Type IV collagen as a potential biomarker of metastatic breast cancer

Clinical & Experimental Metastasis https://doi.org/10.1007/s10585-021-10082-2 RESEARCH PAPER Type IV collagen as a potential biomarker of metastatic breast cancer Moa Lindgren1 · Malin Jansson1 · Björn Tavelin2 · Luc Dirix3 · Peter Vermeulen3 · Hanna Nyström1,3,4 Received: 3 December 2020 / Accepted: 15 February 2021 © The Author(s) 2021 Abstract No reliable, non-invasive biomarker of metastatic breast cancer (mBC) exists: circulating CA15-3 (cCA15-3) is the marker mostly used to monitor mBC. Circulating collagen IV (cCOLIV) has been evaluated in other metastatic cancers and has been found to be a promising biomarker. The overarching aim of this study was to evaluate cCOLIV as a potential biomarker in patients with mBC. The first aim was to determine the levels of cCOL IV and cCA15-3 in patients with healthy controls, primary breast cancer (pBC) and mBC. The second aim was to compare levels of cCOLIV and cCA15-3 in patients with different metastatic sites of BC. The third aim was to investigate the prognostic value of cCOLIV and cCA15-3 for mBC patients. The fourth aim was to analyse whether a combination of the two biomarkers was more accurate in detecting mBC than a single marker. Lastly, we investigated the tissue expression levels of COLIV in BC bone metastases (BM) and liver metastases (LM). Plasma levels of cCOLIV and cCA15-3 from healthy controls and patients with pBC and mBC were measured. COLIV expression in tissue from patients with LM and BM was analysed using immunohistochemistry. Clinical and survival data were collected from medical charts. The levels of cCOLIV and cCA15-3 were significantly elevated in mBC patients compared with healthy controls and pBC patients. No differences in cCOLIV and cCA15-3 levels were found based on the metastatic site. High levels of cCOLIV, but not cCA15-3, correlated with poorer survival. cCOLIV alone and the combination of cCA15-3 and cCOLIV were superior to cCA15-3 at detecting mBC. COL IV was highly expressed in the tissue of LM and BM. Our study suggests that cCOLIV is a potential marker to monitor patients with BC. Keywords Breast cancer · Metastases · Biomarkers · Collagen IV · CA 15-3 Introduction More than 1.5 million women are diagnosed with breast cancer (BC) every year and it is the most common cancer among women [1–3]. In the western world, the prognosis for these patients has steadily improved over the last few decades, with an average five-year overall survival of 90% [4]. BC-associated mortality is mostly related to late diagnosis * Hanna Nyström 1 Department of Surgery, Department of Surgical and Perioperative Sciences, Umeå University, By 10A, Norrlands Universitetssjukhus, Umeå 901 85, Sweden 2 Department of Radiation Sciences, Umeå University, Umeå, Sweden 3 Translational Cancer Research Unit, GZA Hospital Sint Augustinus and Antwerp University, Antwerp, Belgium 4 Wallenberg Centre for Molecular Medicine, Umeå University, Umeå, Sweden and metastatic spread and it is currently estimated that, each year, 500,000 deaths occur worldwide due to metastatic BC (mBC) [3]. Patients with a history of primary BC (pBC) are monitored by clinical examination and imaging to detect metastatic spread [4]. Personalised treatment is essential for patients with BC. Treatment is in fact guided by prognostic and predictive biomarkers, such as the tissue expression of oestrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) [4, 5]. The diagnosis and treatment planning thus rely on invasive procedures and expensive medical imaging. In this context, markers in the blood could be of great use and are highly desirable. Currently, no optimal non-invasive biomarker is available for either pBC or mBC. To our knowledge, MUC1-derived cancer antigen 15-3 (CA 15-3) is the only circulating biomarker used in the clinic. CA 15-3 is derived from tumour cells (TC) and is only used for monitoring advanced disease when standard imaging is unavailable [4, 6]. The limited 13 Vol.:(0123456789) Clinical & Experimental Metastasis applicability of CA 15-3 in explained by its low sensitivity and specificity. High circulating levels are present in patients with both pBC and mBC but also in patients with liver, ovarian and lung cancer and some benign conditions as well [7, 8]. In addition, not all mBC patients have high cCA 15-3 and the reported accuracy is highly dependent on the study set-up [9–14]. One study used a fixed cut-off value of cCA 15-3 (> 40 IU/l) to detect recurrent BC, either local or distant, which resulted in a sensitivity of 36% and a specificity of 97% [15]. In contrast, another study regarded an individual increase in cCA 15-3 of 100% as a strong indicator of metastatic disease and reported a sensitivity of 55.6% and a specificity of 99.8%. It also reported an increase in sensitivity when cCA 15-3 was used in combination with circulating carcinoembryonic antigen (cCEA) compared with when used alone [10]. Furthermore, cCA 15-3 appears to correlate with disease progression, tumour burden and the site of metastasis. It is reported that mBC patients with bone metastases (BM) and liver metastases (LM) present with higher levels of cCA 15-3, compared with lung and lymph node metastases [11, 16]. To conclude, the results show that cCA 15-3 is able to detect BC metastases in some patients but not in all [11]. The ideal biomarker of mBC should have a diagnostic, predictive and prognostic value and should be able to replace other monitoring procedures, including imaging. A biomarker of this kind would be of great clinical significance. The majority of currently available biomarkers in cancer are based on analytes derived from the tumour cells. A source of potential new biomarkers is the tumour stroma. The stroma includes the non-cancerous cells, such as immune cells, vascular cells and fibroblasts, as well as extracellular matrix (ECM) proteins [17]. During tumour progression, the ECM of the stroma is the subject of constant remodelling, contributing to the increased deposition of ECM proteins in the tissue and in the circulation [18]. ECM proteins are potential candidates in the search for new circulating biomarkers. Since a tumour consists of both tumour cells and stroma, a combination of markers representing both compartments could potentially help to improve the sensitivity and specificity of a cancer biomarker. One promising ECM-related biomarker is the basement membrane protein, type IV collagen. Circulating type IV collagen (cCOL IV) is a promising biomarker of other metastatic cancers. Colorectal cancer (CRC) patients with liver metastases (CLM) have higher levels of cCOL IV, compared with patients without metastases, and the level appears to coincide with tumour burden and a poorer prognosis [19, 20]. Furthermore, cCOL IV appears to be superior to the conventional biomarker, circulating carcinoembryonic antigen (cCEA), at identifying CRC patients with a poor prognosis [2 (...truncated)