Background
High-grade serous ovarian cancer has a recurrence and mortality rate of approximately 80%. The impact of disease recurrence on the immune system remains unclear. Understanding this could potentially enhance our future therapeutic strategies.
Materials and Methods
We monitored 27 recurrent ovarian cancer patients treated with second-line chemotherapy (carboplatin-gemcitabine) and/or PARP(Poly(ADP-ribose)polymerase)-inhibition (PARPi) (OV-IMM-2014, S56311, University Hospitals Leuven). PARPi was administered as therapy (CLIO/BGOG-OV10) or in maintenance setting. Serum samples were taken at diagnosis of recurrence and after three or six cycles of treatment. Via Luminex technology, 17 immune-related proteins were assessed in nine sample pairs in the carboplatin-gemcitabine group and in 13 pairs in the PARPi group. The protein concentrations were compared during the treatment course and correlated with clinical parameters.
Results
After six cycles of carboplatin-gemcitabine, CXCL-10 and CCL-5 increased, while CCL-2 and VEGF-A decreased in responding patients. In comparison, PARPi increased CXCL-10 and decreased CCL-2, VEGF-A, MMP-7 and CCL-5. Surprisingly, there was no difference in protein concentration or trends between patients who underwent secondary debulking surgery (SDS) or not, prior to carboplatin-gemcitabine. On the other hand, patients who underwent SDS prior to PARPi exhibited different trends in protein concentrations (galectin-3, MMP-7, TGF-β and CCL-5), compared to patients who did not undergo surgery. There was no difference in protein levels nor trends in patients who received PARPi in platinum-resistant versus platinum-sensitive setting. Comparing patients with a BMI below 25 and patients with a BMI of 25 of higher at initiation of both carboplatin-gemcitabine and PARPi, there was less variation in the levels of proteins in patients with a BMI below 25.
Conclusions
This study is among the first to highlight the immune changes during recurrence. Although populations were small, differences in the tumour immune micro-environment between patient groups were noticeable. Notably, tumour burden influenced the protein levels during PARPi treatment and chemotherapy response decreased immunosuppression. These findings could pave the way for future optimal immunotherapy planning.
J. Kempeneers: None. J. Ceusters: None. G. Thirion: None. S. Claes: None. D. Schols: None. T. Baert: None. A. Coosemans: None.