Estratificación diagnóstica de Globo-H e identificación de DUSP14 como un posible blanco terapéutico en el cáncer colorrectal.

Colorectal cancer (CRC) represents a major global health challenge, ranking among the most common malignancies and standing as the second leading cause of cancer‐related deaths worldwide.^[1], ^[2] Despite advancements in screening and treatment, CRC continues to impose a heavy burden on healthcare systems, particularly in high‐ and middle‐income countries, which account for over 88% of global cases.^[3], ^[4] Another challenge in CRC oncology is the increase in the occurrence of drug‐resistant cancers that limit the effectiveness of anticancer agents.^[5] The drug resistance mechanisms can vary between colon cancer patients and even change within the same patient over time. Thus, personalized and tailored combination therapies are urgently needed in clinical settings. Several innovative strategies were proposed to improve therapeutic outcomes and reduce disease burden.^[6]

Aberrant glycosylation is a hallmark of cancer cells that frequently alters tumor properties.^[7] Among the resulting altered glycans, Globo‐H has emerged as a prominent tumor‐associated carbohydrate antigen (TACA).^[7], ^[8] It is abundant in several epithelial cancers, including breast, ovarian, prostate, colon, and lung tumors, while largely absent in normal adult tissues.^[9] This observation was recapitulated in several human cancer cell lines.^[9] Immunohistochemical analyses have consistently confirmed its expression across malignancies, with minimal presence in normal tissue, primarily limited to the luminal surface of glandular epithelium.^[10] The selective expression and correlation with tumor progression and metastasis make Globo‐H an appealing target for immunotherapy.^[11] The in vitro and biopsy‐based observations led to the development of therapeutic approaches such as Globo‐H based vaccines (e.g., OPT‐822/OPT‐821), which demonstrated promising immunogenicity in clinical trials for breast cancer^[12] and other cancer types.^[13], ^[14], ^[15]

Globo‐H is also found on the cell surface of cancer stem cells (CSCs), offering the potential to target the origin of tumor growth and reduce recurrence.^[10] In vitro studies have shown that cells with high Globo‐H expression exhibit increased proliferation and tumorigenic potential compared to low‐expressing cells.^[16] Furthermore, Globo‐H has been identified as a promoter of angiogenesis and an immune checkpoint molecule, influencing immune and endothelial cell properties.^[10] It was proposed that Globo‐H promoted immunosuppression by downregulating NOTCH1 signaling, thereby impairing immune cell proliferation.^[11] Another promising approach that benefited from the presence of Globo‐H and other antigens on the cell surface (e.g., SSEA3, SSEA4) is that they serve as a basis for antibody‐drug conjugates (ADCs).^[17] The recent clinical evaluation of OBI‐999, a Globo‐H‐targeting ADC, in patients with advanced solid tumors further underscores its therapeutic potential.^[18], ^[19]

This study aims to test the potential of patient heterogeneity while seeking druggable therapeutic targets. We have utilized patient samples to establish an in vitro drug testing platform using two‐dimensional (2D) cultures, and short‐term primary patient‐derived organoids (PDOs) generated from resected colon tumors and adjacent non‐tumor tissues. This system enabled the functional evaluation of the anti‐Globo‐H antibody VK9. Correlating mRNA transcriptome profiling with histological assessment of Globo‐H expression in matched tumor samples, we identified DUSP14 (Dual specificity phosphatase 14) as a key transcript correlated with Globo‐H expression. Pharmacological inhibition of DUSP14 in vitro reduced cell growth, and both baseline activation levels and blockage experiments identified TAK1 (Transforming growth factor‐β‐activated kinase 1) as a putative downstream signal transduction avenue. This work highlights the value of combining patient‐derived models and bioinformatic approaches to uncover candidate antigens and potential therapeutic targets for CRC treatment and diagnosis.

MATERIALS AND METHODS

Sample acquisition

Tissue samples utilized for the establishment of patient‐derived organoids (PDOs) and subsequent tissue analysis were procured from patients who underwent surgical procedures at Otto von Guericke University in Magdeburg. Patient‐derived tissue specimens were meticulously examined and categorized into tumor and non‐tumor tissue by a board‐certified pathologist affiliated with the Department of Pathology, Otto von Guericke University, Magdeburg.### Immunohistochemistry

All procedures were performed in the histology lab of medical pathology. Globo H‐expression was determined using the VK9 primary antibody staining. Detailed protocols for immunohistochemistry (IHC) are provided in Text S1. Globo‐H expression was analyzed on transmural sections of carcinoma for tumor epithelium in general, of the tumor center, and at the invasive margin, as well as for non‐tumorous colon mucosa. For each compartment, five power fields (×40; 0.23 mm2 per field) were evaluated, and the mean values were calculated. Expression was evaluated as the percentage of positive cells at each of the following intensity categories: 0 to 3+ (0 = negative, 1+ = weak, 2+ = moderate, and 3+ = strong). The H‐score was calculated as follows: H‐score = [fraction of cells with intensity grade 1 (%)] + [fraction of cells with intensity grade 2 (%) × 2] + [fraction of cells with intensity grade 3 (%) × 3], with a possible maximum of 300 points. The H‐score threshold for defining high Globo‐H expression was set at VK9‐TT ≥30, with values