RESUMO
Background and purpose: Radiation-induced toxicities are common adverse events in lung cancer (LC) patients undergoing radiotherapy (RT). An accurate prediction of these adverse events might facilitate an informed and shared decision-making process between patient and radiation oncologist with a clearer view of life-balance implications in treatment choices. This work provides a benchmark of machine learning (ML) approaches to predict radiation-induced toxicities in LC patients built upon a real-world health dataset based on a generalizable methodology for their implementation and external validation. Materials and Methods: Ten feature selection (FS) methods were combined with five ML-based classifiers to predict six RT-induced toxicities (acute esophagitis, acute cough, acute dyspnea, acute pneumonitis, chronic dyspnea, and chronic pneumonitis). A real-world health dataset (RWHD) built from 875 consecutive LC patients was used to train and validate the resulting 300 predictive models. Internal and external accuracy was calculated in terms of AUC per clinical endpoint, FS method, and ML-based classifier under analysis. Results: Best performing predictive models obtained per clinical endpoint achieved comparable performances to methods from state-of-the-art at internal validation (AUC ≥ 0.81 in all cases) and at external validation (AUC ≥ 0.73 in 5 out of 6 cases). Conclusion: A benchmark of 300 different ML-based approaches has been tested against a RWHD achieving satisfactory results following a generalizable methodology. The outcomes suggest potential relationships between underrecognized clinical factors and the onset of acute esophagitis or chronic dyspnea, thus demonstrating the potential that ML-based approaches have to generate novel data-driven hypotheses in the field.
RESUMO
This work addresses a scoping review of Feature Selection (FS) methods applied to a Lung Cancer dataset to elucidate parameters' relevance when predicting radiotherapy (RT) induced toxicity. Subsetting-based and Ranking-based FS methods were implemented along with 4 advanced classifiers to predict the onset of RT-induced acute esophagitis, cough, pneumonitis and dyspnea. Their prediction performance was measured in terms of the AUC for each model to find the best FS.
Assuntos
Neoplasias Pulmonares , Lesões por Radiação , Radioterapia , Mineração de Dados , Transtornos de Deglutição/etiologia , Dispneia/etiologia , Esofagite/etiologia , Previsões , Humanos , Neoplasias Pulmonares/radioterapia , Pneumonia/etiologia , Radioterapia/efeitos adversosRESUMO
PURPOSE: In 2005, an application for surgical planning called AYRA[Formula: see text] was designed and validated by different surgeons and engineers at the Virgen del Rocío University Hospital, Seville (Spain). However, the segmentation methods included in AYRA and in other surgical planning applications are not able to segment accurately tumors that appear in soft tissue. The aims of this paper are to offer an exhaustive validation of an accurate semiautomatic segmentation tool to delimitate retroperitoneal tumors from CT images and to aid physicians in planning both radiotherapy doses and surgery. METHODS: A panel of 6 experts manually segmented 11 cases of tumors, and the segmentation results were compared exhaustively with: the results provided by a surgical planning tool (AYRA), the segmentations obtained using a radiotherapy treatment planning system (Pinnacle[Formula: see text]), the segmentation results obtained by a group of experts in the delimitation of retroperitoneal tumors and the segmentation results using the algorithm under validation. RESULTS: 11 cases of retroperitoneal tumors were tested. The proposed algorithm provided accurate results regarding the segmentation of the tumor. Moreover, the algorithm requires minimal computational time-an average of 90.5% less than that required when manually contouring the same tumor. CONCLUSION: A method developed for the semiautomatic selection of retroperitoneal tumor has been validated in depth. AYRA, as well as other surgical and radiotherapy planning tools, could be greatly improved by including this algorithm.
Assuntos
Algoritmos , Imageamento Tridimensional/métodos , Planejamento da Radioterapia Assistida por Computador/métodos , Neoplasias Retroperitoneais/diagnóstico , Adolescente , Adulto , Humanos , Masculino , Neoplasias Retroperitoneais/terapia , Tomografia Computadorizada por Raios X , Adulto JovemRESUMO
AIMS AND BACKGROUND: The objective of this study was to assess the influence of ethnicity on toxicity in patients treated with dynamic arc radiation therapy (ART) for prostate cancer (PC). METHODS: From June 2006 to May 2012, 162 cT1-T3 cN0 cM0 PC patients were treated with ART (primary diagnosis, n = 125; post-prostatectomy/brachytherapy biochemical recurrence, n = 26; adjuvant post-prostatectomy, n = 11) at 2 institutions. Forty-five patients were Latin Americans and 117 were Europeans. The dose prescribed to the prostate ranged between 68 Gy and 81 Gy. RESULTS: The median age was 69 years (range 43-87 years). The median follow-up was 18 months (range 2-74 months). Overall, only 3 patients died, none due to a cancer-related cause. Biochemical recurrence was seen in 7 patients. The rates of acute grade 2 gastrointestinal (GI) and genitourinary (GU) toxicities were 19.7% and 17%, respectively. Only 1 patient experienced acute grade 3 GI toxicity, whereas 11 patients (6.7%) experienced acute grade 3 GU toxicity. Multivariate analysis showed that undergoing whole pelvic lymph node irradiation was associated with a higher grade of acute GI toxicity (OR: 3.46; p = 0.003). In addition, older age was marginally associated with a higher grade of acute GI toxicity (OR: 2.10; p = 0.074). Finally, ethnicity was associated with acute GU toxicity: Europeans had lower-grade toxicity (OR: 0.27; p = 0.001). CONCLUSIONS: Our findings suggest an ethnic difference in GU toxicity for PC patients treated with ART. In addition, we found that ART is associated with a very low risk of severe toxicity and a low recurrence rate.
