Despite the advantages offered by personalized treatments, there is presently no way to predict response to chemoradiotherapy in patients with non-small cell lung cancer (NSCLC). In this exploratory ...study, we investigated the application of deep learning techniques to histological tissue slides (deep pathomics), with the aim of predicting the response to therapy in stage III NSCLC. We evaluated 35 digitalized tissue slides (biopsies or surgical specimens) obtained from patients with stage IIIA or IIIB NSCLC. Patients were classified as responders (12/35, 34.7%) or non-responders (23/35, 65.7%) based on the target volume reduction shown on weekly CT scans performed during chemoradiation treatment. Digital tissue slides were tested by five pre-trained convolutional neural networks (CNNs)-AlexNet, VGG, MobileNet, GoogLeNet, and ResNet-using a leave-two patient-out cross validation approach, and we evaluated the networks' performances. GoogLeNet was globally found to be the best CNN, correctly classifying 8/12 responders and 10/11 non-responders. Moreover, Deep-Pathomics was found to be highly specific (TNr: 90.1) and quite sensitive (TPr: 0.75). Our data showed that AI could surpass the capabilities of all presently available diagnostic systems, supplying additional information beyond that currently obtainable in clinical practice. The ability to predict a patient's response to treatment could guide the development of new and more effective therapeutic AI-based approaches and could therefore be considered an effective and innovative step forward in personalised medicine.
The primary goal of precision medicine is to minimize side effects and optimize efficacy of treatments. Recent advances in medical imaging technology allow the use of more advanced image analysis ...methods beyond simple measurements of tumor size or radiotracer uptake metrics. The extraction of quantitative features from medical images to characterize tumor pathology or heterogeneity is an interesting process to investigate, in order to provide information that may be useful to guide the therapies and predict survival. This paper discusses the rationale supporting the concept of radiomics and the feasibility of its application to Non-Small Cell Lung Cancer in the field of radiation oncology research. We studied 91 stage III patients treated with concurrent chemoradiation and adaptive approach in case of tumor reduction during treatment. We considered 12 statistics features and 230 textural features extracted from the CT images. In our study, we used an ensemble learning method to classify patients' data into either the adaptive or non-adaptive group during chemoradiation on the basis of the starting CT simulation. Our data supports the hypothesis that a specific signature can be identified (AUC 0.82). In our experience, a radiomic signature mixing semantic and image-based features has shown promising results for personalized adaptive radiotherapy in non-small cell lung cancer.
Purpose
Chemoradiation is the standard treatment in patients with locally advanced non-small-cell lung cancer (LA-NSCLC), and thanks to the recent combination with immunotherapy, median survival has ...unexpectedly improved. This study aims to evaluate early changes in cardiac function after chemoradiotherapy (CRT) in LA-NSCLC by multimodal use of advanced imaging techniques.
Materials and methods
This is a prospective, observational cohort study. At the beginning of combined treatment, screening tests including blood samples, electrocardiogram (ECG), echocardiographic examination (TTE), and cardiac magnetic resonance were performed in all patients with LA-NSCLC. ECG and cardiac marker assays were performed weekly during treatment. ECG and TTE were performed at month 1 (M1) and month 3 (M3) after the end of CRT.
Results
This preliminary analysis included thirty-four patients with a mean age of 69.5 years. The median follow-up was 27.8 months. 62% of patients were in stage IIIA. Radiation therapy was delivered with a median total dose of 60 Gy with conventional fractionation. All patients were treated with concurrent CRT, and 65% of cases were platinum-based therapy. Global longitudinal strain (GLS) and ejection fraction (EF) progressively decreased from baseline to M1 and M3. There was a strong correlation between GLS and EF reduction (at M1:
p
= 0.034; at M3:
p
= 0.018). Cardiac arrhythmias occurred in eight patients (23.5%) at a mean follow-up of 15.8 months after CRT.
Conclusions
Reduction in GLS is an early sign occurring after the end of CRT for LA-NSCLC. Future studies are needed to identify variables that can increase the risk of cardiac events in this patient population to implement adequate damage prevention strategies.
