Search

The prediction of response to breast cancer therapy involves assessing the effectiveness of treatment by comparing biomarker levels before and after treatment. Deep learning (DL) models can provide a non-invasive and early way to evaluate the response to therapy based on medical imaging analysis. We conducted this systematic review to investigate the current DL based methods for predicting breast cancer therapy response using medical imaging. This review included 19 studies based on the PRISMA methodology. Some selected studies personalized the Convolutional Neural Network (CNN) architecture to improve its performance in handling medical images, while others used pre-trained models. The accuracy rates range from 0.73 to 0.90, and the Area Under the Curve (AUC) reaches 0.98. Our study's findings suggest that the performance of these approaches varies depending on various medical imaging modalities, the nature of the DL architecture used, and the fusion of training data sources. However, several challenges related to their explainability and generalizability arise. Therefore, it is necessary to develop larger datasets and broaden the scope of current studies to include multi-center studies. [ABSTRACT FROM AUTHOR]

Abstract Cancer is a top-ranked life-threatening disease with intratumor heterogeneity. Tumor heterogeneity is associated with metastasis, relapse, and therapy resistance. These factors contribute to treatment failure and an unfavorable prognosis. Personalized tumor models faithfully capturing the tumor heterogeneity of individual patients are urgently needed for precision medicine. Advances in stem cell culture have given rise to powerful organoid technology for the generation of in vitro three-dimensional tissues that have been shown to more accurately recapitulate the structures, specific functions, molecular characteristics, genomic alterations, expression profiles, and tumor microenvironment of primary tumors. Tumoroids in vitro serve as an important component of the pipeline for the discovery of potential therapeutic targets and the identification of novel compounds. In this review, we will summarize recent advances in tumoroid cultures as an excellent tool for accurate cancer modeling. Additionally, vascularization and immune microenvironment modeling based on organoid technology will also be described. Furthermore, we will summarize the great potential of tumor organoids in predicting the therapeutic response, investigating resistance-related mechanisms, optimizing treatment strategies, and exploring potential therapies. In addition, the bottlenecks and challenges of current tumoroids will also be discussed in this review.

Cancer is a top-ranked life-threatening disease with intratumor heterogeneity. Tumor heterogeneity is associated with metastasis, relapse, and therapy resistance. These factors contribute to treatment failure and an unfavorable prognosis. Personalized tumor models faithfully capturing the tumor heterogeneity of individual patients are urgently needed for precision medicine. Advances in stem cell culture have given rise to powerful organoid technology for the generation of in vitro three-dimensional tissues that have been shown to more accurately recapitulate the structures, specific functions, molecular characteristics, genomic alterations, expression profiles, and tumor microenvironment of primary tumors. Tumoroids in vitro serve as an important component of the pipeline for the discovery of potential therapeutic targets and the identification of novel compounds. In this review, we will summarize recent advances in tumoroid cultures as an excellent tool for accurate cancer modeling. Additionally, vascularization and immune microenvironment modeling based on organoid technology will also be described. Furthermore, we will summarize the great potential of tumor organoids in predicting the therapeutic response, investigating resistance-related mechanisms, optimizing treatment strategies, and exploring potential therapies. In addition, the bottlenecks and challenges of current tumoroids will also be discussed in this review. [ABSTRACT FROM AUTHOR]

