T cell receptor affinity in the age of cancer immunotherapy

Abstract

The strength of the interaction between T cell receptors (TCRs) and their ligands, peptide/MHC complexes (pMHCs), is one of the most frequently discussed and investigated features of T cells in immuno-oncology today. Although there are many molecules on the surface of T cells that interact with ligands on other cells, the TCR/pMHC is the only receptor-ligand pair that offers antigen specificity and dictates the functional response of the T cell. The strength of the TCR/pMHC interaction, along with the environment in which this interaction takes place, is key to how the T cell will respond. The TCR repertoire of T cells that interact with tumor-associated antigens is vast, although typically of low affinity. Here we focus on the low affinity interactions between TCRs from CD8+ T cells and different models used in immuno-oncology.

Introduction

The binding strength of T cell receptors (TCRs) on CD8+ T cells to specific MHC class I molecules complexed with self peptides (a.k.a. tumor-associated antigens, TAAs) is generally weak¹. On average, TCRs bind to these self pMHC molecules at 10 times lower binding affinity than to foreign antigens². These naturally selected self antigen TCR affinities are in the range of 1–100 μM³. Many tumor antigens are, in fact, self antigens that have become dysregulated during the transformation process^4–6. Frequently, these tumor-associated antigens (TAAs) are much more highly expressed in the tumor than the periphery and can be shared among different tumors. For example, expression of the prototype TAA NY-ESO-1 is only found in the immune-privileged testes, but is expressed in many different kinds of tumors⁷. The shared expression of TAAs, in opposition to neoantigens, makes them attractive targets for immunotherapy. In addition to TCRs binding to many peptide-MHC molecules on an antigen presenting cell, numerous interactions between molecules other than TCR-pMHC contribute to the avidity of the interaction. However, in this review we focus primarily on the TCR-pMHC interaction and refer to the strength of this binding as the “affinity.”

T cell fate is determined by TCR engagement and downstream signaling in both the thymus and periphery. To avoid autoimmune responses, but protect from pathogens, T cells in the periphery must be tolerized against self, yet strongly recognize foreign pathogens. In the thymus, the transcription factor AIRE aides expression of self antigens on thymic epithelial cells, facilitating their presentation to developing T cells⁸. Strong interactions lead to negative selection and deletion of self-reactive T cells, or alternatively, T cell differentiation into regulatory T cells^9,10. The K_D of ~6 μM, just about the negative selection threshold, is termed the “danger zone” for autoantigens, as negative selection functions more robustly at higher affinities¹¹. Other experiments examining antitumor immune responses and autoimmunity confirmed these results and identified 10 μM affinity as the optimal affinity for interaction with a TAA¹².

Many TAAs are unmutated self-antigens^13,14, which include post-translationally altered-self, oncofetal embryonic, cancer testes, idiotypic/clonal, and overexpressed tissue differentiation antigens (reviewed in¹⁵). As a result of negative selection, T cells targeting TAAs express TCRs with lower affinity for their antigenic ligands as compared to pathogenic and autoimmune T cells (summarized in Figure 1)¹⁶. For example, the prototypical TAA TRP-1 is expressed in the thymus by the transcription factor AIRE¹⁷. As discussed below, these lower affinity interactions differently stimulate CD8+ T cells and have suboptimal, somewhat truncated immune responses¹⁸. Results in both human and mouse studies have proposed that high functional avidity and affinity thresholds of the TCR are required for efficient antitumor immunity¹⁹. Lower affinity interactions, like those corresponding to normal tissue antigens, can go unrecognized. To analyze the importance of T cell affinity in cancer immunotherapy, appropriate models are necessary.

Figure 1.

Average T cell receptor binding affinities to differing types of antigen-MHC complexes in the thymus and periphery. Self antigen, TAA, and positive selection are considered low-affinity TCR-pMHC interactions.

Detecting T cell receptor affinity

T cell affinity is defined as the probability that a given TCR will interact with a pMHC ligand. The strength of this interaction predicts how the T cell functions when it comes in contact with a specific antigen. Affinity is commonly reported as an equilibrium dissociation constant K_D, which is the ratio of the kinetic off (k_off) and on (k_on) rates. The smaller the K_D value, the stonger the equilibrium binding affinity of the ligand to its receptor. The interaction between the ectodomains of the TCR and pMHC is measured as the amount of free, unbound TCR in equilibrium with bound TCR-pMHC and is independent of ligand:receptor concentrations or kinetic rates^3,20. The half life (t_1/2) or dwell time of an interaction is a kinetic measurement derived from the rate of dissociation of the TCR-pMHC interaction. Typically the K_D is proportional to the t_1/2, but there are also reports of disparate values between the two³. Low affinity interactions, like many TCR-pMHC interations, are often characterized by rapid dissociation rates, which can make it difficult to use kinetic measurements to determine affinity²¹. One method to estimate the K_D under these conditions is to use Scatchard plots. The measurements of K_D, k_off, k_on, and t_1/2 have been used to describe many different facets of TCR and T cell function (reviewed in³). Here, we do not try to cover all of this research, but focus on the K_D, or binding equilibrium constant, of reported “low affinity” interactions.