Background: This study aims to assess the benefit of a deep inspiration breath hold (DIBH) over the standard irradiation technique, and eventually to identify anatomical and/or treatment preplanning ...characteristics correlated with the LAD dose. Methods: Patients with left-sided breast cancer undergoing whole breast radiotherapy with DIBH were analyzed. All patients included in the analysis had plans in DIBH and free-breathing (FB). Receiving operating characteristics (ROC analysis) were used to identify the cut-off point of parameters to predict the LAD maximum dose > 10 Gy and LAD mean dose > 4 Gy, and the areas under the curve (AUCs) were computed. Post-test probability has been performed to evaluate the effect of parameters’ combination. Results: One hundred ninety-seven patients were analyzed. The LAD dose was significantly reduced in DIBH plans with the maximum and mean dose reduced by 31.7% (mean value 3.5 Gy vs. 4.8 Gy, p ≤ 0.001) and 28.1% (mean value 8.2 Gy vs. 12.8 Gy, p ≤ 0.001) in DIBH plans compared to FB plans. The strongest predictor of the LAD dose (maximum > 10 Gy and mean > 4 Gy) was the minimum distance of LAD from tangent open fields. Other parameters were lung volume and heart volume (LAD Dmax > 10 Gy) and lung volume, heart volume, and breast separation (LAD Dmean > 4 Gy). Conclusion: The dosimetric advantage of DIBH is clear in all patients and DIBH should always be preferred.
•The Covid-19 pandemic outbreak has a strong impact on all areas of human existence also affecting the quality of life of cancer patients.•The study aim to identify the psychological impact of the ...Covid-19 pandemic on cancer patients undergoing radiation therapy investigating the possible increased of emotional distress during the pandemic outbreak.•The research showed the presence of a clinically significant anxiety in almost all of the sample analyzed.•The research results show a condition of emotional distress present during the pandemic which makes it necessary to treat patients in a multidisciplinary perspective that includes psychological support in the care plan.
In March 2020, the World Health Organization (WHO) characterized the outbreak of the coronavirus disease 2019 (COVID-19) as a pandemic. The aim of this study was to evaluate the psychological impact of the COVID-19 pandemic on cancer patients undergoing radiotherapy.
Were enrolled 210 patients in treatment and in follow-up who had access to the Radiation Oncology Department of the Campus Bio-Medico University Hospital Foundation between April and May 2020. The sample was subjected to structured interview and validated questionnaires. 37% of patients showed significant levels of distress; depressive symptoms were reported by 22.4% of patients and 99% of sample had clinically significant anxiety symptoms. All patients anxiety worsened during the COVID-19 pandemic (p=< 0.001). Patients on active treatment had higher levels of distress (3.5 vs 2.6; p = 0.04) and anxiety (3.5 vs 2.6; p = 0.04). Lung cancer patients appeared to be more afraid of COVID-19 than other patients (24.2 vs 22.2). This study highlights the presence of clinically significant anxiety in 99% of sample. This conclusion reflects the condition of emotional distress present during the pandemic which makes it necessary to treat patients in a multidisciplinary perspective that includes psychological support in the care plan.
Purpose
The objective of this study was to determine the effect on resection rate and survival of neoadjuvant chemoradiotherapy for primarily unresectable locally advanced pancreatic carcinoma.
...Methods
A systematic review of recently published literature was performed. Resection rates and survival data were derived from reports published from 2000 onwards. Only recent studies, based on radiotherapy with standard dose and fractionation, have been analyzed.
Results
Thirteen studies with a total of 510 patients met selection criteria. A resection rate of 8.3–64.2% was reported (median, 26.5%). Of the operated patients, 57.1–100% (median, 87.5%) had R0 tumor resection. Most papers reported occasional pathological complete responses (CR, 3.0–8.8%). When outcome in all patients was considered, median survival ranged from 9 to 23 (median, 13.3) months, comparing favorably with literature data based on concurrent chemoradiation alone (range, 8.6–13 months). Surprisingly, in patients with unresectable tumor at presentation, median survival after surgery ranged from 16.4 to 32.3 (median, 23.6) months.
Conclusions
The finding of a high proportion of R0 resection among all resections performed confirms the activity of neoadjuvant radiochemotherapy and should not be neglected. Based on these data, patients with unresectable pancreatic cancer without disease progression after chemoradiotherapy should be considered for radical surgery.
Current practice in cancer treatment collects multimodal data, such as radiology images, histopathology slides, genomics and clinical data. The importance of these data sources taken individually has ...fostered the recent rise of radiomics and pathomics, i.e., the extraction of quantitative features from radiology and histopathology images collected to predict clinical outcomes or guide clinical decisions using artificial intelligence algorithms. Nevertheless, how to combine them into a single multimodal framework is still an open issue. In this work, we develop a multimodal late fusion approach that combines hand-crafted features computed from radiomics, pathomics and clinical data to predict radiotherapy treatment outcomes for non-small-cell lung cancer patients. Within this context, we investigate eight different late fusion rules and two patient-wise aggregation rules leveraging the richness of information given by CT images, whole-slide scans and clinical data. The experiments in leave-one-patient-out cross-validation on an in-house cohort of 33 patients show that the proposed fusion-based multimodal paradigm, with an AUC equal to 90.9%, outperforms each unimodal approach, suggesting that data integration can advance precision medicine. The results also show that late fusion favourably compares against early fusion, another commonly used multimodal approach. As a further contribution, we explore the chance to use a deep learning framework against hand-crafted features. In our scenario characterised by different modalities and a limited amount of data, as it may happen in different areas of cancer research, the results show that the latter is still a viable and effective option for extracting relevant information with respect to the former.
