The duration of the neoadjuvant stage of treatment in study group I (total neoadjuvant therapy: courses of neoadjuvant radiotherapy followed by long-term RT) averaged 198,31±8,3 days, in group II (chemoradiotherapy – long-term RT to a TFD of 37.5-40 Gy with capecitabine potentiation) – 99,37±5,42 days, and in group III (intensive radiotherapy – short course of preoperative radiotherapy to a TFD of 20-25 Gy) – 5,31±0,32 days. There were no complications at the first stage of treatment in both groups that would lead to a delay or cancellation of therapy. In all patients of groups I and II, according to the RECIST 1.1 criteria, during neoadjuvant therapy, the overall response was stabilisation or partial response. A statistically significant difference in the time of surgical intervention between the control group (152.7 minutes) and patients with a prolonged neoadjuvant treatment phase of groups I and II was determined – 181.5 and 186.0 minutes, respectively. In the study patients, the 3-year overall survival rate was 90 %, and the 5-year overall survival rate was 81.5 %, respectively. The 3-year overall survival rate in study group I, who received total neoadjuvant therapy, was 75.9±10.0 %; in group II (preoperative chemoradiotherapy) – 88.6±5.7 %; in group III (preoperative intensive radiotherapy) – 81.3±8.5 %; and in the control group, who received surgical treatment at the first stage – 71.5±12.0 %. In study group I, recurrence was detected in 20 % of patients, the average time to recurrence was 25.5 months from the start of treatment, and in group II – 13.3 % (time – 35 months), and in group III – 23.8 % (time – 24.6 months), respectively. It was found that the most common surgical complication for the total number of patients was anastomotic failure. Failure of category A was treated with a conservative method, which was effective in 3 cases (75 %), in the first case after conservative treatment, endoVAC was used. In all cases, the stoma was closed. Category B failure was treated conservatively in 50 % of cases, and with the use of VAC systems – in 50%. Only one patient managed to close the stoma, and in 75% the stoma was not closed due to the prolongation of the disease. A growth in the NLR level is accompanied by an increase in the number of postoperative complications and a decrease in both recurrence-free and overall survival. The 3-year overall and recurrence-free survival rates at the minimum NLR were 96.9 % and 64 %, respectively, and the similar indicators of the predicted 5-year survival rate were 93.9 % and 48 %, at the maximum NLR, namely 3-year overall and recurrence-free survival rates were 75 % and 32 %, respectively, and similar rates of predicted 5-year survival were 58 % and 20 % (p<0.05). Neural networks were used and informative criteria were selected for choosing the optimal variant of the neoadjuvant stage of treatment of patients with stage II-III rectal cancer, which was confirmed by their high accuracy – 80 % - 95 %. Scientific data have been supplemented to show that the growth of the level of neutrophil-lymphocyte index (NLR) is accompanied by an increase in the number of postoperative complications: the average NLR values in patients without complications were 4.5, which is 1.3 times less than in the group of patients with rectal cancer who had complications (p<0.05). The scientific data on the survival results of patients with stage II-III rectal cancer at different NLR levels have been supplemented. The 3-year overall survival rate at the lowest NLR values was 96.9 %, which is 21.9 % higher compared to Cox regressions for the highest NLR values (p<0.05). Scientific data show that the level of NLR correlates with BMI – with an increase in the level of one of the indicators, the other one grows in direct proportion, as well as with the diameter of the stapler used during anastomosis, blood loss during surgery and albumin level before treatment, and is inversely proportional to the duration of preoperative therapy. The possibility of choosing the optimal method of neoadjuvant therapy in the combined treatment of patients with rectal cancer of stage II-III using an artificial neural network model with regard to individual informative criteria was developed and implemented. The risk factors for postoperative complications in patients with rectal cancer were developed and implemented, namely the correlation of NLR with body mass index and the diameter of the stapler used during anastomosis. Key words: rectal cancer, radiation therapy, chemotherapy, chemoradiotherapy, surgical treatment, neutrophil-lymphocyte index, surgical complications, complications of radiation and chemotherapy, treatment results, recurrence, survival, prognosis, risk factors, neural networks, cancer. Branch-Medicsne.