Assuntos
Braquiterapia/efeitos adversos , Trato Gastrointestinal/efeitos da radiação , Hispânico ou Latino/estatística & dados numéricos , Prostatectomia , Neoplasias da Próstata/etnologia , Neoplasias da Próstata/radioterapia , Lesões por Radiação/etnologia , Lesões por Radiação/etiologia , Radioterapia Adjuvante/efeitos adversos , Sistema Urogenital/efeitos da radiação , População Branca/estatística & dados numéricos , Adulto , Idoso , Idoso de 80 Anos ou mais , Biomarcadores Tumorais/sangue , Seguimentos , Humanos , Linfonodos/efeitos da radiação , Metástase Linfática/radioterapia , Masculino , Pessoa de Meia-Idade , Gradação de Tumores , Razão de Chances , Antígeno Prostático Específico/sangue , Neoplasias da Próstata/sangue , Neoplasias da Próstata/patologia , Neoplasias da Próstata/cirurgia , Radioterapia Adjuvante/métodos , Fatores de RiscoRESUMO
AIM: This study evaluates the acute toxicity outcome in patients treated with RapidArc for localized prostate cancer. BACKGROUND: Modern technologies allow the delivery of high doses to the prostate while lowering the dose to the neighbouring organs at risk. Whether this dosimetric advantage translates into clinical benefit is not well known. MATERIALS AND METHODS: Between December 2009 and May 2012, 45 patients with primary prostate adenocarcinoma were treated using RapidArc. All patients received 1.8 Gy per fraction, the median dose to the prostate gland, seminal vesicles, pelvic lymph nodes and surgical bed was 80 Gy (range, 77.4-81 Gy), 50.4 Gy, 50.4 Gy and 77.4 Gy (range, 75.6-79.2 Gy), respectively. RESULTS: The time between the last session and the last treatment follow up was a median of 10 months (range, 3-24 months). The incidence of grade 3 acute gastrointestinal (GI) and genitourinary (GU) toxicity was 2.2% and 15.5%, respectively. Grade 2 acute GI and GU toxicity occurred in 30% and 27% of patients, respectively. No grade 4 acute GI and GU toxicity were observed. Older patients (>median) or patients with V60 higher than 35% had significantly higher rates of grade ≥2 acute GI toxicity compared with the younger ones. CONCLUSIONS: RapidArc in the treatment of localized prostate cancer is tolerated well with no Grade >3 GI and GU toxicities. Older patients or patients with higher V60 had significantly higher rates of grade ≥2 acute GI toxicity. Further research is necessary to assess definitive late toxicity and tumour control outcome.
RESUMO
BACKGROUND: We report the clinical results and prognostic factors of image-guided radiation therapy (RT) with helical tomotherapy (HT) for localized and recurrent prostate cancer (PC). PATIENTS AND METHODS: We evaluated 70 patients with PC (primary diagnosis, n = 48; adjuvant, n = 5; salvage, n = 17) treated with HT from May 2006 through January 2011. The dose prescribed to the prostate/surgical bed ranged between 60 and 78 Gy. Potential risk factors for genitourinary (GU) and gastrointestinal (GI) toxicity were assessed. RESULTS: The median age was 68 years (range 51-87 years). The median follow-up was 37 months (range 3-74 months). The rates of acute grade 2 GI and GU toxicities were 10 and 13%, respectively. Only 1 patient experienced acute grade 3 GU toxicity. The rates of late grade ≥ 2 GI and GU toxicities were 1% each. Multivariate analysis showed an association between rectum mean dose > median (39 Gy) and bladder median dose > median (46 Gy) with a higher grade of acute GI (p = 0.017) and GU (p = 0.019) toxicity, respectively. Additionally, older age was associated with late GU toxicity (p = 0.026). CONCLUSION: Toxicity with HT is low and is associated with higher median/mean doses in organs at risk as well as with older age. A prospective validation would be necessary to confirm these results.
