scholarly journals Pathologic Complete Response Prediction after Neoadjuvant Chemoradiation Therapy for Rectal Cancer Using Radiomics and Deep Embedding Network of MRI

2021 ◽  
Vol 11 (20) ◽  
pp. 9494
Author(s):  
Seunghyun Lee ◽  
Joonseok Lim ◽  
Jaeseung Shin ◽  
Sungwon Kim ◽  
Heasoo Hwang
Keyword(s):  
Rectal Cancer ◽  
Dimension Reduction ◽  
Pathologic Complete Response ◽  
Prediction Method ◽  
Complete Response ◽  
Neoadjuvant Chemoradiation ◽  
Chemoradiation Therapy ◽  
Neoadjuvant Chemoradiation Therapy ◽  
Deep Embedding ◽  
Mri Sequences

Assessment of magnetic resonance imaging (MRI) after neoadjuvant chemoradiation therapy (nCRT) is essential in rectal cancer staging and treatment planning. However, when predicting the pathologic complete response (pCR) after nCRT for rectal cancer, existing works either rely on simple quantitative evaluation based on radiomics features or partially analyze multi-parametric MRI. We propose an effective pCR prediction method based on novel multi-parametric MRI embedding. We first seek to extract volumetric features of tumors that can be found only by analyzing multiple MRI sequences jointly. Specifically, we encapsulate multiple MRI sequences into multi-sequence fusion images (MSFI) and generate MSFI embedding. We merge radiomics features, which capture important characteristics of tumors, with MSFI embedding to generate multi-parametric MRI embedding and then use it to predict pCR using a random forest classifier. Our extensive experiments demonstrate that using all given MRI sequences is the most effective regardless of the dimension reduction method. The proposed method outperformed any variants with different combinations of feature vectors and dimension reduction methods or different classification models. Comparative experiments demonstrate that it outperformed four competing baselines in terms of the AUC and F1-score. We use MRI sequences from 912 patients with rectal cancer, a much larger sample than in any existing work.

Assessment of macroscopic features in relation to tumor response to neoadjuvant chemoradiation therapy in rectal cancer.

2013 ◽  
Vol 31 (4_suppl) ◽  
pp. 507-507
Author(s):  
Amanda Kathleen Arrington ◽  
Julio Garcia-Aguilar ◽  
Marjun Philip Duldulao ◽  
Carrie Luu ◽  
Julian Sanchez ◽  
...  
Keyword(s):  
Rectal Cancer ◽  
Tumor Response ◽  
Locally Advanced ◽  
Pathologic Complete Response ◽  
Complete Response ◽  
Neoadjuvant Chemoradiation ◽  
Chemoradiation Therapy ◽  
Oncologic Outcomes ◽  
Neoadjuvant Chemoradiation Therapy ◽  
Pathologic Features

507 Background: Several studies show locally advanced rectal cancer patients with clinical complete response (cCR) have comparable oncologic outcomes to pathologic complete response (pCR) to neoadjuvant chemoradiation therapy (NCRT). Thus, total mesorectal excision (TME) could potentially be avoided. Our objective was to validate macroscopic features of cCR. Methods: 164 patients with stage II/III rectal cancer were previously enrolled in a phase II trial and treated by NCRT and TME. Tumor response in the surgical specimens was assessed according to AJCC guidelines. A pCR was defined as absence of viable tumor cells. Gross macroscopic features by the pathologist were tabulated and our cohort was applied to previously published cCR criteria. Results: 25.0% (n = 41) had pCR; 75.0% (n = 123) had non-pCR. Descriptors were condensed into 14 macroscopic features by combining terms and excluding those rarely mentioned. Several reports affirm that scarring signifies cCR, while others suggest that fibrosis, edema, ulceration and nonpalpable tumor to be consistent with cCR. In our study, scarring was found in 6.1% of patients, 16.7% of which had pCR. We found that hyperemia, scarring, flat, smooth, and tan-pink mucosa were significantly associated with pCR (p < 0.05). In contrast, a firm lesion and ulceration were frequently observed in patients with non-pCR (p = 0.02 and 0.05 respectively). Conclusions: Our study suggests that macroscopic pathologic features do correlate with pCR. Although cCR has comparable oncologic outcomes as pCR with favorable outcomes, standard criteria of cCR should first be defined so NCRT patients can safely be selected for observation only. [Table: see text]

Totally neoadjuvant chemoradiation therapy with mFOLFOX6 in locally advanced rectal cancer: A single arm phase II study (FOTAC).

