Untangling the mess of CGRP levels as a migraine biomarker: an in-depth literature review and analysis of our experimental experience.

BACKGROUND: Calcitonin gene-related peptide (CGRP) is the most promising candidate to become the first migraine biomarker. However, literature shows clashing results and suggests a methodological source for such discrepancies. We aimed to investigate some of these methodological factors to evaluate the actual role of CGRP as biomarker. METHODS: Previous to the experimental part, we performed a literature review of articles measuring CGRP in migraine patients. Using our 399 bio-bank sera samples, we performed a series of experiments to test the validity of different ELISA kits employed, time of sample processing, long-term storage, sampling in rest or after moderate exercise. Analysis of in-house data was performed to analyse average levels of the peptide and the effect of sex and age. RESULTS: Literature review shows the high variability in terms of study design, determination methods, results and conclusions obtained by studies including CGRP determinations in migraine patients. CGRP measurements depends on the method and specific kit employed, also on the isoform detected, showing completely different ranges of concentrations. Alpha-CGRP and beta-CGRP had median with IQR levels of 37.5 (28.2-54.4) and 4.6 (2.4-6.4)pg/mL, respectively. CGRP content is preserved in serum within the 24 first hours when samples are stored at 4°C after clotting and immediate centrifugation. Storages at -80°C of more than 6 months result in a decrease in CGRP levels. Moderate exercise prior to blood extraction does not modulate the concentration of the peptide. Age positively correlates with beta-CGRP content and men have higher alpha-CGRP levels than women. CONCLUSIONS: We present valuable information for CGRP measurements in serum. ELISA kit suitability should be tested prior to the experiments. Alpha and beta-CGRP levels should be analysed separately as they can show different behaviours even within the same condition. Samples can be processed in a 24-h window if they have been kept in 4°C and should not be stored for more than 6 months at -80°C before assayed. Patients do not need to rest before the blood extraction unless they have performed a high-endurance exercise. For comparative studies, sex and age should be accounted for as these parameters can impact CGRP concentrations.

BRCA1 haploinsufficiency cell-autonomously activates RANKL expression and generates denosumab-responsive breast cancer-initiating cells.

Denosumab, a monoclonal antibody to the receptor activator of nuclear factor-κB ligand (RANKL), might be a novel preventative therapy for BRCA1-mutation carriers at high risk of developing breast cancer. Beyond its well-recognized bone-targeted activity impeding osteoclastogenesis, denosumab has been proposed to interfere with the cross-talk between RANKL-producing sensor cells and cancer-initiating RANK+ responder cells that reside within premalignant tissues of BRCA1-mutation carriers. We herein tested the alternative but not mutually exclusive hypothesis that BRCA1 deficiency might cell-autonomously activate RANKL expression to generate cellular states with cancer stem cell (CSC)-like properties. Using isogenic pairs of normal-like human breast epithelial cells in which the inactivation of a single BRCA1 allele results in genomic instability, we assessed the impact of BRCA1 haploinsufficiency on the expression status of RANK and RANKL. RANK expression remained unaltered but RANKL was dramatically up-regulated in BRCA1mut/+ haploinsufficient cells relative to isogenic BRCA1+/+ parental cells. Neutralizing RANKL with denosumab significantly abrogated the ability of BRCA1 haploinsufficient cells to survive and proliferate as floating microtumors or "mammospheres" under non-adherent/non-differentiating conditions, an accepted surrogate of the relative proportion and survival of CSCs. Intriguingly, CSC-like states driven by epithelial-to-mesenchymal transition or HER2 overexpression traits responded to some extent to denosumab. We propose that breast epithelium-specific mono-allelic inactivation of BRCA1 might suffice to cell-autonomously generate RANKL-addicted, denosumab-responsive CSC-like states. The convergent addiction to a hyperactive RANKL/RANK axis of CSC-like states from genetically diverse breast cancer subtypes might inaugurate a new era of cancer prevention and treatment based on denosumab as a CSC-targeted agent.