Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jan;66(1):51-61.
doi: 10.1007/s00262-016-1919-1. Epub 2016 Oct 25.

High-efficiency Lysis of Cervical Cancer by Allogeneic NK Cells Derived From Umbilical Cord Progenitors Is Independent of HLA Status

Affiliations
Free PMC article

High-efficiency Lysis of Cervical Cancer by Allogeneic NK Cells Derived From Umbilical Cord Progenitors Is Independent of HLA Status

John P Veluchamy et al. Cancer Immunol Immunother. .
Free PMC article

Abstract

Down-regulation of HLA in tumor cells, low numbers and dysfunctionality of NK cells are commonly observed in patients with end-stage cervical cancer. Adoptive transfer of high numbers of cytotoxic NK cells might be a promising treatment approach in this setting. Here, we explored the cytotoxic efficacy on ten cervical cancer cell lines of activated allogeneic NK cells from two sources, i.e., peripheral blood (PBNK) with and without cetuximab (CET), a tumor-specific monoclonal antibody directed against EGFR, or derived from umbilical cord blood (UCB-NK). Whereas CET monotherapy was ineffective against the panel of cervical cancer cell lines, irrespective of their EGFR expression levels and despite their RAS wt status, it significantly enhanced the in vitro cytotoxic efficacy of activated PBNK (P = 0.002). Equally superior cytotoxicity over activated PBNK alone was achieved by UCB-NK (P < 0.001). Both PBNK- and UCB-NK-mediated cytotoxic activity was dependent on the NK-activating receptors natural killer group 2, member D receptor (NKG2D) and DNAX accessory molecule-1 (DNAM-1) (P < 0.05) and unrelated to expression levels of the inhibitory receptors HLA-E and/or HLA-G. Most strikingly, whereas the PBNK's cytotoxic activity was inversely correlated with HLA-ABC levels (P = 0.036), PBNK + CET and UCB-NK cytotoxicity were entirely independent of HLA-ABC expression. In conclusion, this study provides a rationale to initiate a clinical trial for cervical cancer with adoptively transferred allogeneic NK cells, employing either UCB-NK or PBNK + CET for EGFR-expressing tumors. Adoptive transfer of UCB-NK might serve as a generally applicable treatment for cervical cancer, enabled by HLA-, histology- and HPV-independent killing mechanisms.

Keywords: Adoptive NK immunotherapy; CET; Cervical cancer; NK ligands and receptors; Peripheral blood NK cells; Umbilical cord blood stem cell-derived NK cells.

Conflict of interest statement

J. P. Veluchamy and J. Spanholtz are the employees of Glycostem Therapeutics; D. Heideman serves on scientific advisory boards of Amgen and Pfizer. The other authors declare no conflict of interest.