2017 ◽  
Vol 35 (4_suppl) ◽  
pp. TPS816-TPS816
Author(s):  
Jianwei Zhang ◽  
Yue Cai ◽  
Huabin Hu ◽  
Jian Xiao ◽  
Dianke Chen ◽  
...  
Keyword(s):  
Rectal Cancer ◽  
Phase Ii Study ◽  
Locally Advanced ◽  
Pathologic Complete Response ◽  
Complete Response ◽  
Advanced Rectal Cancer ◽  
Locally Advanced Rectal Cancer ◽  
Neoadjuvant Chemoradiation ◽  
Chemoradiation Therapy ◽  
Neoadjuvant Chemoradiation Therapy

TPS816 Background: Preoperative 5-Fluorouracil based chemoradiotherapy is the standard of treatment for locally advanced rectal cancer. About 15% to 18% of patients would achieve pathologic complete response (pCR) after 5-Fluorouracil based chemoradiation. And the survival outcome of patients with pCR was much better than that of non-pCR. In our previous FOWARC study, in the group of preoperative systemic chemotherapy with mFOLFOX6 combined with radiation, the pCR rate was up to 27.5%. In another study, adding mFOLFOX6 after neoadjuvant chemo radiation in locally advanced rectal cancer improve the pCR rate to 38%. This phase II study aimed to explore whether totally neoadjuvant chemoradiation therapy with mFOLFOX6 could further improve the pCR rate in locally advanced rectal cancer. Methods: The primary endpoint is the pathologic complete response rate (pCR).The secondary endpoint included 3-year disease free survival rate, 3-year local recurrence rate, and safety. We hypothesized that totally neoadjuvant chemoradiation therapy with mFOLFOX6 could improve the pCR rate from 18% to 45% with 5% type I error and 80% power. Fifty patients met inclusion criteria will be enrolled in the trial. All patients will receive long term radiation for 25 times and 50Gy before surgery. Four cycles of mFOLFOX6 would be performed every 2 weeks during radiotherapy, and another 4-6 cycles would be added after radiotherapy and before operation. Totally, the patients will receive 8-10 cycles of chemotherapy before surgery. MRI of the pelvic will be performed every 4 cycles of the therapy to assess clinical response. Then the patient will receive total mesorectal excision at least 8 weeks after radiotherapy. The post-operative chemotherapy will be omitted and all the patients go to surveillance. Clinical trial information: NCT02887313.

Use of KRAS codon 13 mutations to predict response to neoadjuvant chemoradiation therapy in patients with rectal adenocarcinoma.

2012 ◽  
Vol 30 (4_suppl) ◽  
pp. 453-453
Author(s):  
Marjun Philip Duldulao ◽  
Wendy Lee ◽  
Zhenbin Chen ◽  
Wenyan Li ◽  
Rebecca A. Nelson ◽  
...  
Keyword(s):  
Rectal Cancer ◽  
Kras Mutation ◽  
Pathologic Complete Response ◽  
Neoadjuvant Chemoradiation ◽  
Chemoradiation Therapy ◽  
Wild Type ◽  
Kras Mutations ◽  
Neoadjuvant Chemoradiation Therapy ◽  
Kras Codon ◽  
Codon 61

453 Background: The predictive role of KRAS mutation in therapeutic management of metastatic colorectal cancer has been much investigated. However, the association between KRAS mutations and rectal cancer response to neoadjuvant chemoradiation therapy (CRT) remains unclear. Our objectives were to determine the incidence of KRAS mutations in patients with rectal cancer and identify potential correlations with response to CRT. Methods: Patients with Stage II-III rectal cancer (n=148) from a multicenter prospective clinical trial were assessed. DNA was extracted from pretreatment tumor biopsies; and genotyping of KRAS was completed by PCR and direct sequencing. All patients underwent pre-operative CRT followed by surgery; and treatment response was evaluated by a pathologist. Surgical specimens without evidence of residual disease were deemed to have a pathologic complete response (pCR). Results: Overall, 60 of 148 (40.5%) patients had KRAS mutation. Breakdown of the different KRAS mutations revealed that 40 (66.7%) mutations occurred in codon 12, 12 (20.0%) in codon 13, 5 (8.3%) in codon 61, and 3 (5.0%) in other locations. When examining the relationship between KRAS and CRT response, we observed that detection of any KRAS mutation was associated with a decreased rate of pCR compared to wild-type KRAS (13.3% vs. 33.0%, respectively; p=0.007). More specifically, patients with KRAS codon 13 mutations did not respond to CRT compared to non-codon 13 mutations (p=0.037). In contrast, the detection of codon 12 or codon 61 mutations was not associated with response to CRT ( Table ). Conclusions: Patients with KRAS mutation have lower response rates to CRT compared to patients with wild-type KRAS. However, our results indicate that this association does not apply to all KRAS mutations, rather to KRAS codon 13 mutation alone. Thus, these data are the first to show that patients with KRAS codon 13 mutations may not derive pCR from neoadjuvant CRT. [Table: see text]

Sign in / Sign up

Export Citation Format

Share Document