Pretreatment clinical findings predict outcome for patients receiving preoperative radiation for rectal cancer

Ensemble learning for glioma patients overall survival prediction using pre-operative MRIs

Objective: Gliomas are the most common primary brain tumors. Approximately 70% of the glioma patients diagnosed with glioblastoma have an averaged overall survival (OS) of only ∼16 months. Early survival prediction is essential for treatment decision-making in glioma patients. Here we proposed an ensemble learning approach to predict the post-operative OS of glioma patients using only pre-operative MRIs.Approach: Our dataset was from the Medical Image Computing and Computer Assisted Intervention Brain Tumor Segmentation challenge 2020, which consists of multimodal pre-operative MRI scans of 235 glioma patients with survival days recorded. The backbone of our approach was a Siamese network consisting of twinned ResNet-based feature extractors followed by a 3-layer classifier. During training, the feature extractors explored traits of intra and inter-class by minimizing contrastive loss of randomly paired 2D pre-operative MRIs, and the classifier utilized the extracted features to generate labels with cost defined by cross-entropy loss. During testing, the extracted features were also utilized to define distance between the test sample and the reference composed of training data, to generate an additional predictor via K-NN classification. The final label was the ensemble classification from both the Siamese model and the K-NN model.Main results: Our approach classifies the glioma patients into 3 OS classes: long-survivors (>15 months), mid-survivors (between 10 and 15 months) and short-survivors (<10 months). The performance is assessed by the accuracy (ACC) and the area under the curve (AUC) of 3-class classification. The final result achieved an ACC of 65.22% and AUC of 0.81.Significance: Our Siamese network based ensemble learning approach demonstrated promising ability in mining discriminative features with minimal manual processing and generalization requirement. This prediction strategy can be potentially applied to assist timely clinical decision-making.

Preoperative short-course radiation therapy for rectal cancer provides excellent disease control and toxicity: Results from a single US institution

PURPOSE

Preoperative short-course radiation therapy (SCRT) has rarely been used for rectal cancer in the United States, although 2 randomized phase 3 trials demonstrate equivalence to conventional chemoradiation (CRT), and recent updates to national guidelines include this regimen as a treatment option. We sought to evaluate the efficacy and safety of preoperative SCRT followed by immediate surgery within 1 week to treat rectal cancer in the US setting.

METHODS AND MATERIALS

All patients treated with preoperative SCRT (4 Gy × 5 fractions for total 20 Gy) followed by planned surgery within 1 week at our institution were retrospectively evaluated. Censored cases with ≥2 years of follow-up were included along with any disease failure or death. Patients with cM1 disease were excluded. Patients with yp stage II/III disease typically received adjuvant chemotherapy from the 1990s onwards. The primary outcomes were actuarial (Kaplan-Meier) 5-year locoregional control (LC), disease-free survival (DFS), and overall survival (OS) as well as late severe (greater than or equal to grade 3) toxicity.

RESULTS

Our analysis included 202 consecutive patients with clinical stage I-III disease treated from 1977 through 2011. Median follow-up was 6.5 years (range, 2-29.2). Five-year disease outcomes were 95.9% ± 1.5% for LC, 76.4% ± 3.1% for DFS, and 84.6% ± 2.6% for OS. For patients with locally advanced rectal cancer (cT3-4 and/or cN+), 5-year LC, DFS, and OS were 95.1% ± 2.1%, 73.3% ± 4.3%, and 80.6% ± 3.7%, respectively. The late severe toxicity rate was 11.4%.

CONCLUSIONS

SCRT followed by immediate surgery is a safe and effective treatment for patients with rectal cancer in the United States. Though SCRT has not been widely adopted, recent updates to the national guidelines for rectal cancer as well as financial pressures to reduce healthcare costs may lead to increased utilization of this treatment regimen in the future.

