P = 0.002 and P = 0.02). Their expression decreased from HGD to IC (respectively, P = 0.03, P = 0.02 and P = 0.37). These activating phosphorylations were inversely correlated with telomere length and TRF1/2 expression. CONCLUSION: In a model of colon multistep carcinogenesis, our data indicate that telomeric length and protein expression levels are inversely correlated with the activation of the DDR pathway.

Assessing abnormal gene promoter methylation in paraffin-embedded sputum from patients with NSCLC.

Aberrant methylation of CpG islands is an important pathway for regulation of gene expression. Recent data suggest that epigenetic abnormalities may occur very early in lung carcinogenesis. We studied the promoters of the four genes, HOX A9, p16(INK4a) (p16), MAGE A1 and MAGE B2 by methylation-specific PCR in matched normal tissue, tumour, and cytological negative sputum samples obtained from 22 patients with non-small cell lung cancer (NSCLC). We further report methylation abnormalities in sputum samples from 56 smokers with differential cytology readouts (negative, inflammatory changes, suspicious, and cancer). Our method was successfully performed on formalin-fixed and paraffin-embedded (FFPE) samples, and was fit to study only few cells obtained by a convenient non-invasive sputum collection and handling. The promoters of MAGE A1 and MAGE B2 had abnormal methylation patterns in, respectively, 50% and 41% of the cytologically negative sputum samples from NSCLC patients, whereas methylation abnormalities of p16 was observed in 27% of negative sputum samples. Interestingly, 95.5% (21 of 22) of the cytologically negative sputum samples from NSCLC patients had abnormal methylation in at least one of the four genes indicating a high sensitivity of this marker system. Moreover, a higher frequency of methylation abnormalities was observed in sputum samples from smokers with high cytological grade compared to low cytological grade. We conclude that the identification of abnormal gene methylation of a limited number of markers in FFPE sputum samples is feasible and may be investigated as a potential system for population-based screening of early stages of lung cancer.