Phase II trial of delta-tocotrienol in neoadjuvant breast cancer with evaluation of treatment response using ctDNA.

Neoadjuvant treatment of breast cancer is applied to an increasing extent, but treatment response varies and side effects pose a challenge. The vitamin E isoform delta-tocotrienol might enhance the efficacy of chemotherapy and reduce the risk of side effects. The aim of this study was to investigate the clinical effect of delta-tocotrienol combined with standard neoadjuvant treatment and the possible association between detectable circulating tumor DNA (ctDNA) during and after neoadjuvant treatment with pathological treatment response. This open-label, randomized phase II trial included 80 women with newly diagnosed, histologically verified breast cancer randomized to standard neoadjuvant treatment alone or in combination with delta-tocotrienol. There was no difference in the response rate or frequency of serious adverse events between the two arms. We developed a multiplex digital droplet polymerase chain reaction (ddPCR) assay for the detection of ctDNA in breast cancer patients that targets a combination of two methylations specific for breast tissue (LMX1B and ZNF296) and one cancer specific methylation (HOXA9). The sensitivity of the assay increased when the cancer specific marker was combined with the ones specific to breast tissue (p < 0.001). The results did not show any association between ctDNA status and pathological treatment response, neither at midterm nor before surgery.

Deep learning detects and visualizes bleeding events in electronic health records.

BACKGROUND: Bleeding is associated with a significantly increased morbidity and mortality. Bleeding events are often described in the unstructured text of electronic health records, which makes them difficult to identify by manual inspection. OBJECTIVES: To develop a deep learning model that detects and visualizes bleeding events in electronic health records. PATIENTS/METHODS: Three hundred electronic health records with International Classification of Diseases, Tenth Revision diagnosis codes for bleeding or leukemia were extracted. Each sentence in the electronic health record was annotated as positive or negative for bleeding. The annotated sentences were used to develop a deep learning model that detects bleeding at sentence and note level. RESULTS: On a balanced test set of 1178   sentences, the best-performing deep learning model achieved a sensitivity of 0.90, specificity of 0.90, and negative predictive value of 0.90. On a test set consisting of 700 notes, of which 49 were positive for bleeding, the model achieved a note-level sensitivity of 1.00, specificity of 0.52, and negative predictive value of 1.00. By using a sentence-level model on a note level, the model can explain its predictions by visualizing the exact sentence in a note that contains information regarding bleeding. Moreover, we found that the model performed consistently well across different types of bleedings. CONCLUSIONS: A deep learning model can be used to detect and visualize bleeding events in the free text of electronic health records. The deep learning model can thus facilitate systematic assessment of bleeding risk, and thereby optimize patient care and safety.

Prognostic impact of serum levels of EGFR and EGFR ligands in early-stage breast cancer.

Epidermal growth factor receptor (EGFR) and its ligands are involved in cancer pathogenesis. The emerging role of treatments co-targeting the EGFR system in breast cancer has increased the need to identify companion biomarkers. The aim of this study is to investigate whether pretreatment serum levels of EGFR and EGFR ligands in early-stage breast cancer patients might provide prognostic information as a stepping stone for further investigation. The study, which included 311 early-stage breast cancer patients, investigated associations between preoperative serum levels of EGFR and EGFR ligands (epidermal growth factor, heparin-binding epidermal growth factor (HBEGF), amphiregulin, transforming growth factor-α and betacellulin) and survival. Cutoffs were determined using Youden's method, and overall survival (OS) and invasive disease-free survival (IDFS) were evaluated using Cox regression. Preoperative S-EGFR < 60.3 ng/mL was associated with shorter OS and IDFS in both univariate analyses and when adjusting for standard prognostic factors (p < 0.05). Preoperative S-HBEGF < 21.4 pg/mL was associated with shorter OS in both univariate and multivariate analyses, whereas association with shorter IDFS could only be demonstrated in the univariate analysis. In conclusion, our study demonstrated shorter survival in early-stage breast cancer patients who had low pretreatment levels of either S-EGFR or S-HBEGF.

Dysregulated EGFR pathway in serum in early-stage breast cancer patients: A case control study.

The epidermal growth factor receptor (EGFR) and its ligands are involved in cancer pathogenesis and they might serve as circulating biomarkers. The current study aims to investigate if abnormal pre-treatment serum levels of EGFR and EGFR ligands are present in women with early-stage breast cancer and if up- or downregulation of EGFR and EGFR ligands occur in defined patient subgroups. Pre-treatment serum samples were obtained from 311 women with newly diagnosed early-stage breast cancer and from 419 healthy women and analysed for EGFR and the ligands: Epidermal growth factor (EGF), heparin-binding epidermal growth factor (HBEGF), betacellulin (BTC), amphiregulin (AREG), and transforming growth factor α (TGF-α). Previously, age-dependent 95% reference intervals for EGFR and the EGFR ligands have been established based on the healthy women population. S-EGFR, S-EGF, S-HBEGF, S-AREG, and S-TGFα were all significantly different in women with breast cancer compared to healthy women (p < 0.05). Elevated S-EGFR, according to the reference intervals, was present in 11.3% of breast cancer patients, whereas decreased S-EGF was found in 11.6%. Elevated S-EGFR was associated with estrogen receptor positivity of tumor (ER+) and a subgroup of ER + breast cancer patients showed markedly elevated S-EGFR (>120 ng/mL).