Prognostic Effect of Pretreatment Serum Carcinoembryonic Antigen Level: A Useful Tool for Prediction of Distant Metastasis in Locally Advanced Rectal Cancer Following Neoadjuvant Chemoradiotherapy and Total Mesorectal Excision

Many studies have reported the prognostic value of pretreatment serum carcinoembryonic antigen (pre-CEA) levels on colorectal cancer outcomes. However, controversy remains concerning the significance of pre-CEA levels in patients with rectal cancer treated with neoadjuvant chemoradiotherapy (CRT). Our aim in this study was to investigate the prognostic role of the pre-CEA level in patients with locally advanced rectal cancer undergoing neoadjuvant CRT followed by total mesorectal excision (TME).A total of 419 patients with stages II and III rectal cancer treated with neoadjuvant CRT followed by TME with available pre-CEA data were included. The outcomes studied were 5-year local recurrence-free survival (LRFS), distant metastasis-free survival (DMFS), and disease-free survival (DFS). Optimal pre-CEA cutoff values to predict DMFS were determined based on current smoking history.The median pre-CEA level of smokers was 3.8 ng/mL, and that of nonsmokers was 2.8 ng/mL (P < 0.01). Pre-CEA levels of 6.6 ng/mL for nonsmokers and 11.4 ng/mL for smokers were determined to best separate patients on the basis of time to distant metastasis by using log-rank statistics. The pre-CEA level was associated with DMFS (hazard ratio = 1.743, 95% confidence interval = 1.129-2.690, P = 0.01). The pre-CEA level was not associated with LRFS or DFS.The pre-CEA level appears to be a significant preoperative prognostic factor. Moreover, it is as valuable as any known pathologic factor. Future studies evaluating oncologic outcomes should take into consideration the pre-CEA level.

Five fractions of radiation therapy followed by 4 cycles of FOLFOX chemotherapy as preoperative treatment for rectal cancer

BACKGROUND

Preoperative radiation therapy with 5-fluorouracil chemotherapy is a standard of care for cT3-4 rectal cancer. Studies incorporating additional cytotoxic agents demonstrate increased morbidity with little benefit. We evaluate a template that: (1) includes the benefits of preoperative radiation therapy on local response/control; (2) provides preoperative multidrug chemotherapy; and (3) avoids the morbidity of concurrent radiation therapy and multidrug chemotherapy.

METHODS AND MATERIALS

Patients with cT3-4, any N, any M rectal cancer were eligible. Patients were confirmed to be candidates for pelvic surgery, provided response was sufficient. Preoperative treatment was 5 fractions radiation therapy (25 Gy to involved mesorectum, 20 Gy to elective nodes), followed by 4 cycles of FOLFOX [5-fluorouracil, oxaliplatin, leucovorin]. Extirpative surgery was performed 4 to 9 weeks after preoperative chemotherapy. Postoperative chemotherapy was at the discretion of the medical oncologist. The principal objectives were to achieve T stage downstaging (ypT < cT) and preoperative grade 3+ gastrointestinal morbidity equal to or better than that of historical controls.

RESULTS

76 evaluable cases included 7 cT4 and 69 cT3; 59 (78%) cN+, and 7 cM1. Grade 3 preoperative GI morbidity occurred in 7 cases (9%) (no grade 4 or 5). Sphincter-preserving surgery was performed on 57 (75%) patients. At surgery, 53 patients (70%) had ypT0-2 residual disease, including 21 (28%) ypT0 and 19 (25%) ypT0N0 (complete response); 24 (32%) were ypN+. At 30 months, local control for all evaluable cases and freedom from disease for M0 evaluable cases were, respectively, 95% (95% confidence interval [CI]: 89%-100%) and 87% (95% CI: 76%-98%). Cases were subanalyzed by whether disease met requirements for the recently activated PROSPECT trial for intermediate-risk rectal cancer. Thirty-eight patients met PROSPECT eligibility and achieved 16 ypT0 (42%), 15 ypT0N0 (39%), and 33 ypT0-2 (87%).

CONCLUSION

This regimen achieved response and morbidity rates that compare favorably with those of conventionally fractionated radiation therapy and concurrent chemotherapy.

