接受新辅助放化疗的直肠癌患者病理完全缓解的发生率及预测因素

Treatment outcomes for rectal cancer have improved significantly over the last few decades. Concerns regarding the morbidity of treatment have led to nuances like total neoadjuvant therapy and nonoperative management. The primary objective was to determine the predictive factors associated with pathological complete response in locally advanced carcinoma rectum undergoing neoadjuvant chemoradiation. The secondary objectives were to estimate the incidence of pCR and cCR in the study population, and to determine the concordance rate of cCR with pCR.

We included 84 patients diagnosed with nonmetastatic rectal cancer who underwent neoadjuvant long-course chemoradiation (CTRT) and radical surgery during September 2018 to March 2020 at the institute. The clinicopathological and treatment details were collected. Rates of clinical (cCR) and pathological complete response (pCR) were analyzed and compared with baseline and histopathological characteristics. Of 84 patients who underwent radical surgery following CTRT, 23 (27. 4%) and 16 (19%) had cCR and pCR, respectively. Mean duration for completion of CTRT and interval to surgery were 35. 33 days and 57. 86 days, respectively. The majority were stage 3 (96. 4%), low-middle rectal (95. 4%), and moderately differentiated (69%) cancers.

Even though cCR was a significant predictor of pCR ( p = . 01), the rate of non pCR in the cCR group was 43. 5% while 4. 9% of the non cCR had pCR in the postoperative pathological specimen. Pre-treatment CEA level and neutrophil: lymphocyte ratio, differentiation of tumor, tumor infiltrating lymphocyte response were found to have a statistical significant association with pCR.

The rate of cCR and pCR was 27. 4% and 19% respectively in our population. The tumor location, T4 and N2 stages, in contrary to the existing literature, do not predict a pCR. With the low rate of pCR and significant discordance between cCR and pCR, stringent selection strategies are mandatory before contemplating a nonoperative management.