Localized fine-tuning and clinical evaluation of deep-learning based auto-segmentation (DLAS) model for clinical target volume (CTV) and organs-at-risk (OAR) in rectal cancer radiotherapy.

BACKGROUND AND PURPOSE: Various deep learning auto-segmentation (DLAS) models have been proposed, some of which have been commercialized. However, the issue of performance degradation is notable when pretrained models are deployed in the clinic. This study aims to enhance precision of a popular commercial DLAS product in rectal cancer radiotherapy by localized fine-tuning, addressing challenges in practicality and generalizability in real-world clinical settings. MATERIALS AND METHODS: A total of 120 Stage II/III mid-low rectal cancer patients were retrospectively enrolled and divided into three datasets: training (n = 60), external validation (ExVal, n = 30), and generalizability evaluation (GenEva, n = 30) datasets respectively. The patients in the training and ExVal dataset were acquired on the same CT simulator, while those in GenEva were on a different CT simulator. The commercial DLAS software was first localized fine-tuned (LFT) for clinical target volume (CTV) and organs-at-risk (OAR) using the training data, and then validated on ExVal and GenEva respectively. Performance evaluation involved comparing the LFT model with the vendor-provided pretrained model (VPM) against ground truth contours, using metrics like Dice similarity coefficient (DSC), 95th Hausdorff distance (95HD), sensitivity and specificity. RESULTS: LFT significantly improved CTV delineation accuracy (p < 0.05) with LFT outperforming VPM in target volume, DSC, 95HD and specificity. Both models exhibited adequate accuracy for bladder and femoral heads, and LFT demonstrated significant enhancement in segmenting the more complex small intestine. We did not identify performance degradation when LFT and VPM models were applied in the GenEva dataset. CONCLUSIONS: The necessity and potential benefits of LFT DLAS towards institution-specific model adaption is underscored. The commercial DLAS software exhibits superior accuracy once localized fine-tuned, and is highly robust to imaging equipment changes.

Exploration of lymph node recurrence patterns and delineation guidelines of radiation field in middle thoracic oesophageal carcinomas after radical surgery: a real-world study.

BACKGROUND: Oesophageal squamous cell carcinoma is one of the most commonly diagnosed carcinomas in China, and postoperative radiotherapy plays an important role in improving the prognosis of patients. Carcinomas in different locations of the oesophagus could have different patterns of lymph node metastasis after surgery. METHODS: In this multicentric retrospective study, we enrolled patients with middle thoracic oesophageal squamous cell carcinomas from 3 cancer centres, and none of the patients underwent radiotherapy before or after surgery. We analysed the lymph node recurrence rates in different stations to explore the postoperative lymphatic recurrence pattern. RESULTS: From January 1st, 2014, to December 31st, 2019, 132 patients met the criteria, and were included in this study. The lymphatic recurrence rate was 62.1%. Pathological stage (P = 0.032) and lymphadenectomy method (P = 0.006) were significant predictive factors of lymph node recurrence. The recurrence rates in the supraclavicular, upper and lower paratracheal stations of lymph nodes were 32.6%, 28.8% and 16.7%, respectively, showing a high incidence. The recurrence rate of the subcarinal node station was 9.8%, while 8.3% (upper, middle and lower) thoracic para-oesophageal nodes had recurrences. CONCLUSIONS: We recommend including the supraclavicular, upper and lower paratracheal stations of lymph nodes in the postoperative radiation field in middle thoracic oesophageal carcinomas. Subcarinal station is also potentially high-risk, while whether to include thoracic para-oesophageal or abdominal nodes needs careful consideration.

Prognostic impact of sarcopenia in patients with locally advanced adenocarcinoma of the esophagogastric junction treated with neoadjuvant chemoradiotherapy.