The role of carcinoembriogenic antigen in predicting response and survival to neoadjuvant chemoradiotherapy for distal rectal cancer

PURPOSE

Carcinoembriogenic antigen (CEA) is the most frequently used tumor marker in rectal cancer. A decrease in carcinoembriogenic antigen after radical surgery is associated with survival in these patients. Neoadjuvant chemoradiotherapy may lead to significant primary tumor downstaging, including complete tumor regression in selected patients. Therefore, we hypothesized that a decrease in CEA after neoadjuvant chemoradiotherapy could reflect tumor response to chemoradiotherapy, affecting final disease stage and ultimately survival.

METHODS

Patients with distal rectal cancer managed by neoadjuvant chemoradiotherapy and available pretreatment and postchemoradiotherapy levels of CEA were eligible for the study. Outcomes studied included final disease stage, relapse, and survival, and these were compared according to initial CEA level, post-chemoradiotherapy CEA level, and the reduction in CEA.

RESULTS

Overall 170 patients were included. Post-chemoradiotherapy CEA levels <5 ng/ml were associated with increased rates of complete clinical response and pathologic response. Additionally, postchemoradiotherapy CEA levels <5 ng/ml were associated with increased overall and disease-free survival (P = 0.01 and P = 0.03). There was no correlation between initial CEA level or reduction in CEA and complete response or survival.

CONCLUSION

A postchemoradiotherapy CEA level <5 ng/ml is a favorable prognostic factor for rectal cancer and is associated with increased rates of earlier disease staging and complete tumor regression. Postchemoradiotherapy CEA levels may be useful in decision making for patients who may be candidates for alterative treatment strategies.

High MUC2 immunohistochemical expression is a predictor of poor response to preoperative radiochemotherapy (RCT) in rectal adenocarcinoma

The purpose of this study is to establish if mucoid differentiation is associated with responsiveness to preoperative radiochemotherapy (RCT) in rectal adenocarcinomas. Thirty-two patients with rectal adenocarcinomas were preoperatively treated with 44 to 46 Gy in 22 to 23 fractions and with 5-fluorouracil (200 to 225 mg/m) before surgery. Mucoid differentiation was searched for both in pre-RCT biopsies with anti-MUC2 antiserum and in postoperative specimens. To evaluate the responsiveness to preoperative RCT, a regression grading was used (grades 0 to 4). Statistical analysis showed a significant negative correlation between immunohistochemical expression of MUC2 in pre-RCT biopsies and regression grade in postoperative specimens (r=-0.529; P=0.002). A significant cutoff value of 60% of MUC2 positive neoplastic cells in pre-RCT biopsies was observed (P=0.018): 13 cases with more than 60% exhibited a poor response to RCT (grade 0 in 5/13, grade 1 in 4/13, grade 2 in 4/13), whereas 19 cases with less than 60% showed a better response to RCT (grade 1 in 6/19, grade 2 in 9/19, grade 3 in 3/19, grade 4 in 1/19).

Oncocytic modifications in rectal adenocarcinomas after radio and chemotherapy

The purpose of the study is to highlight oncocytic modifications in rectal adenocarcinomas and evaluate a possible correlation with preoperative radiochemotherapy (RCT). Twenty-eight cases of advanced rectal carcinoma, treated preoperatively by 5-fluorouracil (200-225 mg/m(2)) and 44-46 Gy in 22-23 fractions, were studied. All patients underwent biopsy before RCT. Surgery was performed within 6 weeks after RCT. In all cases oncocytic modifications were searched for on hematoxylin and eosin (H&E) and at immunohistochemistry using an antimitochondrial antibody. In addition, in two cases, both pre- and post-RCT tissues were examined at electron microscopy. All tumors were adenocarcinomas. In pre-RCT biopsies, oncocytic changes were difficult to find on H&E, while the antimitochondrial antibody strongly stained numerous neoplastic cells (mean 48.4%). In post-RCT surgical specimens, oncocytic changes were detected in 24 out of 28 cases on H&E and the antimitochondrial antibody stained most of the residual neoplastic cells (mean 76.7%). Ultrastructural examination revealed large and bizarre mitochondria inside tumor cells both in pre- and post-RCT tissues. In conclusion, the present data suggest that rectal adenocarcinomas are "mitochondrion-rich" tumors. After preoperative RCT, residual neoplastic cells acquire a definite oncocytic phenotype.