Abstract Purpose To evaluate the applicability and estimate the radiobiological parameters of linear-quadratic Poisson tumour control probability (TCP) model for primary prostate cancer patients for two relevant target structures (prostate gland and GTV). The TCP describes the dose–response of prostate after definitive radiotherapy (RT). Also, to analyse and identify possible significant correlations between clinical and treatment factors such as planned dose to prostate gland, dose to GTV, volume of prostate and mpMRI-GTV based on multivariate logistic regression model. Methods The study included 129 intermediate and high-risk prostate cancer patients (cN0 and cM0), who were treated with image-guided intensity modulated radiotherapy (IMRT) ± androgen deprivation therapy with a median follow-up period of 81.4 months (range 42.0–149.0) months. Tumour control was defined as biochemical relapse free survival according to the Phoenix definition (BRFS). MpMRI-GTV was delineated retrospectively based on a pre-treatment multi-parametric MR imaging (mpMRI), which was co-registered to the planning CT. The clinical treatment planning procedure was based on prostate gland, delineated on CT imaging modality. Furthermore, we also fitted the clinical data to TCP model for the two considered targets for the 5-year follow-up after radiation treatment, where our cohort was composed of a total number of 108 patients, of which 19 were biochemical relapse (BR) patients. Results For the median follow-up period of 81.4 months (range 42.0–149.0) months, our results indicated an appropriate α/β = 1.3 Gy for prostate gland and α/β = 2.9 Gy for mpMRI-GTV. Only for prostate gland, EQD2 and gEUD2Gy were significantly lower in the biochemical relapse (BR) group compared to the biochemical control (BC) group. Fitting results to the linear-quadratic Poisson TCP model for prostate gland and α/β = 1.3 Gy were D 50 = 66.8 Gy with 95% CI [64.6 Gy, 69.0 Gy], and γ = 3.8 with 95% CI [2.6, 5.2]. For mpMRI-GTV and α/β = 2.9 Gy, D 50 was 68.1 Gy with 95% CI [66.1 Gy, 70.0 Gy], and γ = 4.5 with 95% CI [3.0, 6.1]. Finally, for the 5-year follow-up after the radiation treatment, our results for the prostate gland were: D 50 = 64.6 Gy [61.6 Gy, 67.4 Gy], γ = 3.1 [2.0, 4.4], α/β = 2.2 Gy (95% CI was undefined). For the mpMRI-GTV, the optimizer was unable to deliver any reasonable results for the expected clinical D 50 and α/β. The results for the mpMRI-GTV were D 50 = 50.1 Gy [44.6 Gy, 56.0 Gy], γ = 0.8 [0.5, 1.2], α/β = 0.0 Gy (95% CI was undefined). For a follow-up time of 5 years and a fixed α/β = 1.6 Gy, the TCP fitting results for prostate gland were D 50 = 63.9 Gy [60.8 Gy, 67.0 Gy], γ = 2.9 [1.9, 4.1], and for mpMRI-GTV D 50 = 56.3 Gy [51.6 Gy, 61.1 Gy], γ = 1.3 [0.8, 1.9]. Conclusion The linear-quadratic Poisson TCP model was better fit when the prostate gland was considered as responsible target than with mpMRI-GTV. This is compatible with the results of the comparison of the dose distributions among BR and BC groups and with the results achieved with the multivariate logistic model regarding gEUD 2Gy . Probably limitations of mpMRI in defining the GTV explain these results. Another explanation could be the relatively homogeneous dose prescription and the relatively low number of recurrences. The failure to identify any benefit for considering mpMRI-GTV as the target responsible for the clinical response is confirmed when considering a fixed α/β = 1.6 Gy, a fixed follow-up time for biochemical response at 5 years or Gleason score differentiation.

Purpose: To evaluate the applicability and estimate the radiobiological parameters of linear-quadratic Poisson tumour control probability (TCP) model for primary prostate cancer patients for two relevant target structures (prostate gland and GTV). The TCP describes the dose-response of prostate after definitive radiotherapy (RT). Also, to analyse and identify possible significant correlations between clinical and treatment factors such as planned dose to prostate gland, dose to GTV, volume of prostate and mpMRI-GTV based on multivariate logistic regression model.Methods: The study included 129 intermediate and high-risk prostate cancer patients (cN0 and cM0), who were treated with image-guided intensity modulated radiotherapy (IMRT) ± androgen deprivation therapy with a median follow-up period of 81.4 months (range 42.0-149.0) months. Tumour control was defined as biochemical relapse free survival according to the Phoenix definition (BRFS). MpMRI-GTV was delineated retrospectively based on a pre-treatment multi-parametric MR imaging (mpMRI), which was co-registered to the planning CT. The clinical treatment planning procedure was based on prostate gland, delineated on CT imaging modality. Furthermore, we also fitted the clinical data to TCP model for the two considered targets for the 5-year follow-up after radiation treatment, where our cohort was composed of a total number of 108 patients, of which 19 were biochemical relapse (BR) patients.Results: For the median follow-up period of 81.4 months (range 42.0-149.0) months, our results indicated an appropriate α/β = 1.3 Gy for prostate gland and α/β = 2.9 Gy for mpMRI-GTV. Only for prostate gland, EQD2 and gEUD2Gy were significantly lower in the biochemical relapse (BR) group compared to the biochemical control (BC) group. Fitting results to the linear-quadratic Poisson TCP model for prostate gland and α/β = 1.3 Gy were D50 = 66.8 Gy with 95% CI [64.6 Gy, 69.0 Gy], and γ = 3.8 with 95% CI [2.6, 5.2]. For mpMRI-GTV and α/β = 2.9 Gy, D50 was 68.1 Gy with 95% CI [66.1 Gy, 70.0 Gy], and γ = 4.5 with 95% CI [3.0, 6.1]. Finally, for the 5-year follow-up after the radiation treatment, our results for the prostate gland were: D50 = 64.6 Gy [61.6 Gy, 67.4 Gy], γ = 3.1 [2.0, 4.4], α/β = 2.2 Gy (95% CI was undefined). For the mpMRI-GTV, the optimizer was unable to deliver any reasonable results for the expected clinical D50 and α/β. The results for the mpMRI-GTV were D50 = 50.1 Gy [44.6 Gy, 56.0 Gy], γ = 0.8 [0.5, 1.2], α/β = 0.0 Gy (95% CI was undefined). For a follow-up time of 5 years and a fixed α/β = 1.6 Gy, the TCP fitting results for prostate gland were D50 = 63.9 Gy [60.8 Gy, 67.0 Gy], γ = 2.9 [1.9, 4.1], and for mpMRI-GTV D50 = 56.3 Gy [51.6 Gy, 61.1 Gy], γ = 1.3 [0.8, 1.9].Conclusion: The linear-quadratic Poisson TCP model was better fit when the prostate gland was considered as responsible target than with mpMRI-GTV. This is compatible with the results of the comparison of the dose distributions among BR and BC groups and with the results achieved with the multivariate logistic model regarding gEUD2Gy. Probably limitations of mpMRI in defining the GTV explain these results. Another explanation could be the relatively homogeneous dose prescription and the relatively low number of recurrences. The failure to identify any benefit for considering mpMRI-GTV as the target responsible for the clinical response is confirmed when considering a fixed α/β = 1.6 Gy, a fixed follow-up time for biochemical response at 5 years or Gleason score differentiation. [ABSTRACT FROM AUTHOR]