Background: Few studies have evaluated the significance of sarcopenia in predicting the outcomes of patients with adenocarcinoma of the esophagogastric junction (AEG), especially those who received neoadjuvant chemoradiotherapy (NCRT). We aimed to identify the sarcopenic status and its impact on the outcomes of patients with locally advanced AEG who received NCRT followed by radical surgery or systemic therapy. Materials and methods: Patients with T3-4N+M0 AEG with accessible abdominal computed tomography (CT) before and after NCRT were retrospectively analyzed. Body composition parameters, particularly the skeletal muscle index (SMI), were assessed using a CT-based method, and sarcopenia was defined using a predetermined SMI cutoff value. Survival analysis was conducted using the Kaplan-Meier method. A Cox proportional hazards regression model was used to identify independent prognostic factors. Receiver operating characteristic curve analysis was carried out, and the area under the curve (AUC) was calculated to test the prognostic accuracy of different factors. Results: A total of 63 patients were enrolled, 65.1 and 79.4% of whom developed pre- and post-NCRT sarcopenia, respectively. Patients with pre-NCRT sarcopenia had lower radical surgery rates (70.7 vs. 95.5%, p = 0.047) than those without sarcopenia; however, sarcopenic status did not affect other short-term outcomes, including treatment-related toxicity and efficacy. Pre-NCRT sarcopenia was identified as an independent predictive factor for poor overall survival (OS) [adjusted hazard ratio (HR), 6.053; p = 0.002] and progression-free survival (PFS) (adjusted HR, 2.873; p = 0.031). Compared with nutritional indices such as the Nutritional Risk Screening 2002, weight loss during NCRT, and post-NCRT sarcopenia, pre-NCRT sarcopenia was regarded as the best predictive index for the 5-year OS (AUC = 0.735) and PFS rates (AUC = 0.770). Conclusion: Pre-NCRT sarcopenia may be an independent predictive factor for OS and PFS rates in patients with locally advanced AEG receiving multimodal treatment.

Postoperative lymphatic recurrence distribution and delineation of the radiation field in lower thoracic squamous cell esophageal carcinomas: a real-world study.

BACKGROUND: To study lymphatic recurrence distribution after radical surgery in the real world and guide clinical tumor volume delineation for regional lymph nodes during postoperative radiotherapy for lower thoracic squamous cell esophageal carcinomas. METHODS: We enrolled patients who underwent radical esophagectomy, without radiation before or after surgery, at 3 cancer hospitals. Patients were classified into groups according to tumor locations. We included patients with tumors in the lower thoracic segment and analyzed the postoperative lymph node recurrence mode. A cutoff value of 10% was used to differentiate high-risk lymph node drainage areas from others. RESULTS: We enrolled 1905 patients in the whole study series, including 652 thoracic esophageal carcinomas that met our inclusion criteria; there were 241 cases of lower thoracic esophageal carcinomas. 1st, 2nd, 4th, 7th, 8th groups of lymph nodes, according to the 8th edition of the AJCC classification, displayed as high-risk recurrence areas, representing 17.8%, 23.9%, 11.7%, 10.9% and 12.2% of lymph node recurrence. Stage III-IV tumors located in the lower segment of the thoracic esophagus showed a tendency to recur in the left gastric nodes (7.9%) and celiac nodes (10.6%). CONCLUSIONS: According to our results, we recommended including the 4th, 7th and 8th groups of lymph nodes in the radiation field, and for patients with stage III-IV disease, the 17th and 20th groups of nodes should be irradiated during postoperative treatment. Whether including 1st/2nd groups in preventive irradiation needed more proofs.

Establishment of prognostic models for adenocarcinoma of oesophagogastric junction patients with neoadjuvant chemoradiotherapy: a real-world study.

