Clinicopathologic determinants of pathologic treatment response in neoadjuvant treated rectal adenocarcinoma

Machine Learning of Histomorphological Features Predict Response to Neoadjuvant Therapy in Locally Advanced Rectal Cancer

AIM

We hypothesize that machine learning of histomorphological features can predict response to neoadjuvant therapy (NAT) in locally advanced rectal cancer (LARC).

METHOD

This retrospective study included 146 LARC patients who received NAT followed by surgery. The pathologists scanned the H&E slides of pretreatment tumor biopsy into whole slide images (WSIs). We randomly split patients into the primary and validation sets with a ratio of 80%:20%. We cut the WSIs into smaller parts (sample amount: 200-500) and used a convolutional neural network (CNN) to process these blocks directly. Then, a graph neural network (GNN) was applied to train the model in the primary set. The independent validation set was used to assess the performance of the model.

RESULT

Our model could provide indicative information to identify the patients who were most likely to benefit from NAT. When the sample amount reached 500, the tile-level classifier for distinguishing poor response from good response produced an AUC of 0.779 in the primary set and 0.733 in the validation set.

CONCLUSION

In this pilot study, we propose a novel predictive model of therapeutic response to NAT in LARC using a routine diagnostic tool employed in daily practice.

Can histologic features predict neoadjuvant therapy response in rectal adenocarcinoma?

Current standard therapy for locally advanced rectal cancer (LARC) is neoadjuvant therapy followed by surgical resection; however, treatment response is variable among patients. This study aimed to identify histologic features that predict tumor response. This retrospective study included 105 patients with LARC, all of whom underwent biopsy followed by neoadjuvant therapy and subsequent surgical resection. Each patient's initial biopsy was evaluated for tumor grade, tumor budding, intraepithelial lymphocytes, intraepithelial neutrophils, desmoplasia, apoptosis, adjacent stromal lymphocytes, signet ring cells, mucinous features, tumoral Paneth cells, dirty necrosis, microscopic ulceration, and prominent lymphoid aggregates. These histologic features, along with patient age at diagnosis and tumor microsatellite status, were compared to tumor regression grades from the respective resection specimens. No histologic factors in tumor biopsies predictive of treatment response in post-therapy resection specimens were identified. Histologic features in pre-therapy biopsy samples of LARC do not predict subsequent response to neoadjuvant therapy. Effective and reliable methods to predict response to neoadjuvant therapy in rectal cancer remain elusive.

A panel of tumor biomarkers to predict complete pathological response to neo-adjuvant treatment in Locally Advanced Rectal Cancer

Pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer patients is related to a favorable prognosis. The identification of early biomarkers predictive of pathological complete response would help optimize the multimodality management of the patients. A panel of 11 tumor-related proteins was investigated by immunohistochemistry in the pretreatment biopsy of a group of locally advanced rectal cancer patients to identify early biomarkers of pathological complete response to neoadjuvant chemoradiotherapy. A mono-institutional retrospective cohort of 95 stage II/III locally advanced rectal cancer patients treated with neoadjuvant chemoradiotherapy and surgery was selected based on clinical–pathological characteristics and the availability of a pretreatment tumor biopsy. Eleven selected protein marker expression (MLH1, GLUT1, Ki67, CA-IX, CXCR4, COX2, CXCL12, HIF1α, VEGF, CD44, and RAD51) was investigated. The optimal cutoff values were calculated by receiver operating characteristic curve analysis. Classification and regression tree analysis was performed to investigate the biomarker interaction. Patients presenting either Ki-67 or HIF1α or RAD51 below the cutoff value, or CXCR4 or COX2 above the cutoff value, were more likely to get a pathological complete response. Classification and regression tree analysis identified three groups of patients resulting from the combination of Ki-67 and CXCR4 expression. Patients with high expression of Ki-67 had the lowest chance to get a pathological complete response (18%), as compared to patients with low expression of both Ki-67 and CXCR4 (29%), and patients with low Ki-67 and high CXCR4 expression (70%). Pretreatment Ki-67, CXCR4, COX2, HIF1α, and RAD51 in tumor biopsies are associated with pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer. A combined evaluation of Ki-67 and CXCR4 would increase their predictive potential. If validated, their optimal cutoff could be used to select patients for a tailored multimodality treatment.

Adenoma-like adenocarcinoma: clinicopathologic characterization of a newly recognized subtype of colorectal carcinoma