Pathological and molecular predictors of the response of rectal cancer to neoadjuvant radiochemotherapy

AIMS

The prediction of sensitivity and resistance to neoadjuvant therapy has great potential value for many tumour sites. A neoadjuvant regimen is increasingly the gold standard in rectal cancer management and the aim of this review was to highlight predictive markers currently assessed and evaluate their clinical utility.

METHODS

A systematic search of Medline was conducted using the following keywords 'colorectal', 'neoadjuvant', 'molecular', 'predict' and 'radiotherapy'. Original manuscripts from all relevant listings were sourced. These were hand searched for further articles of relevance.

RESULTS

Conventional indices including tumour stage and grade were unable to predict histological response. Immunohistochemical assessment of P53 gene, Bcl 2, Bax and microsatellite instability are of no predictive value. Studies utilising molecular response predictors from archival pre-treatment tumour tissues have identified several promising predictive markers including p21, spontaneous apoptosis and direct sequencing of the p53 gene. Global gene expression from fresh pre-treatment tissue using cDNA microarray has only recently been assessed but identified expression differences between 54 genes and was able to predict response with 78% sensitivity and 86% specificity.

CONCLUSIONS

Currently there are no clinically useful predictors of response based on standard pathological assessment and immunocytochemistry. Direct gene sequencing of p53, studies of apoptosis and global gene sequencing may hold promise.

Usefulness of magnetic resonance volumetric evaluation in predicting response to preoperative concurrent chemoradiotherapy in patients with resectable rectal cancer

PURPOSE

We performed magnetic resonance (MR) volumetry before and after neoadjuvant chemoradiation for evaluating response to therapy in T3 and T4 rectal cancer. To investigate the utility of MR volumetry for predicting the response to neoadjuvant chemoradiation, we compared results from MR volumetry before chemoradiation with those after chemoradiation.

METHODS AND MATERIALS

A total 112 patients with T3 or T4 rectal cancer who successfully underwent MR volumetry and completed neoadjuvant chemoradiation followed by radical resection for cure were identified. MR volumetries were performed before and after chemoradiation. We compared pre- and postchemoradiation tumor volume and % volume reduction rates of patients whose tumors were down-staged with those of patients that were not down-staged. The same analyses were also performed between those patients having a complete histologic regression and those with residual disease in the operative specimen. We assessed the difference of % volume reduction rate according to Dworak's rectal cancer regression grades.

RESULTS

Fifty-seven patients (50.9%) demonstrated a tumor down-staging after chemoradiation therapy. Both pre- and posttreatment MR tumor volumes were significantly less in patients whose tumors were down-staged than in patients that were not down-staged (p = 0.04, 0.031), and % volume reduction rates were significantly higher in patients whose tumors were down-staged (p = 0.024). Sixteen patients (14.3%) showed pathologically complete tumor regression. The differences of MR tumor volumes before and after chemoradiation and % volume reduction rates were not significantly different between patients having a complete histologic regression and those with residual disease (p = 0.688, 0.451, and 0.480). The differences of % volume reduction rates according to Dworak's grades were statistically significant (p = 0.03).

CONCLUSION

The MR volumetric examinations before and after chemoradiation demonstrated the significant difference of tumor volume and % volume reduction rate between patients whose tumors were down-staged and those that were not down-staged. The volume reduction rates were significantly different among groups according to Dworak's grades. However, the MR volumetric evaluation could not identify any differences between those patients having a complete histologic regression and those with residual disease.

Technical aspects of image-based treatment planning of rectal carcinoma

The role of 3-dimensional treatment planning in the management of rectal cancer is not well defined. This report reviews technical guidelines for simulation, defining target volumes, and suggested beam configurations. Outcome from pilot clinical trials, with emphasis on the impact of volume and dose on tumor response and treatment morbidity is discussed.

Urinary tract involvement by colorectal cancer

PURPOSE

Involvement of the urinary tract by colorectal cancer is sufficiently rare to be encountered by an individual surgeon on an infrequent basis. The aim of this review is to highlight technical and oncologic issues that should be considered when dealing with complex colorectal cancer that involves the urinary tract.

