Serum thymidine kinase 1 and its kinetics in HER2-positive breast cancer: Results from the Swedish phase II PREDIX HER2 trial.

person Yajing Zhu Karolinska Institute, Stockholm, Sweden info_outline Yajing Zhu, Kang Wang, Ioannis Zerdes, Alexios Matikas, Mattias Bergqvist, Ellinor Elinder, Ana Bosch, Henrik Lindman, Zakaria Einbeigi, Anne Andersson, Lena Carlsson, Ann Charlotte Dreifaldt, Erika Isaksson-Friman, Mats Hellstrom, Hemming Johansson, Jonas C. S. Bergh, Thomas Hatschek, Theodoros Foukakis

Authors person Yajing Zhu Karolinska Institute, Stockholm, Sweden info_outline Yajing Zhu, Kang Wang, Ioannis Zerdes, Alexios Matikas, Mattias Bergqvist, Ellinor Elinder, Ana Bosch, Henrik Lindman, Zakaria Einbeigi, Anne Andersson, Lena Carlsson, Ann Charlotte Dreifaldt, Erika Isaksson-Friman, Mats Hellstrom, Hemming Johansson, Jonas C. S. Bergh, Thomas Hatschek, Theodoros Foukakis Organizations Karolinska Institute, Stockholm, Sweden, Karolinska Institute, Huddinge, Sweden, Karolinska Institute and Karolinska University Hospital, Stockholm, Sweden, Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden, Biovica International, Uppsala, Sweden, Department of Oncology, South Hospital, Stockholm, Stockholm, Sweden, Department of Hematology, Oncology and Radiation Physics, Skåne University Hospital, Lund, Sweden, Department of Oncology, Uppsala University, Uppsala, Sweden, Southern Älvsborg Hospital, Borås, Sweden, Department of Radiation Sciences, Oncology Unit, Umeå University Hospital, Umea, Sweden, Department of Oncology, Sundsvall Hospital, Sundsvall, Sweden, Orebro University Hospital, Orebro, Sweden, Department of Oncology, St Göran Hospital, Stockholm, Sweden, Central Trial Office, Clinical Trial Unit, Karolinska University Hospital, Stockholm, Sweden, Department of Oncology-Pathology, Karolinska Institutet and Breast Cancer Centre, Cancer Theme, Karolinska University Hospital, Karolinska Comprehensive Cancer Center, Stockholm, Sweden, Karolinska University Hospital, Stockholm, Sweden Abstract Disclosures Research Funding Other Background: Thymidine kinase 1 (TK1) plays a pivotal role in DNA synthesis and cellular proliferation. TK1 has been studied as a prognostic marker and as an early indicator of treatment response in HER2-negative early and metastatic breast cancer (BC). Here we present the first report of TK1 in HER2-postitive BC. Methods: In the PREDIX HER2 trial, 202 patients with HER2-positive BC were randomized to 6 cycles of neoadjuvant trastuzumab, pertuzumab and docetaxel or trastuzumab emtansine every three weeks followed by surgery and adjuvant epirubicin and cyclophosphamide. Serum was prospectively collected from all participants at multiple timepoints: At baseline, after cycles 2, 4 and 6, at end of adjuvant therapy and then annually for 5 years and in case of recurrence. TK1 activity was measured by DiviTum assay (Biovica, Sweden), blinded to treatment allocation, patient characteristics and outcome. TK1 activity was correlated with baseline characteristics, pathologic complete response (pCR) and event-free survival (EFS). Results: Baseline TK1 activity as categorical variable was not associated with tumor size or hormone receptor expression. An increase of TK1 activity was seen in the majority of cases after treatment (mean TK1 at baseline = 111.7, after cycle 2 = 1257.0, after cycle 4 = 1157.0, after cycle 6 = 1178.0). Neither baseline TK1, on-treatment TK1 nor its change from baseline to cycle 2 were significantly associated with pCR in multivariable logistic regression analysis (Table). Baseline TK1 activity was not predictive for differential benefit to the study treatments ( p interaction = 0.19). After a median follow-up of 52.5 months, 21 patients had EFS events. There was no significant correlation between baseline TK1 activity and EFS in multivariable Cox regression analysis. Conclusions: Serum TK1 activity in HER2-positive BC increased following treatment with neoadjuvant therapy but was not correlated to pCR rates or EFS. TK1 No. of Patients No. of pCR Cases Logistic regression Odds ratio (95% confidence interval) Unadjusted Model p value Adjusted Model a p value at baseline Tertile1 64 23 1.00(reference) 1.00(reference) Tertile2 65 33 1.84 [0.91, 3.75] 0.09 2.26 [1.07, 4.85] 0.03 Tertile3 64 31 1.67 [0.83, 3.43] 0.15 1.79 [0.85, 3.84] 0.13 p trend b 0.16 0.14 after cycle 2 Tertile1 65 29 1.00(reference) 1.00(reference) Tertile2 64 29 1.03 [0.51, 2.06] 0.94 1.07 [0.51, 2.23] 0.87 Tertile3 65 28 0.94 [0.47, 1.88] 0.86 0.94 [0.41, 2.11] 0.87 p trend b 0.86 0.89 change from baseline to cycle 2 Tertile1 64 26 1.00(reference) 1.00(reference) Tertile2 63 32 1.51 [0.75, 3.06] 0.25 1.58 [0.75, 3.36] 0.23 Tertile3 63 27 1.10 [0.54, 2.23] 0.80 1.15 [0.50, 2.65] 0.75 p trend b 0.80 0.68 a Adjusted for intervention arm, tumor subtype, and tumor size. b Linear trends across tertiles of TK1 were tested by modeling the median value in each quintile as a continuous variable in Logistic regression.