The 5th edition of the World Health Organization (WHO) Classification of Tumours (Digestive System) recognizes a new subtype of colorectal adenocarcinoma, called adenoma-like adenocarcinoma. In this study, we sought to determine its clinicopathologic associations and how it is comparable with adenocarcinoma, of no special type (NOS). We retrospectively reviewed all available archival slides of stage I-III colonic adenocarcinoma resection specimens at our institution from 2013 to 2016.Ninety-one cases were classified as adenoma-like adenocarcinoma, and 251 cases were classified as adenocarcinoma, NOS. Of the adenoma-like adenocarcinoma cases, a majority (65 cases, 71%) were composed exclusively of adenoma-like features, designated as pure adenoma-like adenocarcinoma, whereas in the rest, the component of adenoma-like morphology was more than 50% but less than 100%, designated as mixed adenoma-like adenocarcinoma. Compared with adenocarcinoma, NOS, adenoma-like adenocarcinoma cases were significantly associated with the absence of tumor budding (P < 0.001), the absence of an immature/myxoid desmoplastic reaction (P < 0.001), the presence of intraepithelial tumor-infiltrating lymphocytes (P = 0.006), involvement of fewer lymph nodes ( P < 0.001), fewer tumor deposits (P = 0.042), lower pT stage (P = 0.047), lower pN stage (P < 0.001), and consequently the pTNM prognostic group (P < 0.001), as well as better recurrence-free survival (RFS), as per univariate analysis than adenocarcinoma, NOS cases (P = 0.026) but not as per multivariate analysis. However, mixed adenoma-like adenocarcinoma had a worse RFS than pure adenoma-like adenocarcinoma (hazard ratio = 1.639, 95% confidence interval = 0.494-5.437). Our findings not only support the importance of distinguishing this new subtype of colorectal adenocarcinoma but also raise the question whether mixed adenoma-like adenocarcinoma cases should be included in this category, and if so, whether 50% is an appropriate cutoff, as currently defined by the WHO.

Intraepithelial tumour infiltrating lymphocytes are associated with absence of tumour budding and immature/myxoid desmoplastic reaction, and with better recurrence-free survival in stages I-III colorectal cancer

AIMS

Tumour budding (TB), desmoplastic reaction (DR) and intraepithelial tumour infiltrating lymphocytes (iTILs) are recently recognised prognostic factors in colorectal cancer (CRC). In this study, we evaluated their significance and relationship to each other and their cumulative effect on survival.

METHODS AND RESULTS

A total of 372 stages I-III CRC cases from 2013 to 2016 were included. Low TB was identified in 302 (81%) cases, immature/myxoid DR in 67 (18%) cases and iTILs in 130 (35.0%) cases. iTILs was associated with low budding (P = 0.0247), non-myxoid DR (P = 0.0004), poorly differentiated histology (P = 0.0015), absence of perineural invasion (P = 0.0367) and loss of mismatch repair proteins (P = 0.0002). Absence of iTILs and presence of immature/myxoid DR were associated with a worse recurrence-free survival (RFS) [hazard ratio (HR) = 2.191, 95% confidence interval (CI) = 1.232-3.895; and HR = 5.706, 95% CI = 3.632-8.964, respectively]. A competing risk analysis showed statistically significant prognostic groups combining iTILs and TB (P < 0.0001). Cases with iTILs and low TB were associated with better RFS compared to cases without iTILs and with intermediate/high TB (HR = 0.214, 95% CI = 0.109-0.421). Similarly, combining iTILs and DR revealed statistically significant prognostic groups (P < 0.0001). Cases with iTILs and a non-myxoid DR had better RFS compared to cases without iTILs and immature/myxoid DR (HR = 0.113, 95% CI = 0.056-0.230). On multivariate cause-specific analysis, patients' age (P = 0.0045), iTILs (P = 0.0345), DR (P < 0.0001) and pTNM prognostic groups (P < 0.0001) were associated with RFS.

CONCLUSIONS

Our study validates the association of iTILs and DR as independent prognostic finding in CRC, and propose a prognostic model using the combinations of iTILs with TB and stromal reaction in CRC.

Effect of aspirin use on neoadjuvant chemoradiotherapy for rectal cancer: a meta-analysis with trial sequential analysis

BACKGROUND

Numerous studies have reported the preventive and protective effects of aspirin in patients with rectal cancer. However, it is not clear whether aspirin can be used as an assistance drug in preoperative neoadjuvant chemoradiotherapy. Therefore, this study will explore the efficacy of aspirin as an adjuvant agent in rectal cancer neoadjuvant chemoradiotherapy.

METHODS

A literature search was performed using the electronic platforms to obtain relevant research studies published up to Jan 2020. The search was limited to papers published in English or Chinese language. Confidence intervals of research endpoints in each study were extracted and merged. The meta-analysis was performed using Stata12.0 software. Furthermore, we performed trial sequential analysis (TSA) to evaluate the robustness of our findings and to obtain a more conservative estimation.

RESULTS

A total of 5 studies including 977 patients were identified to be eligible for this meta-analysis. Compared with control group, aspirin group significantly increased pathologic complete response rate from 16.5 to 22.3% (RR 1.41, 95% CI 1.01-1.96, P = 0.041), partial remission rate from 21.8 to 45.7% (RR 1.87, 95% CI 1.37-2.54, P < 0.001), and tumor down-staging rate from 44.4 to 63.8% (RR 1.43, 95% CI 1.17-1.75, P = 0.001). Moreover, aspirin group can reduce local recurrence rate (RR 0.37, 95% CI 0.17-0.84, P = 0.017), improve 3-year survival rate (RR 1.24, 95% CI 1.12-1.36, P < 0.001), and 5-year survival rate (RR 1.29, 95% CI 1.14-1.46, P < 0.001). TSA shows that the meta-analysis results of pathologic complete response rate and local recurrence rate may be a false positive. Furthermore, the meta-analysis results of other study endpoints were further confirmed by TSA.

CONCLUSION

Aspirin, as an adjuvant agent, can enhance the effect of neoadjuvant chemoradiotherapy and improve the prognosis of patients with rectal cancer. Neoadjuvant therapy combined with aspirin may be considered a better option for preoperative rectal cancer patients.