One standard technique to determine these binding constants is surface plasmon resonance (SPR). SPR allows a ligand in solution to associate to an anchored receptor (or solution phase receptor to bind to an anchored ligand), resulting in a change in mass on the immobilized molecule. The affinity constants are determined by measuring the change in instrument response by varying the concentration of the ligand in solution in real time. Although SPR is valuable to make affinity measurements, T cell activation depends on numerous other factors (such as other molecules, membrane confinement, immune synapse formation) for signaling, and affinity measurements do not account for any additional forces, such as co-receptor binding²². The CD8 co-receptor modifies the measured affinity of the T cell receptor up to ten-fold²³. As mentioned above, the equilibrium constant K_D by definition only measures the affinity of the TCR ectodomain portion of the T cell receptor to a given ligand. However, the measured affinity between the entire TCR and pMHC is modulated in the context of the entire TCR-CD3 multi-subunit complex. Therefore, depending on the question being asked, techniques to predict relative binding in the context of the whole T cell interaction with an antigen presenting cell may be more appropriate, such as FRET, FRAP, or the 2D micropipette adhesion frequency assay (2D-MP)^24,25. Lastly, in addition to SPR, affinity measurements for low-affinity interactions can be enhanced using thermodynamic techniques, such as titration calorimetry²¹.

In addition to direct measurements, affinity is often indirectly detected by manipulation of the TCR ligands. Such measurements of the collective strength of the functional and physical interactions leading to T cells responses are also known as “functional avidity.” Detection of an antigen-specific T cell is commonly performed using soluble recombinant multimeric pMHC. pMHC molecules are standardly multimerized into dimers, tetramers, pentamers or octamers²⁶; typically, monomeric pMHC does not bind with high enough affinity to detect binding. Multimers are conjugated to a fluorophore for detection via flow cytometry or via a barcode for single cell analysis. Intensity of staining with multimers can act as a proxy for affinity measurements; the higher affinity T cells bind the tetramer better than lower affinity T cells, assuming similar expression of TCR and co-receptor^27–29. However, higher tetramer binding does not always correlate with functional T cell responses^30,31. For example, upon activation, TRP1-tetramer^LO and TRP1-tetramer^HI T cells similarly arrest tumor growth in a B16 model³⁰. In addition, if the T cells being compared express different amounts of TCR or co-receptor, the affinity correlation may not hold true. Multimer detection has recently been described to introduce a bias towards high-affinity interactions, underestimating the lower affinity components²⁵. Because these reagents are multimerized, the avidity of the interaction biases the interpretation of what the affinity of multimer-specific T cells may be. Therefore, a more accurate interpretation of multimer-binding would be detection of functional avidity, not binding affinity.

Many signaling molecules and transcription factors act as rheostats downstream of the TCR-pMHC interaction to control T cell function. These molecules translate the strength of the interaction between TCR and pMHC into a T cell response. Many of these rheostat experiments were performed using altered peptide ligands (APLs) of the SIINFEKL peptide, which is encoded by the chicken ovalbumin protein³². These peptides complex with the murine H-2K^b MHC molecule and are recognized by the CD8+ OT-1 transgenic T cell. Negative costimulatory molecules such as CD5^33–35 and tyrosine phosphatase PTPN22³⁶ have been found to inhibit T cell signaling in an affinity-dependent manner using the OT-1 system. CD5 expression in the periphery appears to be linked to THEMIS, a TCR signalosome component. THEMIS is constitutively associated with the phosphatases Shp1 and Shp2 to regulate signals from very low-affinity pMHC interactions by altering cellular metabolism that drive CD8⁺ T cell proliferation^37,38. THEMIS also plays an important role in the positive selection of low affinity T cells (reviewed in³⁹). The phosphatase PTPN22 was found to limit conjugate formation between LFA-1 on the T cell and the ICAM ligand molecule on the antigen presenting cell; knocking out the LFA-1 gene in the T cell altered the activation threshold³⁶. Similarly, using the OT-1 APL system, it was determined that the transcription factor IRF4 is critical for increasing glycolysis and metabolic function required for T cell activation of high-affinity T cells; low-affinity T cells show diminished glucose uptake⁴⁰. Lastly, the orphan receptor Nur77 is commonly used as a marker for TCR signal strength^10,41. In the OT-1 system using a Nur77^GFP-reporter mouse, Nur77 expression increased as TCR-pMHC affinity increased in the thymus¹⁰. The effect of these rheostats was determined using transgenic mice specific for a foreign antigen and elucidates the usefulness of these approaches.

