Complete regression of metastatic cervical cancer after treatment with human papillomavirus-targeted tumor-infiltrating T cells

Abstract

Purpose: Metastatic cervical cancer is a prototypical chemotherapy-refractory epithelial malignancy for which better treatments are needed. Adoptive T-cell therapy (ACT) is emerging as a promising cancer treatment, but its study in epithelial malignancies has been limited. This study was conducted to determine if ACT could mediate regression of metastatic cervical cancer.

Patients and methods: Patients enrolled onto this protocol were diagnosed with metastatic cervical cancer and had previously received platinum-based chemotherapy or chemoradiotherapy. Patients were treated with a single infusion of tumor-infiltrating T cells selected when possible for human papillomavirus (HPV) E6 and E7 reactivity (HPV-TILs). Cell infusion was preceded by lymphocyte-depleting chemotherapy and was followed by administration of aldesleukin.

Results: Three of nine patients experienced objective tumor responses (two complete responses and one partial response). The two complete responses were ongoing 22 and 15 months after treatment, respectively. One partial response was 3 months in duration. The HPV reactivity of T cells in the infusion product (as measured by interferon gamma production, enzyme-linked immunospot, and CD137 upregulation assays) correlated positively with clinical response (P = .0238 for all three assays). In addition, the frequency of HPV-reactive T cells in peripheral blood 1 month after treatment was positively associated with clinical response (P = .0238).

Conclusion: Durable, complete regression of metastatic cervical cancer can occur after a single infusion of HPV-TILs. Exploratory studies suggest a correlation between HPV reactivity of the infusion product and clinical response. Continued investigation of this therapy is warranted.

Trial registration: ClinicalTrials.gov NCT01585428.

Conflict of interest statement

Authors' disclosures of potential conflicts of interest are found in the article online at www.jco.org. Author contributions are found at the end of this article.

Published by the American Society of Clinical Oncology.

Figures

Fig 1.

Complete tumor responses in two patients with metastatic cervical cancer treated with tumor-infiltrating T cells selected for human papillomavirus E6 and E7 reactivity (HPV-TILs). Change from pretreatment baseline in longest diameter of individual metastatic tumors after treatment with HPV-TILs for patients (A) 3 and (B) 6. Contrast-enhanced computed tomography scans obtained before treatment and at most recent follow-up for these patients. (C) Patient 3 had disease involving para-aortic, bilateral hilar, subcarinal, and left iliac lymph nodes (gold arrows). Left hilar tumor seen on second and third images from top is same tumor at different slice levels. Patient had no evidence of disease 22 months after treatment. (D) Patient 6 had metastatic disease in para-aortic lymph node, abdominal wall, aortocaval lymph node, left pericolic pelvic mass, and right ureteral nodule (gold arrows). Additional tumors were present on liver surface and were visualized best with magnetic resonance imaging (Data Supplement). Patient had no evidence of disease 15 months after treatment. Red arrowhead indicates ureteral stent that was removed after right ureteral tumor regressed.

Fig 2.

Human papillomavirus (HPV) reactivity of infused T cells. Infusion products for each patient were assessed for reactivity against HPV type–specific E6 and E7 oncoproteins using (A) interferon gamma (IFN-γ) enzyme-linked immunospot, (B) CD137 upregulation, and (C) IFN-γ production assays. For patients 3 and 4, HPV-16–positive oncoprotein peptide pools were used; for patients 1, 2, 5, 6, 7, 8 and 9, HPV-18–positive oncoprotein peptide pools were used (Table 1). Values shown represent sum of E6 and E7 reactivity after background subtraction (data for each antigen and negative control are provided in Data Supplement). Data are representative of ≥ two independent experiments, each performed in duplicate wells. NR, nonresponding patient; R, responding patient.

Fig 3.

Repopulation of patients with human papillomavirus (HPV) –reactive T cells after treatment with tumor-infiltrating T cells selected for HPV E6 and E7 reactivity. Peripheral blood (PB) samples from before and after treatment were assessed by interferon gamma (IFN-γ) enzyme-linked immunospot for reactivity against HPV oncoproteins. For patients 3 and 4, HPV-16–positive oncoprotein peptide pools were used; for patients 1, 2, 5, 6, 7, 8 and 9, HPV-18–positive oncoprotein peptide pools were used (Table 1). (A) HPV reactivity of PB T cells from responding (blue) and nonresponding (gold) patients before and 19 to 42 days after treatment. (B, C, D) HPV reactivity of PB T cells from responding patients (B) 2, (C) 3, and (D) 6 at late time points after treatment. Error bars represent standard deviations of duplicate wells in same experiment except in patients 3 (month 4) and 6 (months 1, 3, and 11), which are single determinations. This experiment was repeated once with slightly different time points and similar results.