Growing evidence shows that many patients with cancer benefit from immunotherapy (1). The immune system can eliminate cancer cells by recognizing cancer antigens expressed in malignant cells. This cancer immunosurveillance concept is now clearly described by a seven-step cancer immunity cycle (2). Most cancers adopt strategies to evade the immune system after a long struggle between malignant cells and the immune system (3–5). Thus, reactivation of the antitumor immune response and upregulation of the cancer-immunity cycle are critical to ensure improved clinical response. Immune checkpoint inhibitors are currently the most promising treatment for upregulating the cancer-immunity cycle; however, their clinical responses vary from patient to patient due to the complexity of tumor-immune interactions (6). Recent analysis of the mechanisms of immune suppression in cancer revealed that different steps in the cancer-immunity cycle by which tumors escape immunosurveillance are likely to differ among patients (7, 8). Therefore, cancer immunotherapy needs to be personalized to identify the rate-limiting steps in individual patients, and a combination of strategies should be used to overcome these hurdles. To realize personalized precision cancer immune therapy, a technique that can monitor the cancer-immunity cycle will be a powerful tool for treatment.

The immunogenicity of cancer cells is determined by antigen peptides present on MHC class I, and CD8⁺ cytotoxic lymphocytes eliminate cells by recognizing this complex (9). Cancer antigens can be classified into two groups: aberrantly expressed tumor-associated antigens (TAAs) (10) and cancer/testis antigens (CTAs) (11, 12), or neoantigens derived from mutated gene products (13, 14). Both TAAs and CTAs are known as shared antigens that are universally detectable in different patients. In contrast, somatic mutation-derived neoantigens show patient-specific individual variations (15).

During the cancer-immunity cycle activation, antigens released from cancer cells are then captured by dendritic cells, but not all the TAA- or CTA-derived peptides can present on MHC class I. However, these aberrantly expressed antigens could be involved in the humoral immune response. Hence, serum autoantibodies to cancer antigens reflect the current immune response level associated with tumor volume and antigenicity ( Figure 1 ). Once cancer cells are destroyed, many cancer antigens are released from them, captured by antigen-presenting cells, and induce T and B cell immune responses. The activation of the antitumor immune response is accompanied by an increase in the number of autoantibodies against various cancer antigens, referred to as antigen spreading (16–22) which could be a critical pharmacodynamic biomarker for the clinical outcome of cancer immunotherapy (18, 22–26). Accumulating evidence indicates that antibody-antigen immune complex uptake through Fcγ receptors on antigen-presenting cells induces cross-presentation, stimulating long-term antitumor cellular immunity (27). Thus, a simple blood test-based evaluation of antigen spreading with cancer-immunity cycle activation could predict systemic cancer immunity.

The diagnosable set of cancer antigens requires a comprehensive array because the expression pattern, antigenicity, and epitopes vary in individual patients (23, 28, 29). Full-length cancer antigen preparation is favored for monitoring antigen spreading based on these requirements. However, many recombinant CTAs appear to show an aggregation-favored unstable property (30). This property is consistent with the bioinformatics prediction of the structure of CTAs, which shows that the majority of CTAs are intrinsically disordered proteins (IDPs) (31). These IDPs, or IDP regions, lack rigid tertiary structures under physiological conditions in vitro; however, they can fold after binding to target macromolecules in vivo ( 32). Thus, recombinant CTA proteins in a disordered conformation frequently form inclusion bodies in host cells (33).