Current murine affinity models

Altered peptide ligands of both human and mouse native antigens are used to model TAAs. Low-affinity and/or positive selecting ligands in mice transgenic for the antigen and/or the TCRs may mimic endogenous low-affinity interactions. Much of the TCR affinity research has been performed in genetically pliable mice and has investigated the T cell response to model foreign antigens. However, the foreign transgenes used in these studies present thymic-selection-related limitations. Additionally, the binding affinities of bona fide tumor antigens are often lower than what a transgenic or altered peptide model can provide. A table of the affinities of native MHC class I-restricted peptides and their corresponding APLs for use in transgenic CD8+ T cell models is provided (Table 1^42–52.) This list summarizes the low-affinity altered peptides and TCRs available for use in murine in vivo oncoimmunotherapy affinity models.

Table 1.

Transgenic CD8+ T cell mouse models with known high and low affinity TCR-pMHCI interactions

TCR	Peptide/MHC	Peptide Sequence	KD (μM)	Ref.
OT-1	OVA/Kb	SIINFEKL	6.1	Yachi et al.
OT-1	A2/Kb	SAINFEKL	4.4	Yachi et al.
OT-1	G4/Kb	SIIGFEKL	10.1	Yachi et al.
OT-1	E1/Kb	EIINFEKL	20.8	Yachi et al.
OT-1	R4/Kb	SIIRFEKL	48.9	Yachi et al.
CT	A5/Ld	SPSYAYHQF	1.2	Slansky et al.
CT	AH1/Ld	SPSYVYHQF	5	Slansky et al.
ID4	A5/Ld	SPSYAYHQF	5	Burhman et al.
ID4	AH1/Ld	SPSYVYHQF	>100	Burhman et al.
2C	p2Ca/Ld	LSPFPFDL	3.3	Garcia et al.
2C	QL9/Ld	QLSPFPFDL	3.9	Garcia et al.
2C	SIY/Kb	SIYRYYGL	31.9	Garcia et al.
2C	dEV8/Kb	EQYKFYSV	84	Garcia et al.
AHIII	p1058/Db	FAPGFFPYL	81.4	Buslepp et al.
AHIII	p1049/HLA-A2	ALWGFFPVL	11.3	Buslepp et al.
P14	gp33/Db	KAVYNFATC	12.2	Tian et al.
P14	Y4A/Db	KAVANFATC	63	Tian et al.
P14	KIMC9M/Db	KIMYNFATM	74.5	Tian et al.
P14	KIR/Db	KIRYNFATC	81.6	Tian et al.
P14	KISC9M/Db	KISYNFATM	267	Tian et al.
P14	V3LC9M/Db	KALYNFATM	264	Tian et al.
P14	Y4SC9M/Db	KAVSNFATM	264.1	Tian et al.
HY	HY/Db	KCSRNRQYL	171 *	Maile et al.
HY	C2A/Db	KASRNRQYL	221 *	Maile et al.
HY	K1A/Db	ACSRNRQYL	1159 *	Maile et al.
TRP1	M9/Db	TAPDNLGYM	NA - peptide stabilization	Clancy-Thompson et al.
TRP1	K8/Db	TAPDNLGKM	NA - peptide stabilization	Clancy-Thompson et al.
PMEL	hgp100/Db	KVPRNQDWL	NA - peptide stabilization	Overwijk et al.
PMEL	mgp100/Db	EGSRNQDWL	NA - peptide stabilization	Overwijk et al.
CL4	HA/Kd	IYSTVASSL	NA - EC50 410pM	Hartemann-Heurtier et al.
CL4	A517G/Kd	IYSTVVSSL	NA - EC50 15pM	Hartemann-Heurtier et al.
N-TgN HI	RNEU/Dq	PDSLRDLSVF	NA – tetramer	Weiss et al.
N-TgNLO	RNEU/Dq	PDSLRDLSVF	NA - tetramer	Weiss et al.