Assuntos
Gastroenteropatias/etiologia , Doenças Urogenitais Masculinas/etiologia , Neoplasias da Próstata/radioterapia , Lesões por Radiação/etiologia , Radioterapia Guiada por Imagem/métodos , Radioterapia de Intensidade Modulada/efeitos adversos , Radioterapia de Intensidade Modulada/métodos , Idoso , Idoso de 80 Anos ou mais , Gastroenteropatias/prevenção & controle , Humanos , Masculino , Doenças Urogenitais Masculinas/prevenção & controle , Pessoa de Meia-Idade , Neoplasias da Próstata/diagnóstico por imagem , Lesões por Radiação/prevenção & controle , Proteção Radiológica/métodos , Dosagem Radioterapêutica , Radioterapia Guiada por Imagem/efeitos adversos , Estudos Retrospectivos , Tomografia Computadorizada Espiral/métodos , Resultado do TratamentoRESUMO
PURPOSE: To investigate the extent of change in pulmonary function over time after definitive radiotherapy for non-small-cell lung cancer (NSCLC) with modern techniques and to identify predictors of changes in pulmonary function according to patient, tumor, and treatment characteristics. PATIENTS AND METHODS: We analyzed 250 patients who had received ≥ 60 Gy radio(chemo)therapy for primary NSCLC in 1998-2010 and had undergone pulmonary function tests before and within 1 year after treatment. Ninety-three patients were treated with three-dimensional conformal radiotherapy, 97 with intensity-modulated radiotherapy, and 60 with proton beam therapy. Postradiation pulmonary function test values were evaluated among individual patients compared with the same patient's preradiation value at the following time intervals: 0-4 (T1), 5-8 (T2), and 9-12 (T3) months. RESULTS: Lung diffusing capacity for carbon monoxide (DLCO) was reduced in the majority of patients along the three time periods after radiation, whereas the forced expiratory volume in 1 s per unit of vital capacity (FEV1/VC) showed an increase and decrease after radiation in a similar percentage of patients. There were baseline differences (stage, radiotherapy dose, concurrent chemotherapy) among the radiation technology groups. On multivariate analysis, the following features were associated with larger posttreatment declines in DLCO: pretreatment DLCO, gross tumor volume, lung and heart dosimetric data, and total radiation dose. Only pretreatment DLCO was associated with larger posttreatment declines in FEV1/VC. CONCLUSIONS: Lung diffusing capacity for carbon monoxide is reduced in the majority of patients after radiotherapy with modern techniques. Multiple factors, including gross tumor volume, preradiation lung function, and dosimetric parameters, are associated with the DLCO decline. Prospective studies are needed to better understand whether new radiation technology, such as proton beam therapy or intensity-modulated radiotherapy, may decrease the pulmonary impairment through greater lung sparing.
Assuntos
Carcinoma Pulmonar de Células não Pequenas/radioterapia , Neoplasias Pulmonares/radioterapia , Pulmão/efeitos da radiação , Terapia com Prótons , Capacidade de Difusão Pulmonar/efeitos da radiação , Radioterapia Conformacional/métodos , Adulto , Idoso , Idoso de 80 Anos ou mais , Algoritmos , Análise de Variância , Monóxido de Carbono/metabolismo , Carcinoma Pulmonar de Células não Pequenas/fisiopatologia , Feminino , Volume Expiratório Forçado/fisiologia , Humanos , Pulmão/fisiopatologia , Neoplasias Pulmonares/fisiopatologia , Masculino , Pessoa de Meia-Idade , Capacidade de Difusão Pulmonar/fisiologia , Radioterapia de Intensidade Modulada , Testes de Função Respiratória , Estudos Retrospectivos , Capacidade Vital/fisiologiaRESUMO
UNLABELLED: We evaluated potential associations between maximum standardized uptake value (SUV(max)) on (18)F-FDG PET before and after radiation therapy (RT) and survival outcomes for patients with locally advanced non-small cell lung cancer. METHODS: Patients with stage III non-small cell lung cancer (n = 49) who had undergone (18)F-FDG PET at the M.D. Anderson Cancer Center both before and up to 3.5 mo after undergoing radiochemotherapy were studied; exclusion criteria were patients with a history of thoracic surgery, RT, or other cancer or those who had received a total radiation dose less than 60 Gy. We assessed associations between overall survival (OS) or disease-free survival (DFS) and post-RT SUV(max) and the extent of decrease in SUV(max) in the primary tumor (PT) and regional lymph nodes (LNs). SUV(max) was assessed as a continuous variable by Cox proportional hazards regression analysis. RESULTS: Univariate and multivariate analyses showed that having a high post-RT SUV(max) (either PT or LNs) was associated with a higher risk of death (univariate analyses: hazard ratio [HR] for PT SUV(max), 1.27, P < 0.0001; HR for LN SUV(max), 1.32, P = 0.004) and disease recurrence (univariate analyses: HR for PT SUV(max), 1.16, P = 0.004; HR for LN SUV(max), 1.32, P = 0.001). Moreover, after definitive RT, the greater the decrease in SUV(max) in the lesion that had the highest SUV(max) at diagnosis, the longer the OS (HR, 0.06; P = 0.002), DFS (HR, 0.03; P = 0.001), local-regional control (HR, 0.04; P = 0.002), and distant metastasis-free survival (HR, 0.07; P = 0.028). CONCLUSION: The post-RT SUV(max) in both the PT and the LNs was a predictor of survival-specifically, the higher the residual SUV(max) after RT, the poorer the OS and DFS; and the greater the decrease in SUV(max) in the lesion with the highest SUV(max) at diagnosis, the longer the OS and DFS. This information should help to identify patients who are at high risk of recurrence and for whom additional treatments can be designed accordingly.