In order to quantitatively evaluate antigen spreading, we designed a multiple S-cationized antigen-immobilized bead array (MUSCAT) assay system (33). S-cationization techniques are employed as a powerful solubilization tool by conjugation of cationic moieties in sulfhydryl groups in denatured protein (34–36). Full-length and water-soluble S-cationized antigens were covalently immobilized onto Luminex magnetic beads via the activated carboxylic acid of the COOH radical group (33, 37). Although immobilized S-cationized antigen on beads modified all Cys residues and limited amino-groups employed for immobilization, specific antibodies raised in cancer patients were quantitatively detected by the Luminex assay with high sensitivity. This MUSCAT assay system can detect polyclonal antibodies recognizing the linear epitope. Epitope-mapping study of anti-CTA autoantibodies in patient sera revealed that these antibodies are polyclonal and recognize individually different linear epitopes (28). Most CTAs are predicted to have no rigid ordered conformation (31), so antibodies recognizing conformational epitopes are most likely rare. Furthermore, recent knowledge-based and in silico analyses of linear epitopes showed that Cys is a minor frequent amino acid residue (38). Chemically modified Cys has minimal effect on the antibody binding efficiency. Thus, the MUSCAT assay system is a powerful strategy for quantifying antigen spreading to diagnose cancer immunotherapy.

Monitoring antigen spreading by the level of autoantibody biomarker requires validated positive control to ensure diagnostic accuracy. This study demonstrated that water-soluble reversibly S-cationized CTAs can be employed in both antigens recognized by serum antibodies on the MUSCAT assay and antigens to immunize rabbits to develop antibody-positive controls. Subsequently, this quality-certified serum autoantibody monitoring system also demonstrated successful immune monitoring using clinical samples. In a clinical trial, adenovirus−mediated REIC/Dkk−3 (Ad-REIC) gene therapy has shown a successful clinical response in metastatic castration-resistant prostate cancer (39–41). Ad-REIC is known to induce cancer cell-specific apoptosis (42) and activate the antitumor immune response (43, 44), so this serum sample was used to evaluate the MUSCAT assay system in the current study. It was confirmed that the elevated autoantibody biomarker closely related to the clinical response showed potential to assist clinical decisions by monitoring the level of the cancer-immunity cycle.

A recent oncological study revealed that cancer is heterogeneous (49). Heterogeneity in cancer is not only limited to different patients but also occurs within a single patient (50). Certain tumors can be analyzed by molecular profiling yielding clinically relevant diagnostic and prognostic results (49). However, intrapatient or intratumoral heterogeneity remains a significant challenge for deciding the course of clinical cancer treatment. As such, activating the cancer-immunity cycle is a reasonable strategy to overcome this heterogeneity issue (49). The death of the heterogeneous population of cancer cells can be a trigger for priming naïve T cells by newly exposed antigens, resulting in the activation of the cancer-immunity cycle. Antigen exposure from cancer cells also enhances pre-existing memory T and B cells. Upon induction of these immune responses, T cell-mediated cellular immunity targeting various tumor antigens may help regress heterogeneous cancer. Humoral immune responses also increase during the cancer-immunity cycle activation ( Figure 1 ); therefore, anti-CTA/TAA IgGs are biomarkers that can reflect this anticancer immunity. The IgG autoantibody biomarker is superior in physicochemical aspects because IgGs in peripheral blood are relatively stable proteins (51). Furthermore, the bias in sampling or storage conditions are nearly negligible.

The preparation of a comprehensive array set of CTAs/TAAs is preferable for reliable evaluation of antigen spreading. Each antigen-specific antibody assay must be validated for its specificity and sensitivity for clinical use. In this study, the water-soluble full-length reversibly S-cationized TAPS-antigen was demonstrated to employ both antigens to capture specific antibodies in assay and antigens to develop the antibody in the immunized animals. Purified antibodies reliably validated control in the MUSCAT assay system. Most CTAs have unstable and aggregation-favored properties (30); therefore, S-cationization is a powerful technique to prepare highly purified and water-soluble antigens. We are currently preparing comprehensive sets of CTAs/TAAs and each specific antibody for use as a validation tool using these procedures.