^*binding affinity in female mice

In addition to these reported transgenic models, there are many other systems in which affinity has been modeled. When searching for native ligands and APLs for human T cells, CD4+ T cells, or modulating affinity in a system by altering the T cell instead of the peptide, a variety of manually curated databases exist. Webservers such as ATLAS (Altered TCR Ligand Affinities and Structures) serve as an important resource for searching binding affinities between wild type or mutant TCRs and their corresponding antigens⁵³. Additional databases such as VDJdb⁵⁴, iMatrix⁵⁵, PAComplex⁵⁶ and MPID-T2⁵⁷ are sequence-structure-function based tools that allow users to interrogate TCR, peptide and MHC sequences, their structures, and cellular outcomes. Some of these programs can input the structural models and associated TCR and pMHC data to predict peptide binding for antigen discovery. Such programs are being developed to take bioinformatic predictions and translate them into “immunogenicity scores” for discovery of potent epitopes⁵⁸. These transgenic models and predictive databases allow researchers to model affinity and can be advantageous when modeling novel and known TAAs.

Differences between low- and high-affinity TCR-pMHC interactions

Low-affinity T cells play an active role in immune responses²⁵. The affinity of the signals thymocytes receive during thymic development dictate their pathway of differentiation. Using the OT-1 APL peptides, an affinity of 20–60 μM was found to induce positive selection, while peptides of stronger affinities (K_D ~6 μM) stimulate negative selection^59,60. Compartmentalization of the Ras/MAPK signaling pathway detects these threshold differences⁶¹. Once matured and in the periphery, T cells are much less sensitive to signals received through the TCR than their thymocyte counterparts.

Outcomes of signaling in mature T cells depend on the affinity of the pMHC interaction with the TCR. Early studies using mature OT-1 CD8+ T cells and APLs reported weak TCR interactions preferentially inducing homeostatic proliferation, suggesting differences in signaling⁶². In both the T1 and OT-1 APL mouse models, only agonist peptides form the interactions required to couple TCR binding and signaling^63,64. Similarly, only the agonist peptide in this system induces CD3 conformational change in the CD3 subunits of the TCR⁶⁵. Using the OT-1 APL system, low-affinity TCR-pMHC interactions are sufficient to activate naive cells, induce proliferation, and stimulate T cell maturation into effector and memory cells³². However, sustaining these signals is impaired in the low-affinity interactions and activation is slowed⁶⁶. Interestingly, OT-1 transgenic mice with a point mutation in the TCRβ chain may also provide insights into how affinity influences T cell fate⁶⁷. Although memory T cells are generated in these mice, memory responses to high-affinity interactions are deficient. Low-affinity ligands trigger more central memory CD8+ precursors, which express less T-bet and more Eomes transcription factors⁶⁷. Weak interactions result in efficient NF-κB signaling that persist in immune responses. Using the OT-1 APL system, low-affinity-primed memory cells produce more IL-2 and fully differentiated memory cells that are CD45RB^{HI 68}. Using specific T cells with low and high affinity for the NY-ESO-1 TAA peptide complexed with HLA-A2, the high-affinity interactions resulted in transient signal activation leading to poor MAPK production and low proliferation ability⁶⁹. Taken together, these data suggest that low-affinity T cells have a separate, but necessary role in T cell function (summarized in Figure 2).

Figure 2.

Functional differences in CD8+ T cells between low- and high-affinity TCR-pMHC interactions. Relative to high-affinity T cells, low-affinity T cells play a distinct role in development, effector function, memory function and in the tumor microenvironment. Both are crucial contributors to the immune response.

The CD8 co-receptor modulates the strength of the TCR-pMHC interaction by binding to the alpha 3 domain of the MHC class I molecule⁷⁰. Although the affinity of the TCR-pMHC interaction is unchanged by CD8, the overall stability of the complex is enhanced as the number of receptor-ligand interactions is increased. Low-affinity APLs delay CD8 co-receptor interactions, suggesting that CD8 binding contributes to low-affinity signals more than high-affinity signals^23,42. Consistent with this framework, in a naturally occurring HCV model a high-affinity human TCR is more capable of CD8-independent functional responses than it’s low affinity counterpart⁷⁰. These findings have practical implications: detection of low-affinity antigens using TCR multimers may depend more on CD8 binding than the high-affinity T cells.