EGFR and EGFR ligands in serum in healthy women; reference intervals and age dependency.

Background The epidermal growth factor receptor (EGFR) system is involved in cancer pathogenesis and serves as an important target for multiple cancer treatments. EGFR and its ligands epidermal growth factor (EGF), heparin-binding epidermal growth factor (HB-EGF), betacellulin (BTC), amphiregulin (AREG) and transforming growth factor α (TGF-α) have potential applications as prognostic or predictive serological biomarkers in cancer. The aim was to establish EGFR and EGFR ligand reference intervals in healthy women. Methods EGFR and EGFR ligands were measured in serum from 419 healthy women aged 26-78 years. The need for age partitioned reference intervals was evaluated using Lahti's method. EGFR and EGF were analyzed using ELISA assays, whereas HB-EGF, BTC, AREG and TGF-α were analyzed using the highly sensitive automated single molecule array (Simoa) enabling detection below the lower reference limit for all six biomarkers. Results Reference intervals for EGFR and the EGFR ligands were determined as the 2.5th and 97.5th percentiles. All six biomarkers were detectable in all serum samples. For EGFR, EGF, HB-EGF and TGF-α, reference intervals were established for women <55 years and for women >55 years, whilst common reference intervals were established for AREG and BTC including women aged 26-78 years. Conclusions Age specific reference intervals were determined for EGFR, EGF, HB-EGF, BTC, AREG and TGF-α.

Development of a three-plex single molecule immunoassay enabling measurement of the EGFR ligands amphiregulin, betacellulin and transforming growth factor α simultaneously in human serum samples.

BACKGROUND: Prior to large studies in breast cancer patients and healthy individuals we established a sensitive three-plex immunoassay to measure the EGFR ligands amphiregulin (AR), betacellulin (BTC) and transforming growth factor α (TGF-α) simultaneously in human serum samples. METHOD: The three-plex immunoassay was developed using single molecule array (Simoa) technology and requires only 20 µL of serum. RESULTS: AR, BTC and TGF-α were first established as three single-plex assays. Multiplexing the three single-plex assays showed no significant cross reactivity between the reagents. The concentrations of the ligands in serum samples showed correlations r2 ≥ 0.84 between the single-plex and three-plex methods. The three-plex assay demonstrated limit of detection levels at 0.16 ng/L for AR, 0.23 ng/L for BTC and 0.22 ng/L for TGF-α. Total coefficients of variations were 8.5%-31% for AR, 11%-21.8% for BTC and 12.4%-16.2% for TGF-α. Spiking experiments showed a mean recovery of 97% for AR, 86% for BTC and 81% for TGF-α. The concentrations of the EGFR ligands did not change significantly after series of freeze thaw cycles or incubation at 22 °C for up to 24 h. CONCLUSION: This robust three-plex assay with up to 40-fold increase in sensitivity relative to conventional ELISA is the first published method that has the required sensitivity to measure AR, BTC and TGF-α simultaneously in human blood samples.

Prognostic and predictive value of EGFR and EGFR-ligands in blood of breast cancer patients: a systematic review.

Epidermal growth factor receptor (EGFR) serves as a co-target for dual/pan-EGFR-inhibitors in breast cancer. Findings suggest that EGFR and EGFR-ligands are involved in resistance towards certain breast cancer treatments. The aim is to explore the validity of EGFR and EGFR-ligands in blood as prognostic and predictive biomarkers in breast cancer. The systematic review was conducted in accordance to the PRISMA guidelines. Literature searches were conducted to identify publications exploring correlations between EGFR/EGFR-ligands in serum/plasma of breast cancer patients and prognostic/predictive outcome measures. Sixteen publications were eligible for inclusion. Twelve studies evaluated EGFR, whereas five studies evaluated one or more of the EGFR-ligands. Current evidence indicates associations between low baseline serum-EGFR and shorter survival or reduced response to treatment in patients with advanced breast cancer, especially in patients with estrogen and/or progesterone receptor positive tumors. The prognostic and predictive value of EGFR and EGFR-ligands in blood has only been investigated in highly selected subsets of breast cancer patients and most studies were small. This is the first systematic review evaluating the utility of EGFR and EGFR-ligands as predictive and prognostic biomarkers in blood in breast cancer. Further exploration in large well-designed studies is needed.