Cytolytic T-cell response against Epstein-Barr virus in lung cancer patients and healthy subjects

Evidence suggests that cancer patients present with a compromised immune response of multifactorial origin, including the tumor itself. It seems that the early stages of tumor growth appear not to elicit systemic immune deficiency and are sometimes associated with antigen-specific tolerance, while generalized immunodeficiency can arise during the late stages of tumor development [1]. Related data are mainly derived either from in vitro experiments or from DTH measurements in the context of cancer immunotherapy [2]. Therefore, the existing evidence remains inconclusive, while the significance of the described immune alterations in relation to the ability of cancer patients to mount effective responses against pathogens has not been clarified. Finally, there is existing controversy regarding the efficacy of influenza vaccination in patients with cancer [3, 4].

This study was scheduled in order to examine whether, at diagnosis, EBV seropositive patients with lung cancer, have a compromised virus-specific CTL response, as compared to age-matched healthy controls. A group of younger healthy individuals was also examined to ascertain whether a possible reduction in the anti-EBV CTL responses of the above patients and age-matched controls could be attributed to senescence. Lung cancer was selected because although such cancers express several tumour antigens [5] and T cells infiltrating these tumours have been identified [6], the outcomes of specific immunotherapy for patients with lung cancer is rather poor [7].

EBV-specific CTL responses were detected in the peripheral blood of 8/19 lung cancer patients (42%) and 5/14 (36%) aged-matched controls (p = 0.713). Both of these proportions were statistically significantly different than 86% (6/7) of younger healthy individuals (p = 0.048 and p = 0.031, respectively) that presented with an EBV-specific CTL response (Figure 1). When we examined whether corresponding alterations could be observed against other viruses such as CMV, our findings indicated that the anti-CMV response was similar to that described in the literature [9]. Hence, although all subjects had prior exposure to CMV as determined by serology, younger individuals appeared to have a lesser response as compared to aged individuals (p = 0.046) and aged individuals had a higher response than that observed with patients (p = 0.025) (Table 1).

Despite the above, the frequency of EBV specific CTL amongst peripheral CD8 T cells was almost the same in the subjects of all groups (Table 2). In all groups, the response against the BMLF1.A2 peptide was more frequent than that against the EBNA3C.A24 peptide (7 patients out of the possible 13, 3 aged-matched controls out of the possible 9 and 6 younger healthy individuals out of the possible 7).

In the process of determining the pCTL frequencies in the peripheral blood, we collected and evaluated flow cytometric data obtained from the analysis of each individual MLPCs. Interestingly, although MLPC containing a multimer positive population, amongst all three groups appeared to have similar multimer positive populations (Figure 2), interesting findings were observed when these were analysed in detail. In particular, the mean percentage of multimer⁺CD8⁺ T cells inside the positive MLPCs was found significantly higher (p < 0.0001) in age-matched healthy subjects (26.6 ± 26.4%, range 0.4--80.7%) than in lung cancer patients (2.7 ± 3.3%, range 0.1-19.0%) and younger healthy individuals (2.4 ± 1.7%, range 0.2-7.0%) (Figure 3A). This reflects an increased proliferative capacity against the antigenic stimulus of the peptide-specific pCTLs in the older healthy subjects. On the other hand, no statistically significant difference was observed among the three groups with respect to the intensity of multimer binding by each multimer positive population (patients; MFI 6.9 ± 12.3, range 2-115, older healthy subjects; MFI 6.0 ± 4.1, range 2-23, younger healthy subjects; MFI 5.1 ± 3.7, range 2-19) (Figure 3B). This indicates that all antiviral T cells had TCR with a similar avidity towards the peptide/MHC complex and no difference in the kinetics of interaction between TcR and multimer complexes could be observed [10]. Regarding the above, a significant correlation was observed between the percentage of multimer⁺CD8⁺ and the multimer MFI within the patient (r = 0.15, p < 0.0001) and the aged-matched healthy individual group (r = 0.504, p < 0.0001) but not within the young healthy individual group (r = 0.016, p = 0.435).

Finally, we examined whether the presence of an anti-EBV CTL response lung cancer patients correlated with any clinicopathological parameter (age, sex, performance status, loss of weight, stage of disease etc). No significant correlations were uncovered with either group (Table 3).

This study provides direct evidence that lung cancer patients dispose an EBV-specific CTL response equivalent to that of age-matched healthy counterparts. Moreover, it was demonstrated that the EBV-specific CTL response mounted by subjects of this age group, either with cancer or not, was twice as less than that elicited by younger healthy individuals. Regarding the healthy individuals, our results are in accordance to those reported recently by Colonna-Romano et al [11] demonstrating an inverse correlation between age and the percentage of circulating EBV-specific CTLs. Most likely, these observations can be explained in the context of the complex process of T cell immunosenescence [9, 12].

With respect to cancer patients, it is interesting that they present with the same age-related alteration of EBV-specific CTL response as their healthy counterparts. In other words, neither the antigenic burden of the tumor nor any other cancer-related factor affected their ability to mount a CTL response against the virus. Assuming that the CTL response of cancer patients against other pathogens follows a similar pattern of alterations, no special vaccination strategy [4] is required other than that followed for elderly people in general, except when they are under the influence of immunosuppressive therapies. To this end, it must be noted that considering the low frequencies detected in our study population (3-60/million CD8), one has no other alternative but to attempt to amplify these cells first in order to understand their reactivity. This is not unusual since other we and others have confirmed that in most cases CTL responses detected after in vitro stimulation reflect the true number of these cells circulating in vivo [13, 14].

Beyond differences observed in the specific pCTL frequency related to age, cancer patients also appeared with a decreased proliferative capacity of virus specific pCTL. Most likely these differences could be explained by replicative senescence [15, 16], whereby viral specific CTL in patients have multiplied several times over their lifetime and present with a reduced ability to further respond to an antigenic stimulus. This does not exclude their presence but rather supports the fact that T cell clonal exhaustion results in the accumulation of oligoclonal dysfunctional cells followed by repertoire shrinkage due to clonal deletion, maintaining however, the actual number of dysfunctional cells [17], as has recently being demonstrated in patients with renal cell cancer [18].

Many investigators relate the immune dysfunction of cancer patients with both the inefficient anti-tumor response and a reduced efficacy of immunotherapy [19, 20]. To this end, we have recently identified that patients with lung cancer present with a tenfold higher number of anti-tumor CTL as compared to the age-matched controls [13]. These results suggest that such patients do not have an immunocompromised CD8 T cell response but the ineffective anti-tumor response, is most likely a reflection of the age-associated changes that take place in individuals [21] impacting on their capacity to respond effectively against the tumor. Under the light of the data presented herein, it is worth examining whether young individuals have a more robust anti-tumor response, as is the case with the anti-EBV response.

In conclusion, this study provides evidence that lung cancer patients dispose an EBV-specific CTL response equivalent to that of age-matched healthy counterparts. Our study suggests that possibly the poor outcome of cancer immunotherapeutic approaches in lung cancer can be a result of the underlying effects of senescence on the immune system rather than an inefficient anti-tumor response. These data warrant the examination of whether young individuals have a more robust anti-tumor response, as is the case with the anti-EBV response.

This work was supported by (a) a European Union - European Social Fund (75%) and the Greek Ministry of Development-GSRT (25%) (ENTER 04EP09) grant and (b) a Marie Curie Incoming International Fellowship within the 6th European Community Framework Programme (FP6 Contract MIF1-CT-2006-021795, IRTALUNG) grant.