Current dogma on CD8+ TCR affinity predicts that high-affinity tumor infiltrating lymphocytes (TILs) are the cells most important for tumor clearance. In a transgenic T cell model against HA, high-affinity T cells show increased effector function, express lower levels of inhibitory receptors, higher levels of activation markers, less pro-apoptotic gene expression, and more expression of effector molecule expression⁷¹. However, low-affinity TILs play a distinct role and are also necessary for tumor clearance. Using the OT-1 APL system with the B16 melanoma tumor model, an intermediate duration of TCR-pMHC interaction promoted release of cytotoxic vesicles and release of effector cytokines⁷². Peptides that provide too short or too long of an interaction altered tumor infiltration and provided little anti-tumor activity. TCR recognition of weak peptides during the priming phase can also change the effector function of antigen-specific TILs⁴⁹. Using two lines of mice with TCRs that recognize the TAA TRP1 bound to H-2D^b with differing affinities (TRP1^HI and TRP1^LO) and a panel of APLs in the B16 tumor model, it was shown that priming with the high-affinity peptide leads primarily to IFNγ production, while priming with a low-affinity peptide leads to a more cytolytic cell, although adoptive transfer of either significantly delays tumor growth⁴⁹.

Using ID8 ovarian carcinoma cells expressing ovalbumin, adoptive cell therapy with high- and low-affinity T cell interactions were examined with T cells from OT-1 and OT-3 transgenic mice, respectively. T cells were transferred into mice that express a membrane-bound form of ovalbumin as a “self” antigen in kidney and pancreas. The OT-3 T cells mediate tumor regression with remarkable specificity for the tumor antigen and no evidence of autoimmunity⁷³. The specificity of the interaction of the T cell with only the tumor cell is a key component of the low-affinity CD8+ T cell response in the tumor. The specificity, combined with the findings that low-affinity TAA-specific CD8+ T cells in the tumor are an active, cytolytic component of the immune response responsible for tumor clearance, suggests a necessary role for low-affinity TILs⁷⁴.

Although high-affinity CD8+ T cells show decreased levels of inhibitory receptors in a transgenic model, high antigen levels in chronic infections also induce an exhausted phenotype⁷⁵. T cells exhibiting high functional avidity show more signs of exhaustion when responding to an infection or tumors⁷⁶. Additionally, IRF4 establishes and sustains T cell exhaustion⁷⁷. With a previously described role for IRF4 in CD8+ T cell affinity⁴⁰, higher affinity T cells may sustain an exhausted T cell phenotype. However, in several studies, exhaustion markers are upregulated in both low and high-affinity interactions in the tumor^73,78. Therefore, the functionality of low- and high-affinity T cells may be abrogated in confounding ways by the tumor microenvironment. Taken together, low-affinity T cells play a distinct role in development, effector function, memory function, and in the tumor microenvironment and are crucial contributors to a fully functional immune response.

Altering the T cell activation threshold

Understanding the role of low-affinity interactions is imperative in learning how to augment responses to these interactions in the context of cancer immunotherapy. Various laboratories have reported that subtle changes in the variable domain of TCRβ can increase the affinity of the receptor using techniques such as directed molecular evolution^79–82, structure based design^83–85, and computational design^86–88, among others. However, strategies to increase affinity of the receptor may not improve efficacy if a threshold affinity is already met¹². Additional factors, such as antigen density, TCR-pMHC stabilization and CD8 co-receptor signaling is also necessary for predicticting anti-tumor therapeutic potential⁸⁹. Adoptive cell therapy of high-affinity CD8+ T cells has had mixed results^18,90,91. Enhancing affinities can also have off-target results; as the affinity for the tumor antigen increases, so does the TCR’s affinity for structurally similar peptides⁹². For example, an anti-MAGE-A3 engineered T cell receptor induced a large inflammatory response resulting in neuronal cell destruction and death⁹³. Techniques such as TCR “fingerprinting” and mRNA engineered T cells have been developed to limit receptor-crossreactivity and off-target effects^94,95. On the other hand, adoptive transfers of endogenous TCRs from TIL can often lead to immune escape. To overcome this hurdle, T cells expressing two additional T cell receptors (TETARs) have been developed and shown to have better efficacy than traditional adoptive transfers⁹⁶. In this system, two TCRs to melanoma TAAs evade immune escape. Altering affinity at the TCR level is an important step in overcoming central tolerance.

