Biomarker directed selection of targeted anti-neoplastic agents such as immune checkpoint inhibitors, small molecule inhibitors and monoclonal antibodies form an important aspect of cancer treatment. Immunohistochemistry (IHC) analysis of the tumor tissue is the method of choice to evaluate the presence of these biomarkers. However, a significant barrier to biomarker testing on tissue is the availability of an adequate amount of tissue and need for repetitive sampling due to tumor evolution. Also, tumor tissue testing is not immune to inter- and intra-tumor heterogeneity. We describe the analytical and clinical validation of a Circulating Tumor Cell (CTC) assay to accurately assess the presence of PD-L1 22C3 and PD-L1 28.8, ER, PR and HER2, from patients with solid tumors to guide the choice of suitable targeted therapies. Analytically, the test has high sensitivity, specificity, linearity and precision. Based on a blinded case control study, the clinical sensitivity and specificity for PD-L1 (22C3 and 28.8) was determined to be 90% and 100% respectively. The clinical sensitivity and specificity was 83% and 89% for ER; 80% and 94% for PR; 63% and 89% for HER2 (by ICC); and 100% and 92% for HER2 (by FISH), respectively. The performance characteristics of the test support its suitability and adaptability for routine clinical use.

Real time monitoring of disease status is an essential part of cancer management. The low sensitivity and specificity of serum markers and the constraints and risks associated with radiological scans prompt the need for accurate non-invasive means to monitor minimal residual disease (MRD) in solid tumors. In this study we describe MRD evaluation via profiling of patient-specific gene variants in cell free tumor DNA (i.e., ctMRD). We evaluate the feasibility of this approach for real time monitoring of tumor load dynamics in response to anticancer treatments. We prioritized 162 hot spot mutations for designing ctMRD assays based on literature review. These ctMRD assays were evaluated in 436 plasma specimens with a median of 6 (range 3-18) longitudinal evaluations in a cohort of 48 patients with various solid tumors. In patients with partial radiological response (PR), Mutant Allele Fraction (MAF) showed high correlation (84%) with radiological response and tumor volume (cm3) compared to conventional CA markers (53%). Total plasma ctDNA level was significantly higher in patients with 2-5 metastatic sites compared with single metastatic site (P = 0.04) and discriminated patients with stable disease (SD) and progressive disease (PD) from patients with partial response (PR) (P = 0.01 and P = 0.04, respectively). Collectively, the present study shows that changes in mutation burden evaluated using patient specific ctMRD assays is a highly sensitive approach for monitoring of therapy response.

Invasive procedures for diagnosis of CNS malignancies carry inherent risks of high morbidity and mortality. Although circulating biomarkers such as cell free DNA (cfDNA) and microvesicle (MV) borne nucleic acids have been proposed as potential diagnostic aids, their stand-alone utility has inherent limitations. However, Circulating Glial Cells (CGCs) combined with cfDNA could offer a viable alternative to invasive biopsies for diagnosis of CNS malignancies; yet the technological challenge in the detection of CGCs in glioma patients presents a formidable challenge. In this study, we evaluated the feasibility of harvesting CGCs from suspected cases of Glioma. From a cohort of 23 suspected cases of CNS malignancies, we used 15ml of peripheral blood and used the CellWizard™ process and related protocol for isolation of CGCs. CellWizard™ is an epigenetically active media with paradoxical chemo-toxicity that selectively induces lethality in normal cells which have a functionally responsive cell death (apoptosis) mechanism, while simultaneously conferring survival privilege on apoptosis resistant cells typically released from a malignant tumor. This paradoxical cytotoxicity of the medium leads to selective elimination of most leukocytes thus facilitating a label free negative enrichment of CGCs, which can be harvested and further characterized. Patients included 11 Glioblastoma, 3 Anaplastic astrocytoma, 2 Medulloblastoma, 5 Oligodendroglioma, 1 Gliosarcoma and 1 meningioma patient. Characterization of CGCs was performed using GFAP, S100 and CD45 markers. CGCs were detected in 16 out of 23 (69.6 %) patients and could be stained positively for both GFAP and S100 and negatively for CD45. Detection of viable CGCs in cases of CNS malignancies can be used for characterization of markers related to the diagnosis.

The diagnosis of Central Nervous System (CNS) malignancies such as Gliomas in individuals presenting with Intracranial Space Occupying Lesions (ICSOL) is based on histopathological examination (HPE) of tumor tissue obtained by an invasive brain biopsy. However, brain biopsies are resource intensive and are associated with procedural risks such as haemorrhage, morbidity and mortality. The present study evaluated a non-invasive approach for diagnosis of CNS-M in symptomatic individuals based on evaluation of circulating tumor analytes in peripheral blood. The non-invasive multi-platform approach for diagnosis of CNS-M included Immunocytochemistry (ICC) profiling and Fluorescence in situ Hybridization (FISH) of Circulating Tumor Cells (CTCs) and Digital Droplet PCR (ddPCR) of cell-free tumor DNA (ctDNA) and exosomal mRNA. Performance characteristics of each platform were evaluated using blood and tissue samples from 445 individuals including 227 known cases of CNS-M, 47 known cases of benign CNS conditions (CNS-B), 141 known cases of other cancers with brain metastases (OTH-M) and 30 asymptomatic individuals (ASYM). In a set of 37 samples from individuals with radiological ICSOL, suspected of malignancy (CNS-S) complete diagnostic work-up was performed with ICC, FISH and ddPCR. Glial CTCs were detected in 88.8% of 227 CNS-M and undetectable in 89.4% of 47 CNS-B or 100% of 141 OTH-M, indicating high sensitivity and specificity, respectively. The multi-analyte approach discerned CNS-M from CNS-B as well as OTH-M with 91.7% accuracy and accurately inferred lineage in 84.6% of cases. This non-invasive multi-analyte approach can diagnose CNS-M with an accuracy not inferior to standard HPE, can substitute invasive biopsies in most cases and is particularly helpful in cases where a biopsy is not viable.

Biomarker directed selection of targeted anti-neoplastic agents such as immune check-point inhibitors, small molecule inhibitors and monoclonal antibodies form an important aspect of cancer treatment. Immunohistochemistry (IHC) analysis of the tumor tissue is the method of choice to evaluate the presence of these biomarkers. However, a significant barrier to biomarker testing on tissue is the availability of an adequate amount of tissue and need for repetitive sampling due to tumor evolution. Also, tumor tissue testing is not immune to inter- and intra-tumor heterogeneity. We describe the analytical and clinical validation of a Circulating Tumor Cell (CTC) assay to accurately assess the presence of PD-L1 22C3 and PD-L1 28.8, ER, PR and HER2, from patients with solid tumors to guide the choice of suitable targeted therapies. Analytically, the test has high sensitivity, specificity, linearity and precision. Based on a blinded case control study, the clinical sensitivity and specificity for PD-L1 (22C3 and 28.8) was determined to be 90% and 100% respectively. The clinical sensitivity and specificity was 83% and 89% for ER; 80% and 94% for PR; 63% and 89% for HER2 (by ICC); and 100% and 92% for HER2 (by FISH), respectively. The performance characteristics of the test support its suitability and adaptability for routine clinical use.