BACKGROUND: Multimodal therapies based on surgical resection have been recommended for the treatment of adenocarcinoma of the oesophagogastric junction (AEG). We aimed to evaluate prognostic factors in AEG patients receiving neoadjuvant chemoradiotherapy and to build predictive models. METHODS: T3 - T4N + M0 AEG patients with resectable Siewert type II/III tumours were enrolled in this study. All patients underwent neoadjuvant chemoradiation, followed by radical surgery or systemic therapy according to clinical response. Survival analysis was performed using the Kaplan-Meier method; multivariate analysis using the Cox proportional hazards method was also conducted. The Harrell concordance index (C-index) was used to test the prognostic value of models involving prognostic factors, and consistency between actual and predicted survival rates was evaluated by calibration curves. RESULTS: From February 2009 to February 2018, 79 patients were treated with neoadjuvant chemoradiotherapy; 60 patients of them underwent radical surgery. The R0 resection rate was 98.3%, and 46.7% of patients achieved a major pathologic response (MPR), namely, a residual tumour issue less than 10%. The 5-year overall survival (OS) rate was 63%, and the 5-year progression-free survival (PFS) rate was 48%. The incidence of grade 3 complications was 21.5%, and no grade 4 complications were reported. According to the results of univariate and multivariate analyses, we included the neutrophil-lymphocyte ratio (NLR), prognostic nutrition index (PNI), eosinophilic granulocyte (EOS) and postoperative pathologic stage in nomogram analysis to establish prediction models for OS and PFS; the C-index of each model was 0.814 and 0.722, respectively. Both the C-index and calibration curves generated to validate consistency between the actual and predicted survival indicated that the models were well calibrated and of good predictive value. CONCLUSIONS: AEG patients achieved favourable downstaging and pathologic response after neoadjuvant chemoradiation, with acceptable adverse effects. Inflammation-based and nutrition-related factors and postoperative pathologic stage had a significant influence on OS and PFS, and the predictive value was verified through prognostic models.

Lymph nodes metastasis in cervical cancer: Incidences, risk factors, consequences and imaging evaluations.

AIM: The aim of this study was to determine the incidence, risk factors, consequences and accuracy of imaging evaluation of lymph node (LN) metastasis in a cohort of 406 patients treated with radical hysterectomy (RH), lymphadenectomy of pelvic LN (PLN) and para-aortic LN (PALN), which was performed primarily by one physician. METHODS: From February 2001 to November 2015, patients with cervical cancer of FIGO stage IB were included, if they received RH of class III or type C performed by Dr. M. Wu in Peking Union Medical College Hospital. Follow-up ended in December 2016. Incidences and accuracy of imaging evaluation of LN metastasis were described, and predictive factors of LN metastasis and its impact on survival outcomes were determined in univariate and multivariate models. RESULTS: Among 406 patients with clinical stage IB, 57 (14.0%) had lymphatic metastasis. In multivariate model, positive parametrium was independent factor for general LN metastasis (odds ratio [OR] 5.1; 95% confidence interval [95% CI], 2.1-12.1) and PLN metastasis (OR 5.3; 95% CI, 2.2-12.8). Positive PLN was independent factor for metastasis to common iliac LN and PALN. After adjusted with clinico pathologic factors, general and site-specific LN metastases were independent risk factors of progression-free survival and overall survival (all P values <0.05). Preoperative imaging evaluation had low sensitivity but high specificity for predicting LN metastasis. Various imaging methods had similar predictive accuracy. CONCLUSION: Lymphatic metastasis was significantly related to the clinico pathologic characters and survival of cervical cancer. More sensitive preoperative evaluation is needed for predicting the LN metastasis.

Surgical Management of Congenital Arteriovenous Fistula near Lateral Cranial Base.

An arteriovenous fistula is an aberrant communication between an artery and a vein. Here is a case in which risk factors including multiple fistulae, dangerous location, and high-flow blood occurred simultaneously. The patient is a 13-year-old boy diagnosed with bilateral arteriovenous fistulae between the jugular vein and carotid artery separately, with cardiac dilatation and pulmonary artery hypertension. We performed right arteriovenous fistulae resection, ligated the involved small vessels, and tightened the internal jugular vein. The tiny left arteriovenous fistulae were treated with transarterial embolism. The blood flow bruit and swelling of his neck completely disappeared, with clear improvement of his life quality, the pulmonary artery pressure has dropped from 54 mmHg to 39 mmHg. Surgery is one of the main therapies of arteriovenous fistula currently, we need to make a balance between good effect and low risk, aiming at early treatment to avoid serious complications.