Ibrutinib plus venetoclax is a highly effective combination in mantle cell lymphoma. However, strategies to enable the evaluation of therapeutic response are required. Our prospective analyses of patients within the AIM study revealed genomic profiles that clearly dichotomized responders and nonresponders. Mutations in ATM were present in most patients who achieved a complete response, while chromosome 9p21.1-p24.3 loss and/or mutations in components of the SWI-SNF chromatin-remodeling complex were present in all patients with primary resistance and two-thirds of patients with relapsed disease. Circulating tumor DNA analysis revealed that these alterations could be dynamically monitored, providing concurrent information on treatment response and tumor evolution. Functional modeling demonstrated that compromise of the SWI-SNF complex facilitated transcriptional upregulation of BCL2L1 (Bcl-xL) providing a selective advantage against ibrutinib plus venetoclax. Together these data highlight important insights into the molecular basis of therapeutic response and provide a model for real-time assessment of innovative targeted therapies.

Several novel therapeutics are poised to change the natural history of chronic lymphocytic leukaemia (CLL) and the increasing use of these therapies has highlighted limitations of traditional disease monitoring methods. Here we demonstrate that circulating tumour DNA (ctDNA) is readily detectable in patients with CLL. Importantly, ctDNA does not simply mirror the genomic information contained within circulating malignant lymphocytes but instead parallels changes across different disease compartments following treatment with novel therapies. Serial ctDNA analysis allows clonal dynamics to be monitored over time and identifies the emergence of genomic changes associated with Richter's syndrome (RS). In addition to conventional disease monitoring, ctDNA provides a unique opportunity for non-invasive serial analysis of CLL for molecular disease monitoring.

The diagnosis and monitoring of myelodysplastic syndromes (MDSs) are highly reliant on bone marrow morphology, which is associated with substantial interobserver variability. Although azacitidine is the mainstay of treatment in MDS, only half of all patients respond. Therefore, there is an urgent need for improved modalities for the diagnosis and monitoring of MDSs. The majority of MDS patients have either clonal somatic karyotypic abnormalities and/or gene mutations that aid in the diagnosis and can be used to monitor treatment response. Circulating cell-free DNA is primarily derived from hematopoietic cells, and we surmised that the malignant MDS genome would be a major contributor to cell-free DNA levels in MDS patients as a result of ineffective hematopoiesis. Through analysis of serial bone marrow and matched plasma samples (n = 75), we demonstrate that cell-free circulating tumor DNA (ctDNA) is directly comparable to bone marrow biopsy in representing the genomic heterogeneity of malignant clones in MDS. Remarkably, we demonstrate that serial monitoring of ctDNA allows concurrent tracking of both mutations and karyotypic abnormalities throughout therapy and is able to anticipate treatment failure. These data highlight the role of ctDNA as a minimally invasive molecular disease monitoring strategy in MDS.

The traditional maximum likelihood estimator (MLE) is often of limited use in complex high-dimensional data due to the intractability of the underlying likelihood function. Maximum composite likelihood estimation (McLE) avoids full likelihood specification by combining a number of partial likelihood objects depending on small data subsets, thus enabling inference for complex data. A fundamental difficulty in making the McLE approach practicable is the selection from numerous candidate likelihood objects for constructing the composite likelihood function. In this article, we propose a flexible Gibbs sampling scheme for optimal selection of sub-likelihood components. The sampled composite likelihood functions are shown to converge to the one maximally informative on the unknown parameters in equilibrium, since sub-likelihood objects are chosen with probability depending on the variance of the corresponding McLE. A penalized version of our method generates sparse likelihoods with a relatively small number of components when the data complexity is intense. Our algorithms are illustrated through numerical examples on simulated data as well as real genotype single nucleotide polymorphism (SNP) data from a case–control study.

Abstract Women who carry a germline BRCA1 or BRCA2 mutation are at high risk of developing ovarian cancer in their lifetime. While risk can be reduced by prophylactic bilateral salpingo-oopherectomy (RRSBO), for those high-risk women who delay, or choose not to undergo RRSBO, current surveillance options have not shown sufficient specificity or sensitivity to prevent an advanced stage diagnosis (Buys et al (2011) JAMA). As such there is a need to identify novel, sensitive biomarkers to employ in this at-risk population. We recently found that BRCA1/2 mutation carriers predominantly develop high-grade serous cancers (HGSC) (Alsop et al (2012) J Clin Oncol), therefore, identifying a biomarker of HGSC will be useful for screening BRCA1/2 carriers. Mutation in TP53 represents the ideal biomarker of HGSC; mutation occurs early in tumour development (Crum et al (2007) Clin Med Res), it is present in almost 100% of tumours (Ahmed, et al (2010) J Pathol, TCGA (2011) Nature), and pathogenic mutations can be readily distinguished from passengers (Olivier et al (2010) Cold Spring Harb Perspect Biol). Tumour specific TP53 mutations have been identified in the circulating DNA of late stage ovarian cancer patients by Sanger sequencing (Swisher et al (2005) Am J Obstet Gynecol). Next generation sequencing (NGS) provides increased sensitivity for identification of rare mutant alleles (Flaherty et al (2011) Nucl Acids Res) and has been used to monitor disease burden in advanced stage cases (Murtaza et al (2013) Nature). Supported by a Marsha Rivkin Challenge Grant, we explored the feasibility of detecting free circulating mutant TP53 DNA in plasma from HGSC patients, using multiplex PCR to amplify TP53 DNA from the circulating DNA followed by ultra-deep NGS. The main challenge has been in developing bioinformatics filtering processes to increase specificity and sensitivity. We utilised multiple variant callers to identify low frequency, high quality, pathogenic variants that are then compared to the IARC TP53 database and COSMIC. Variants called in our sequence data that are absent from these databases are flagged as lower confidence calls. Consistent sequencing artefacts were removed from consideration. High confidence variants were tested for their presence in multiple amplicons, to generate a final list of variants ranked in terms of confidence based on multiple lines of bioinformatics evidence. Using this methodology, we have been able to detect tumour specific TP53 mutations in circulating DNA of 40% of predominately late stage cases. In those cases where the mutation was not called in the circulating DNA, the mutant allele frequency was less than 0.1% and could not be distinguished from the background errors associated with NGS. The extremely low frequency of mutations in the plasma of women with late stage HGSC, against a background of normal TP53 sequences, currently hampers immediate adoption of circulating TP53 mutations as an early detection biomarker. Our work provides proof of principle for the approach, however, we need to develop additional strategies to push detection into a range that is useful for screening at risk women. Citation Format: Elizabeth Christie, Tane Hunter, Maria Doyle, Australian Ovarian Cancer Study, David Bowtell. Circulating TP53 mutations as a biomarker of high-grade serous cancer [abstract]. In: Proceedings of the 10th Biennial Ovarian Cancer Research Symposium; Sep 8-9, 2014; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2015;21(16 Suppl):Abstract nr POSTER-CTRL-1203.

This article illustrates many features of a mathematics document created with Scientific Word or Scientifi cW orkPlace.