Figures

Fig. 1
Fig. 1
PBNK cytotoxicity against cervical cancer cells alone and in combination with CET. a Cytotoxicity levels (Δ7AAD) of activated PBNK (open bars) and PBNK + cetuximab (CET) (closed bars) against ten cervical cancer cell lines, b arranged in order of EGFR expression level. Bars are means of triplicate values from four experiments with four different PBNK donors for C33A, HeLa, SiHa, CC11B, CC11A, CC10B, CC10A, CaSki and two experiments with two different PBNK donors for CSCC7 and CC8. Bars represent mean ± SEM. c Significantly higher cytotoxicity levels (Δ7AAD) were observed in all cell lines after co-culture with PBNK + CET compared to PBNK, except for C33A (open circle). *P < 0.05 and **P < 0.01 calculated with paired t test
Fig. 2
Fig. 2
PBNK and UCB-NK cytotoxicity against cervical cancer cells. a Cytotoxicity levels (Δ7AAD) of PBNK (open bars) and UCB-NK (hatched bars) against ten cervical cancer cell lines. Bars are means of triplicate values from four experiments with four different PBNK donors for C33A, HeLa, SiHa, CC11B, CC11A, CC10B, CC10A, CaSki and two experiments with two different PBNK donors for CSCC7 and CC8 and five experiments for UCB-NK using five different UCB-NK donors for all cell lines; Bars represent mean ± SEM. PBNK data used to compare with UCB-NK in a are the same dataset as Fig. 1a. b Significantly higher cytotoxicity levels (Δ7AAD) were observed in all cell lines after co-culture with UCB-NK compared to PBNK. c Significantly higher levels of NK degranulation (ΔCD107a) in PBNK + cetuximab (CET) and UCB-NK conditions compared to PBNK only condition. Triangles denote cell lines with low EGFR levels, i.e., C33A, HeLa and SiHa. *P < 0.05, **P < 0.01 and ***P < 0.001 calculated in A and B with unpaired t test, in C with one-way ANOVA, Bonferroni’s multiple comparison test
Fig. 3
Fig. 3
NK-activating receptors in PBNK and UCB-NK and their ligand expression in cervical cancer cell lines and their influence on NK cytotoxicity. a Percentage of positive cells within the NK cell population for NK-activating receptors DNAM-1 and NKG2D for PBNK only, PBNK stimulated with cytokines (IL-2 + IL-15) and UCB-NK only were determined by flow cytometry. The data presented is from three representative donors for PBNK and UCB-NK. PBNK only are denoted by open circles, PBNK (IL-2 + IL-15) are denoted by closed circles and UCB-NK only by closed squares. b Representative example of histograms showing geometric mean fluorescence intensity (MFI) for NK-activating ligands PVR (ligand of DNAM-1 receptor), MICA/B and ULBP1, −3 and −2/5/6 (ligands of NKG2D receptor). c PBNK and UCB-NK were coated with NKG2D and/or DNAM-1 blocking antibodies and incubated with C33A and SiHa cells. Cytotoxicity levels (Δ7AAD) were measured from 7AAD + C33A and SiHa cells at the end of a 4 h assay. Data presented are means of triplicate values from three different PBNK and three different UCB-NK donors; Bars represent mean ± SEM. *P < 0.05 and **P < 0.01 calculated with paired, two-way ANOVA multiple comparisons of column means
Fig. 4
Fig. 4
Effects of NK inhibitory ligands on NK cytotoxicity against cervical cancer cells. a Percentage of positive cells within the NK cell population for NK inhibitory receptors KIR2D and CD94/NKG2A for PBNK only, PBNK stimulated with cytokines (IL-2 + IL-15) and UCB-NK only were determined by flow cytometry. The data presented is from three representative donors for PBNK and UCB-NK. PBNK only are denoted by open circles, PBNK (IL-2 + IL-15) are denoted by closed circles and UCB-NK by closed squares. b Representative histogram plots showing geometric mean fluorescence intensity (MFI) of NK inhibitory ligands HLA-ABC, HLA-E and HLA-G on cervical cancer cells; representative plots of 2–3 separate analyses are shown. Correlation analysis of MFI of HLA-ABC with % cytotoxicity (Δ7AAD) by c PBNK, d PBNK + cetuximab(CET) and e UCB-NK. Dotted lines represent 95 % confidence interval of the regression line. Four experiments with four different PBNK donors for C33A, HeLa, SiHa, CC11B, CC11A, CC10B, CC10A, CaSki, two experiments with two different PBNK donors for CSCC7 and CC8 and five experiments with five different UCB-NK donors were used for this experiment. P value was calculated with Pearson’s analysis

Similar articles

See all similar articles

Cited by 8 articles

See all "Cited by" articles

References

    1. Trottier H, Franco EL. The epidemiology of genital human papillomavirus infection. Vaccine. 2006;24(Suppl 1):S1–15. - PubMed
    1. Woodman CB, Collins SI, Young LS. The natural history of cervical HPV infection: unresolved issues. Nat Rev Cancer. 2007;7:11–22. doi: 10.1038/nrc2050. - DOI - PubMed
    1. Steenbergen RD, Snijders PJ, Heideman DA, Meijer CJ. Clinical implications of (epi)genetic changes in HPV-induced cervical precancerous lesions. Nat Rev Cancer. 2014;14:395–405. doi: 10.1038/nrc3728. - DOI - PubMed
    1. Zagouri F, Sergentanis TN, Chrysikos D, Filipits M, Bartsch R. Molecularly targeted therapies in cervical cancer: a systematic review. Gynecol Oncol. 2012;126:291–303. doi: 10.1016/j.ygyno.2012.04.007. - DOI - PubMed
    1. Diaz-Padilla I, Monk BJ, Mackay HJ, Oaknin A. Treatment of metastatic cervical cancer: future directions involving targeted agents. Crit Rev Oncol Hematol. 2013;85:303–314. doi: 10.1016/j.critrevonc.2012.07.006. - DOI - PubMed

Publication types

Feedback