The preparation of CTA-specific antibodies also allowed us to gain an understanding of the structural properties of the CTAs ( Figure 4 ). TAPS-CTAs are fully denatured, so antibodies immunized by them recognize linear epitopes. However, these antibodies also recognize intracellular endogenous CTAs under non-denaturing conditions. This strongly suggests that the conformation of CTAs has no rigid conformation in cells. CTAs are originally found in testicular cells and are predicted to form a cooperative structure (31). However, aberrantly expressed CTAs in cancer lack this cooperative expression. Therefore, the CTAs released from apoptotic cancer cells are supposed to be easy to denature or decompose. Antibody-producing cells proliferate that bind to cancer antigens released from dead cancer cells; thus, most anti-CTAs bind to the denatured form of protein. A class switch from IgM to IgG also requires antigen binding and cytokine stimulation (52). This mechanism explains why anti-CTAs are recognized preferably in linear epitopes ( Figure 1 ).

Many CTAs, including MAGE-A4, NY-ESO-1, and XAGE-1b, are located on the X chromosome and are epigenetically regulated. Lack of X chromosome inactivation in cancer cells induces aberrant epigenetic changes and CTAs expression. Highly immunogenic cancer cells, expressing CTAs, tend to be targeted by the host immune system; thus, suppression of CTAs by DNA methylation could be a mechanism of immune escape in cancer (53, 54). DNA methylation inhibitor treatment, to upregulate the immunogenicity of tumors, would therefore facilitate targeting by the host immune system (55). Ovarian cancer cell line SK-OV-3, has been reported as a model for reactivation of NY-ESO-1 by DAC (56). Increased levels of CTAs after DAC treatment were detected using anti-CTAs antibodies ( Figures 4C, D ). The polyclonal antibodies, immunized full-length CTAs, detected endogenous antigens and their transcript variants. Interestingly, XAGE-1b includes various uncharacterized immune reactive variants ( Supplementary Figure 2 ). Determination of these target antigens is crucial for enhancing the accuracy of the immune monitoring system.

As discussed above, both antigen exposure by inducing apoptosis in cancer cells and releasing immune-suppressing machinery is critical to upregulate the cancer-immunity cycle. Intratumoral injections of Ad-REIC therapy are potent in inducing both effects (39). In this therapy, Ad-REIC was injected into the metastatic lymph nodes. This strategy may contribute to the upregulation of the cancer immunity cycle because the released antigens from apoptotic cells exist in lymph nodes and stimulate T and B cells efficiently ( Figure 1 and Figure 6 ). Although the validation of the MUSCAT assay has not yet been completed by positive control of antibodies, this panel showed successful immune monitoring by detecting upregulation of several antibodies in the high responders by Ad-REIC cancer gene therapy. It is crucial for the diagnostic use of the MUSCAT assay that antigen spreading be observed before tumor regression. Baseline analysis of these autoantibodies showed individual variations in ten healthy donors ( Figure 6A ) although it is unclear whether these variations reflect the immune history of the individuals. However, monitoring autoantibody changes during therapy will evaluate the generation of cancer-immunity cycle and autoimmune responses. The concept of immunologically “hot” or “cold” from the outcome of immune checkpoint inhibitor therapy is widely accepted today, and the drug discovery to achieve “cold-to-hot” tumor conversion is a great challenge (57). Thus, a tool to monitor this conversion is critical. A highly accurate and quantitative autoantibody monitoring system will be an excellent tool for developing a therapeutic strategy to overcome refractory cancer.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Okayama University Hospital. The patients/participants provided their written informed consent to participate in this study.

AM, KK, and JF conceived the project. AM, TH, MM, and RK performed the experiments. AM and JF prepared the manuscript. HK analyzed the clinical data. All authors contributed to the article and approved the submitted version.

This work is partially supported by JST START Grant Number JPMJST1918 (JF) and supported by Science and Technology Promotion grants (2019–2021) in Okayama Prefecture, Japan.

Okayama University and Medinet Co. Ltd. are holding patents on the method for MUSCAT assay.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.