A prospective dose escalation trial was developed to evaluate the maximum tolerated dose of stereotactic body radiotherapy (SABRT) to primary breast cancer in stage IV disease. The aim of the present ...report was to describe safety and outcome of the first dose level cohort of patients.
Patients with histologically confirmed diagnosis of invasive breast carcinoma (biological immuno-histochemical profile: luminal and/or HER2 positive) and distant metastatic disease not progressing after 6 months of systemic therapy with a tumor CT or 5FDG-PET detectable were deemed eligible. The starting dose was 40 Gy in 5 fractions (level 1) because this dose proved to be safe in previous dose-escalation trial on adjuvant stereotactic body radiotherapy. The maximum dose level was chosen as 45 Gy in 5 fractions. Dose limiting toxicity was any grade 3 or worse toxicity according to CTCAE v.4. Time-to-event Keyboard (TITE-Keyboard) design (Lin and Yuan, Biostatistics 2019) was used to find the maximum tolerated dose (MTD). MTD was the dose of radiotherapy associated with a ≤ 20% rate pre-specified treatment-related dose-limiting toxicity (DLT).
To date 10 patients have been treated at the starting dose level. Median age was 80 years (range 50-89). 7 patients had a luminal disease, while 3 patients had an HER2 positive disease. No patient suspended ongoing systemic treatment. No protocol defined DLTs were observed. Grade 2 skin toxicity occurred in 4 patients with diseases located close to or involving the skin. Median follow-up was 13 months and all 10 patients were evaluable for response: 5 achieved a complete response, 3 achieved a partial response and 2 showed a stable disease, all with a clinical benefit (resolution of skin retraction, bleeding and pain). The mean reduction in the sum of the largest diameters of target lesions was of 61.4% (DS=17.0%).
SABR to primary breast cancer seems feasible and is associated with symptoms reduction. Continued accrual to this study is needed to confirm the safety and assess the MTD.
ClinicalTrials.gov, identifier NCT05229575.
In breast cancer volumetric-modulated arc therapy (VMAT) planning, the rotation of the gantry around the target implies a greater dose spreading to the whole heart, compared to tangential-field ...standard treatment. A consecutive cohort of 121 breast cancer patients treated with the VMAT technique was investigated. The correlation of breast volume, heart volume and lung volume with mean heart dose (mHD) and mean and maximum LAD dose (mLAD dose, MLAD dose) was tested, and a subsequent a linear regression analysis was carried out. VMAT treatment plans from 56 left breast cancer and 65 right breast cancer patients were analyzed. For right-sided patients, breast volume was significantly correlated with mHD, mLAD and MLAD dose, while for left-sided patients, breast volume was significantly correlated with mHD and mLAD, while heart volume and lung volume were correlated with mHD, mLAD and MLAD dose. Breast volume was the only predictor of increased heart and LAD dose (
≤ 0.001) for right-sided patients. In left-sided patients, heart and lung were also predictors of increased mHD (
= 0.005,
≤ 0.001) and mean LAD dose (
= 0.009,
≤ 0.001). In this study, we observed an increase in heart and LAD doses in larger-breasted patients treated with VMAT planning. In right-sided patients, breast volume was shown to be the only predictor of increased heart dose and LAD dose.
Brain metastases (BMs) are common among patients affected by HER2+ metastatic breast cancer (>30%). The management of BMs is usually multimodal, including surgery, radiotherapy, systemic therapy and ...palliative care. Standard brain radiotherapy (RT) includes the use of stereotactic radiotherapy (SRT) for limited disease and whole brain radiotherapy (WBRT) for extensive disease. The latter is an effective palliative treatment but has a reduced effect on brain local control and BM overall survival, as it is also associated with severe neurocognitive sequelae. Recent advances both in radiation therapy and systemic treatment may change the paradigm in this subset of patients who can experience long survival notwithstanding BMs. In fact, in recent studies, SRT for multiple BM sites (>4) has shown similar efficacy when compared to irradiation of a limited number of lesions (one to three) without increasing toxicity. These findings, in addition to the introduction of new drugs with recognized intracranial activity, may further limit the use of WBRT in favor of SRT, which should be employed for treatment of both multiple-site BMs and for oligo-progressive brain disease. This review summarizes the supporting literature and highlights the need for optimizing combinations of the available treatments in this setting, with a particular focus on radiation therapy.
PDF