“Peptide velcro” has recently been described as a strategy to increase the affinity of low-affinity ligands in the context of cancer, but still maintain the specificity that low-affinity interactions possess⁹⁷. Here, through a combinatorial library approach, a secondary interaction site is engineered independently of the interface between the TCR and pMHC. This site provides a 10-fold affinity gain between TCR and pMHC, enhancing even extremely low affinities of > 100 μM. Similar to adoptive cell therapy, this requires genetic manipulation of patient T cells and infusion of the corrected or higher affinity cells back into the patient.

Peptide vaccination with either endogenous or mimetic antigen has both progressed and had challenges⁹⁸. Vaccines with adjuvant and naturally occurring TAAs stimulate activation and proliferation of antigen-specific cells at a rate of <1% total circulating CD8+ T cells, much less than vaccines to foreign antigen⁹⁹. However, vaccines to specific oncogenic viral antigens such as the HPV vaccine have been successful¹⁰⁰. Mimetic peptides often induce greater immune responses than naturally occurring peptides and have had some success. However, T cells are sensitive to the structural differences; T cells primed with a mimetic peptide do not always recognize the wild type peptide^101,102. An important consideration when using these mimetic peptides as vaccines is determining the cross-reactivity of the T cell repertoire responding to the native peptide relative to the mimetic¹⁰³. Lastly, vaccination of low-affinity mTERT epitopes exhibited potent anti-tumor effects in a murine tumor model, while high-affinity epitopes provided no protection, suggesting that weak binding may be compensated for¹⁰⁴.

In addition to affinity of the TCR to pMHC, high affinity between the peptide for MHC (10 nM) is often considered necessary for tumor regression¹⁰⁵; although this is controversial as neoantigens do not demonstrate strong affinity towards MHC class I¹⁰⁶. Weak binding of the peptide to the MHC molecule may be compensated with increased antigen, or high affinity TCRs as mentioned previously. Changing the affinity of the peptide for MHC and not the affinity of the TCR/pMHC interaction may prove to be an additional strategy; however, as mentioned above, one must consider the extent of cross-reactivity of the T cell repertoire responding to the native peptide relative to the peptides changed to improve binding to the MHC molecule^101,107.

Low-dose radiation decreases the activation threshold of T cells, promoting anti-tumor immunity. Two to four Gy directly affects the T cells themselves, lowering the TCR activation threshold and skewing towards a Th1 population¹⁰⁸. CD8+ T cells are especially sensitive to glycosylation differences because the CD8 molecule is highly glycosylated. Reduction of glycosylation through either enzyme treatment or pretreatment of T cells with neuraminidase increases receptor clustering and sensitivity, similarly decreasing the activation threshold of T cells without the need for adoptive cell therapy^109,110.

Manipulation of TCR triggering can also lower the activation threshold in CD8+ T cells. Induction of TCR clustering by a magnetic field induces aggregation of nanoparticles bound to T cells¹¹¹. This nanoparticle treatment provides a clinical application against B16 melanoma. Because bivalent or multivalent complexes are preferentially triggered at lower affinity interactions¹¹², this treatment may work by lowering the activation threshold of the T cells. Similarly, inducing CD3 conformational change in the TCR-CD3 complex through the use of a monomeric fab fragment against CD3 also decreases the activation threshold¹¹³. In the OT-1 APL model, non-agonist stimulation was enhanced against peptides within a specific range of affinities. The many rational therapeutic designs currently being studied with the goal of augmenting low-affinity TCR-pMHC interactions in TILs highlights the importance of signal strength in immuno-oncology.

Conclusion

An important role for low-affinity T cells in cancer immunotherapy is coming to light. Understanding the hurdles with low-affinity T cells - central tolerance, weak binding, competition with higher affinity receptors and antigen - will delineate how to better design cancer immunotherapeutics against TAAs. Importantly, when studying low-affinity T cells, low affinity must be defined in the context of the model being used. Because of the many tools existing to measure affinity, as well as indirect markers of affinity, confounding results about the important role of low-affinity T cells exist. When recapitulating T cell affinity in the tumor, using antigen models more similar to TAAs may better translate to efficacy in the clinic. The contribution of studying well designed models of low-affinity T cells has significant repercussions in the age of cancer immunotherapy.

Abbreviations

2D-MP

two dimensional micropipette adhesion frequency assay

AIRE

autoimmune regulator

pMHC

peptide-major histocompatibility complex

SPR

surface plasmon resonance

TAA

tumor associated antigen

TCR

T cell receptor

TETARS

T cells expressing two additional T cell receptors

THEMIS

Thymocyte-expressed molecule involved in selection

TIL

tumor infiltrating lymphocyte