Comparison and quantification of different concurrent chemotherapy regimens with radiotherapy in locally advanced non-small cell lung cancer: Clinical outcomes and theoretical results from an extended LQ and TCP model

To develop a new radiobiological model and compare the efficacy of four concurrent chemotherapy regimens administered with radiotherapy in locally advanced non-small-cell lung cancer (LANSCLC) by in-field locoregional progression-free survival (LPFS).151 LANSCLC patients were reviewed and divided into 5 groups according to their concurrent chemotherapy regimens, including 24 patients treated with radiotherapy alone, 30 treated with concurrent 4-week etoposide-cisplatin (EP), 26 with 3-week pemetrexed-cisplatin (AP), 37 with weekly paclitaxel-cisplatin (TP) and 34 with weekly docetaxel-cisplatin (DP). In-field LPFS and toxicities were compared among groups. A novel tumor control probability (TCP) model, LQRGC, incorporating four "R"s of radiobiology, Gompertzian tumor growth and chemotherapeutic effect, was related to in-field LPFS. Chemo-induced biologically effective doses (BEDs) in LQRGC/TCP model were used to quantify the concurrent chemotherapeutic efficacy.The median follow-up time was 54.5 months. The weekly DP and 4-week EP groups had favorable median in-field LPFS (EP:46.2 months, AP:30.3 months, TP:12.2 months, DP: not reached, radiotherapy alone: 12.2 months, p = 0.001). The 4-week EP group had a higher incidence of ≥grade 3 leukopenia (EP:76.7%, AP:15.4%, TP:24.3%, DP:14.7%, radiotherapy alone: 12.5%, p < 0.001) than the other four. The LQRGC/TCP model fitted well with the in-field LPFS with the average absolute and relative fitting errors of 6.36% and 12.12%. The chemo-induced BEDs of EP, AP, TP and DP were 5.17, 0.63, 1.89 and 2.52 Gy, respectively.The LQRGC/TCP model achieved promising fitting accuracy for in-field LPFS. As quantified by the model, the 4-week EP and weekly DP showed higher chemo-induced BEDs when concurrently administered with radiotherapy in LANSCLC.