Colorectal cancer (CRC) remains a major cause of cancer-related morbidity and mortality worldwide, and patient survival is highly dependent on the stage at which the disease is detected. While five-year survival exceeds 90% for individuals diagnosed with localized disease, outcomes decline dramatically once CRC progresses to advanced stages. These stark differences highlight a critical unmet need for minimally invasive biomarkers capable of enabling early detection of CRC and identifying high-risk precancerous lesions in asymptomatic populations, where intervention is most effective. 

​

Blood-based biomarkers offer a compelling and clinically actionable approach to CRC screening, risk stratification, and early detection. To date, the development of blood-based biomarkers has focused on the detection of circulating tumor DNA and abnormal plasma cytokines. However, tests of circulating tumor DNA under-perform in the early stages of colon cancer, likely because of low levels of tumor DNA in the peripheral blood. The measurement of plasma cytokines is promising because of the ease of testing; however, blood is a “mixing bowl” for cytokines produced anywhere in the body, from any physiologic process.  Cytokine signals from an early-stage tumor, not yet having remodeled surrounding tissue, may be extensively diluted in the peripheral blood, to the point that they cannot be detected. Moreover, cytokines produced in early- stage tumors may be consumed locally by immune cells.  

​

The immune system is a well-recognized defense mechanism for the body, performing constant surveillance for transformed (cancerous) cells. The proteins produced by cancer cells, known as tumor antigens, mark these cells for elimination as an initiating step in an anti-tumor immune response.  CD4+ T-cells, specifically recognizing portions of the tumor antigens, proliferate to increase in number and release cytokines. These cytokines recruit other immune cells and amplify local anti-tumor responses. In addition, CD8+ T-cells identify transformed cells displaying tumor antigens and kill them.  These early immune responses launch a race between the tumor – to evade the immune response and grow – and the body’s ultimate detection and killing system. 

​

Regardless of the outcome of this race, the T-cells involved in the early, anti-tumor immune response persist in the body (particularly in the blood), through a mechanism known as immunologic memory. We can detect these cells in the laboratory by re-stimulating the cells with tumor antigens, and observing the up-regulation of cellular activation markers on the subset of cells that recall the tumor antigens. We hypothesize that individuals who harbor transformed (pre-cancer or cancerous) cells will have a higher frequency of T-cells expressing activation markers than healthy individuals in our recall tumor-antigen stimulation assay. 

​

The stimulation assay central to our testing is well-established across a wide-range of settings, invented and first reported in a Nature Medicine paper in 2005 by Talon’s founder and CEO, even adapted for highly regulated vaccine trials. In developing the platform, Talon100 carefully considered what tumor antigens would be used for the assay. To make the assay platform broadly applicable, in the widest range of test subjects, lysates from colorectal cell cancer lines will be used. Three concerns could be raised by use of lysates from cell culture lines. First, immune cells from healthy (control) individuals may mount responses against the tumor lysates, which will be seen as non-self (foreign). This concern is mitigated by the fact that healthy donors have not had in vivo exposure to the lysate; thus, these responses will be low level, infrequent naïve (non-recall) responses.  In contrast, test subjects with colorectal tumors should have much stronger, higher-frequency recall responses. Second, a theoretical concern is that the tumor lysates collected from colorectal cell cancer lines will not match in vivo tumor antigens. In fact, published analyses have shown an average of 70% similarity in gene upregulation between colon cancer lines and patient tumor samples, mitigating this concern.  Third, evolution of tumor antigens occurs in vivo, leading to profiles that are unique to a patient over time (and thus discordant with cell culture antigens common to all individuals). This concern is mitigated by the fact that memory immune responses are retained for decades against antigens, even after they are cleared by the body. In sum, the rationale for using a platform reliant on a patient’s own T-cell responses against tumor cell lines is strong. 

 

The work presented here is the first step toward future studies validate immunologic profiling, comparing to liquid biopsy-based tests of circulating tumor DNA and methylation patterns. These platforms may be combined in the future not only to increase test sensitivity, but also to stratify patients, identifying subsets of individuals at greatest risk for colon cancer, and in whom early detection tests might be most valuable. In fact, either or both tests may be most sensitive in individuals pre-disposed to colorectal cancer, based on genetic risk factors.  These combined testing modalities offer additional clinical and commercial pathways.