METHODS

The relevant literature from 1975 to 2001 was identified using the MEDLINE database of the U.S. National Library of Medicine and reviewed. Because of the diversity of forms of presentation of urologic involvement, few randomized, controlled trials are available, with most evidence derived from retrospective studies.

RESULTS

Three distinct clinical situations in which the urinary tract may be affected by colorectal cancer were identified: involvement by primary colorectal cancer, involvement by recurrent cancer, and unexpected intraoperative findings of urinary tract involvement. Management strategies to identify and treat locally advanced primary or recurrent colorectal cancer involving the urinary tract improve survival with acceptable morbidity and mortality. Careful preoperative assessment of all patients with colorectal cancer will reduce unexpected identification of urinary tract invasion at the time of surgery. In patients in whom cure is not possible, endourologic techniques combined with judicious surgical resection can provide high-quality palliation. Optimal care of many of these conditions is facilitated by specialist urologic advice.

CONCLUSIONS

The wide spectrum of possible urinary tract involvement by colorectal cancer requires individual patient-specific and disease-specific consideration. The literature offers important guidelines that aid decision making and improve management of these challenging problems.

The addition of continuous infusion 5-FU to preoperative radiation therapy increases tumor response, leading to increased sphincter preservation in locally advanced rectal cancer

PURPOSE

To compare the outcome from preoperative chemoradiation (CXRT) and from radiation therapy (RT) in the treatment of rectal cancer in two large, single-institutional experiences.

PATIENTS AND METHODS

Between 1978 and 1995, 403 patients with localized, nonmetastatic, clinically staged T3 or T4 rectal cancer patients were treated with preoperative RT alone at two institutions. Patients at institution 1 (n = 207) were treated with pelvic CXRT exclusively, and patients at institution 2 were treated (except for 8 given CXRT) with pelvic RT alone (n = 196). In addition, a third group (n = 61) was treated with CXRT at institution 2 between 1998 and 2000 after a policy change. Both institutions delivered 45 Gy in five fractions as a standard dose, but institution 2 used 20 Gy in five fractions in selected cases (n = 26). At both institutions, concurrent chemotherapy consisted of a continuous infusion of 5-fluorouracil (5-FU) at a dosage of 1500 mg/m(2)/week. The end points were response, sphincter preservation (SP), relapse-free survival (RFS), pelvic disease control (PC), and overall survival (OS).

RESULTS

Median follow-up was 63 months for all living patients at institution 1 and in the primary group of institution 2. Multivariate analysis of the patients in these groups showed that the use of concurrent chemotherapy improved tumor response (T-stage downstaging, 62% vs. 42%, p = 0.001, and pathologic complete response, 23% vs. 5% p < 0.0001), but did not significantly improve LC, RFS, or OS. Follow-up for the secondary group at institution 2 was insufficient to allow the analysis of these endpoints. In the subset of patients receiving 45 Gy who had rectal tumors < or /=6 cm from the anal verge (institution 1: n = 132; institution 2 primary: n = 79; institution 2 secondary: n = 33), there was a significant improvement in SP with the use of concurrent chemotherapy (39% at institution 1 compared with 13% in the primary group at institution 2, p < 0.0001). A logistic regression analysis of clinical prognostic factors indicated that the use of concurrent chemotherapy independently influenced SP in these low tumors (p = 0.002). This finding was supported by a 36% SP rate in the secondary group at institution 2. Thus SP increased after the addition of chemotherapy at institution 2.

CONCLUSIONS

The use of concurrent 5-FU with preoperative radiation therapy for T3 and T4 rectal cancer independently increases tumor response and may contribute to increased SP in patients with low rectal cancer.

Surgical treatment of rectal cancer

Rectal cancer should no longer be thought of as only a surgically treated disease. Centers that treat large numbers of rectal cancer patients should provide state of the art radiotherapy and chemotherapy as well as offer anatomic tumor-specific operations for advanced-stage cancers and local treatment options for favorable, early lesions.