Background: Bladder cancer is a common and highly heterogeneous malignant tumor with a relatively poor prognosis. Thus, personalized treatment strategies for bladder cancer are essential for improving patient outcomes., Materials and Methods: We developed an efficient 3-dimensional in vitro organoid culture system for bladder cancer organoids (BCOs), which maintains the homology with the original patient tumors and the heterogeneity between different individuals. In addition, we constructed chimeric antigen receptor (CAR)-T cells targeting B7H3 and evaluated the antitumor function of CAR-T cells by coculturing them with BCOs., Results: The BCOs closely resembled the characteristics of human tumors and were used to test individual sensitivity to platinum-based drugs and olaparib therapy. Coculture with CAR-T cells demonstrated specific antigen recognition and immune activation, indicating their potential in immunotherapy., Conclusions: Our study highlights the potential of BCOs to facilitate the development of personalized medicine for bladder cancer and improve the efficiency of drug discovery for bladder cancer therapy., Competing Interests: DT is the managing editor of Current Urology. This article was accepted after a normal external review. No conflict of interest has been declared by the other authors., (Copyright © 2023 The Authors. Published by Wolters Kluwer Health, Inc.)

Background: Core needle biopsy plays a crucial role as diagnostic tool for BC. Both Ki67 and likely tumorinfiltrating lymphocytes (TILs) in the near future are determining the kind of systemic therapy. The role of TILs in BC is still an issue for clinical research, albeit preliminary results of neoadjuvant and adjuvant clinical studies already now highlight the crucial impact of TILs on therapy response and survival. Methods: Evaluation of related publications (pubmed) and meeting abstracts (ASCO, SABCS). Results: The monoclonal antibody Ki67 recognizing a nuclear antigene in proliferating cells is a positive marker of therapy response and superior survival. Endocrine responsive tumors of low proliferation (Ki67 < 14%/11%) respond to tamoxifen, in contrast postmenopausal tumors with higher proliferation respond better to aromatase-inhibitors. Pathological complete response (pCR)-rates increase in tumors with higher proliferation (Ki67 > 19%) vs. tumors with lower proliferation after neoadjuvant chemotherapy (NAC). pCR-rates of up to 60% can be seen in TNBC and HR-, HER2+BC, lower pCR-rates, however, in HR+, HER2- BC. Increased stromal TILs are found in 30% of TNBC and in 19% of HR-, HER2+BC. The percentage of TILs is a significant independent parameter for pCR after NAC. Lymphocyte-predominant BC (LPBC) respond with higher pCR-rates than non-LPBC or tumors without any TILs. Increased TILs in TN and HR-, HER2+ subtypes predict benefit from addition of carboplatin to NAC. TILs are also associated with improved DFS and OS among patients with TNBC and HR-, HER2+ BC. Conversly and interestingly increased TILs in patients with HR+, HER2-(luminal) BC are associated with a 10% higher risk of death per 10% increase of TILs. Interactions between immune system and cancer are complex. The cancer-immunity cycle characterizes these interactions. BC subtypes with higher number of mutations such as TNBC and HR-, HER2+BC are considered to provide a raising number of tumor-associated antigens, thereby capable to build up a higher endogenous immune response. TILs may serve as surrogate marker of both an existing endogenous immune response and the probability to respond to cancer immune therapies. As cancer co-opt immune checkpoint-pathways as a major mechanism of immune resistance, in particular, against cytotoxic T-cells, blockades of checkpoint-pathways by antibodies are one of the goals of the current cancer immunotherapy studies. Therapy studies with antigene-based strategies (vaccines) and antibodies against the immune checkpoints PD-1 and CTLA-4 and their inhibitory pathways in order to enhance cytotoxic T-cell activities against cancer cells with or without chemotherapy are underway. Conclusions: It can be suggested that the use of multigene expression testing will increase in order to select more clearly primary HR+, HER2- BC patients with intermediate recurrence risk who likely may benefit from chemotherapy. Furthermore Ki67 and the multigene expression test Oncotype DX can act as dynamic markers to avoid cytostatic overtreatment and endocrine undertreatment. A data-derived optimal Ki67 cut point for pCR and DFS as well as OS is currently not feasible. The integration of stromal TILs into the immunohistopathological report after their evaluation has been standardized is likely helpful to determine patients who profit by additional carboplatin chemotherapy. Oncologists need an enlarged information about the tumormicroenvironment in future. The preliminary results of current BC immunotherapy studies are encouraging. [ABSTRACT FROM AUTHOR]

