In recent years, there has been emerging evidence from both pre-clinical and clinical studies, including our own, which challenge the one-sided view of the role of IFN-γ producing T cells in protecting the host against cancers. For example, IFN-γ was shown to promote immune editing and subsequent tumor escape in the CT26 colon carcinoma by down-regulation of the expression of gp70 immunogenic tumor antigen at the mRNA levels [
1]. Recently, it was demonstrated that inhibiting expression of IFN-γ receptor (R) by forcing expression of the dominant negative IFN-γ R reduced the ability of renal carcinoma cells to metastasize [
2]. We have also reported that immunotherapy of rat neu expressing breast cancer elicits neu-specific IFN-γ producing CD8+ T cells that in turn facilitate breast cancer recurrence of neu
A ntigen
N egative
V ariant (ANV) tumors following initial rejection of the neu positive
M ouse
M ammary
C arcinoma (MMC) tumor cells in immunocompetent mice [
3,
4]. The tumor antigen loss was due to hypermethylation of the neu promoter and loss of neu both at mRNA and protein levels [
3,
5], resulting in escape of the tumor from further neu-specific immune responses. On the clinical front, elevated serum levels of IFN-γ in uveal melanoma patients correlate with the spread of metastasis and represent a negative prognostic marker [
6]. It was recently shown that HER-2/neu-targeted vaccination of patients who had HER-2/neu+ ductal carcinoma
in situ (DCIS) elicited HER-2/neu-specific IFN-γ producing CD8+ T cell responses which resulted in HER-2/neu antigen loss [
7]. Although the authors considered this HER-2/neu loss a positive outcome of the immune response, no follow-up studies have been performed to determine whether patients with HER-2/neu loss in their DCIS tumors might end up with recurrence of invasive HER-2/neu positive or negative breast cancers. Several other correlative or
in vitro studies suggest potentially negative effects of some immune responses in breast cancer. Matkowski and Sheu both showed in cohorts of 88 and 24 patients with operable breast cancer, respectively, that relapse or disease progression was associated with strong CD8+ T cell infiltration [
8,
9]. Interestingly, it was reported that HER-2/neu+ human prostate tumor cell lines, DU145 and PC-3, that responded to IFN-γ (because of the expression of IFN-γ Rα), showed down-regulation of HER-2/neu expression whereas another prostate tumor cell line, LNCaP, that failed to respond to IFN-γ did not show any change in the expression of HER-2/neu [
10]. Such failure of the LNCaP to respond to IFN-γ was later shown to be due to the lack of JAK1 expression [
11]. These findings prompted us to determine whether HER-2/neu-specific IFN-γ producing T cell responses may be associated with HER-2/neu loss and progression to HER-2/neu negative breast carcinoma. To test this hypothesis, we conducted pilot studies in patients with HER-2/neu positive and HER-2/neu negative breast carcinoma to determine whether patients with HER-2/neu negative tumors had HER-2/neu-specific T cell responses that had been induced by HER-2/neu positive pre-malignant lesions in the past.