Our purpose was to evaluate and compare the clinical utility of simultaneously obtained quantitative 18 F-FDG PET and diffusion-weighted MRI datasets for predicting the histopathologic response of soft-tissue sarcoma (STS) to neoadjuvant isolated limb perfusion (ILP). Methods: In total, 37 patients with a confirmed STS of the extremities underwent 18 F-FDG PET/MRI before and after ILP with melphalan and tumor necrosis factor-α. For each patient, the maximum tumor size, metabolic activity (SUV), and diffusion restriction (apparent diffusion coefficient, ADC) were determined in pre- and posttherapeutic examinations, and percentage changes during treatment were calculated. Mann-Whitney U testing and receiver-operating-characteristic analysis were used to compare the results of the different quantitative parameters to predict the histopathologic response to therapy. Results from histopathologic analysis after tumor resection served as the reference standard, and patients were defined as responders or nonresponders based on the grading scale by Salzer-Kuntschik. Results: Histopathologic analysis categorized 22 (59%) patients as responders (grades I-III) and 15 (41%) as nonresponders (grades IV-VI). Under treatment, tumors in responders showed a mean reduction in size (-9.7%) and metabolic activity (SUV peak , -51.9%; SUV mean , -43.8%), as well as an increase of the ADC values (ADC min , +29.4%; ADC mean , +32.8%). The percentage changes in nonresponders were -6.2% in tumor size, -17.3% in SUV peak , -13.9% in SUV mean , +15.3% in ADC min , and +14.6% in ADC mean Changes in SUV and ADC mean significantly differed between responders and nonresponders (<0.01), whereas differences in tumor size and ADC min did not (>0.05). The corresponding AUCs were 0.63 for tumor size, 0.87 for SUV peak , 0.82 for SUV mean , 0.63 for ADC min , 0.84 for ADC mean , and 0.89 for ratio of ADC mean to SUV peak Conclusion: PET- and MRI-derived quantitative parameters (SUV and ADC mean ) and their combination performed well in predicting the histopathologic therapy response of STS to neoadjuvant ILP. Therefore, integrated PET/MRI could serve as a valuable tool for pretherapeutic assessment as well as monitoring of neoadjuvant treatment strategies of STS., (© 2021 by the Society of Nuclear Medicine and Molecular Imaging.)

We present a probabilistic approach to identify patients with primary and secondary hepatic malignancies as responders or nonresponders to yttrium-90 radioembolization therapy. Recent advances in computer-aided detection have decreased false-negative and false-positive rates of perceived abnormalities; however, there is limited research in using similar concepts to predict treatment response. Our approach is driven by the goal of precision medicine to determine pretherapy fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography imaging parameters to facilitate the identification of patients who would benefit most from yttrium-90 radioembolization therapy, while avoiding complex and costly procedures for those who would not. Our algorithm seeks to predict a patient's response by discovering common co-occurring image patterns in the lesions of baseline fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography scans by extracting invariant shape and texture features. The extracted imaging features were represented as a distribution of each subject based on the bag-of-feature paradigm. The distribution was applied in a multinomial naive Bayes classifier to predict whether a patient would be a responder or nonresponder to yttrium-90 radioembolization therapy based on the imaging features of a pretherapy fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography scan. Comprehensive published criteria were used to determine lesion-based clinical treatment response based on fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography imaging findings. Our results show that the model is able to predict a patient with liver cancer as a responder or nonresponder to yttrium-90 radioembolization therapy with a sensitivity of 0.791 using extracted invariant imaging features from the pretherapy fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography test. The sensitivity increased to 0.821 when combining extracted invariant image features with variable features of tumor volume.