Onco Preventer

Category: Uncategorized

  • Fostering Multidisciplinary Collaboration in Brain Tumor Management

    Optimal management of brain tumors necessitates a cohesive and collaborative approach involving neurosurgeons, medical oncologists, and radiation oncologists, according to Jennifer Moliterno, MD, FAANS. The inherent complexities in treating patients with diverse tumor types demand a nuanced understanding of how to strategically integrate surgery, radiation therapy, and systemic therapies to maximize patient outcomes.

    Moliterno, chief of Neurosurgical Oncology; clinical director of the Chênevert Family Brain Tumor Center; director of the Susan Beris, MD, Neurosurgical Oncology Program; surgical director of the Facial Pain and Spasm Program; and director of Neurosurgical Oncology Fellowship, Neurosurgical Oncology at Yale School of Medicine, spoke with CancerNetwork® about invaluable insights into fostering effective collaboration among these specialties.

    She addressed the critical considerations in determining the ideal timing and sequencing of different treatment modalities for various brain tumors. The unique characteristics of each tumor type, including its histology, location, and molecular profile, influence the collaborative decision-making process.

    Moliterno further delved into the specific nuances of coordinating surgical resection with adjuvant therapies. She discussed how factors such as the extent of resection, postoperative neurological status, and the anticipated benefits and risks of subsequent radiation and systemic treatments are carefully weighed in multidisciplinary tumor boards.

    Transcript:

    One of the unique things about [our institution], in addition to our expertise for maximizing and safely improving the extent of resection, is that all our tumors undergo what we call whole-exome sequencing so we can understand the tumor from a molecular genetic perspective. We’re unique with that comprehensive next-generation sequencing technique that we do for all of our patients who have surgery here. We can then use that information in our multidisciplinary group. I lead the multidisciplinary brain tumor board, as well as the precision brain tumor board, where, after surgery, we can then use those molecular data for each person to personalize more precise treatment. A lot of times, there’s the standard of care for glioma; for instance, with radiation and with chemotherapy. Sometimes, there can be opportunities for more personalized treatment. That’s where we then coordinate with our neuro-oncologists and our radiation oncologists. We do have a seamless team that allows us to then transition the patient’s care from the surgeon, then to have the [neuro-oncologist] and medical doctor as the quarterback for the patient. Then, [we can] incorporate the radiation doctors as well.

  • Fostering Multidisciplinary Collaboration in Brain Tumor Management

    Managing brain tumors is a remarkably complex endeavor that requires not only expert knowledge but also a well-orchestrated team effort. Dr. Jennifer Moliterno, a leading figure in neurosurgical oncology, underscores the importance of a collaborative approach involving neurosurgeons, medical oncologists, and radiation oncologists to optimize treatment outcomes. Brain tumors are no monolith; their treatment success hinges on understanding their diverse histologies, precise locations, and unique molecular profiles. This multifaceted nature means that a personalized and dynamic method combining surgery, radiation, and systemic therapies is essential to provide the best care.

    At the forefront of this approach at Yale School of Medicine’s Neurosurgical Oncology division, Dr. Moliterno leads several key programs focused on pioneering treatment and patient care innovation. Beyond just her surgical skills, she champions the integration of cutting-edge techniques such as whole-exome sequencing, offering a molecular-level insight into each tumor. This genomic data reveals insights that transcend traditional imaging and pathology, enabling the team to tailor treatments that go beyond the standard care. It is this precision medicine philosophy, utilizing personalized genetic information, that exemplifies the next generation of neuro-oncology. By leveraging these molecular clues, the multidisciplinary brain tumor and precision brain tumor boards at Yale can recommend therapies tailored specifically to the genetic makeup of a patient’s tumor.

    The collaboration doesn’t come together haphazardly but through carefully structured tumor board discussions where specialists deliberate over each patient’s unique case. Dr. Moliterno highlights how decisions on the timing and sequencing of surgery, radiation, and systemic therapy require a delicate balance of factors. They consider the extent of tumor resection possible—aiming to remove as much of the tumor as safely feasible—while also evaluating the patient’s postoperative neurological function. These decisions also involve weighing the risks and benefits of adjuvant therapies such as radiation and chemotherapy to minimize harm while maximizing treatment efficacy. Such boards are a prime example of evidence-based medicine meeting personalized care, where the collective wisdom of multiple specialists improves patient outcomes.

    One particularly fantastic aspect of Dr. Moliterno’s institution is the seamless transition of patients between specialists. After surgical resection, the neuro-oncologist assumes the role of “quarterback,” coordinating the patient’s ongoing care plan, including the integration of systemic therapies. Meanwhile, radiation oncologists join the team strategy, ensuring treatments are synchronized and optimized. This smooth continuum of care exemplifies what many hospitals strive for but few achieve at this level of precision and coordination. It removes the potential pitfalls of fragmented care and fosters a patient-centered environment, where everybody is aligned on a unified treatment strategy.

    In addition to the collaborative clinical approach, the use of next-generation sequencing technologies at Yale represents a remarkable leap forward in brain tumor care. Whole-exome sequencing—examining all protein-coding genes in the human genome—provides an unprecedented depth of tumor characterization. This data guides innovative therapeutic strategies, exploring avenues beyond conventional radiation and chemotherapy. For instance, some tumor genetic profiles may indicate susceptibility to novel targeted treatments or immunotherapies, opening doors to clinical trials or off-label uses supported by molecular rationale. As the field of neuro-oncology continues to evolve, embracing such comprehensive molecular analyses not only personalizes therapy but also fuels research that may in time transform the prognosis for what have been historically devastating conditions.

    #BrainTumor #Neurosurgery #PrecisionMedicine #OncologyCollaboration #CancerTreatment #MolecularGenetics #NeuroOncology

    Source

  • US Approves First At-Home Cervical Cancer Screening Device

    The U.S. Food and Drug Administration (FDA) has recently approved an innovative at-home cervical cancer screening tool, marking a significant advancement in women's health care. This new device offers an alternative to the traditional Pap smear, a procedure usually performed by a gynecologist during a clinical visit. By enabling women to collect samples from the comfort and privacy of their own homes, this screening method is expected to enhance accessibility and convenience for cervical cancer screening, ultimately aiming to increase screening rates and improve early detection outcomes. Cervical cancer remains a serious health concern for women worldwide, with human papillomavirus (HPV) recognized as a major causative factor. Regular screening is vital for early identification and treatment of precancerous changes, significantly reducing the incidence and mortality rates associated with cervical cancer. Traditionally, women have relied on in-person Pap smears to detect abnormal cervical cells. However, barriers such as limited access to healthcare providers, discomfort with pelvic exams, time constraints, and apprehension have led to lower screening rates in some populations. The newly approved at-home screening tool addresses many of these challenges by allowing women to collect cervical cell samples independently at home. After collection, the samples are sent to a laboratory for analysis, where they are tested for the presence of HPV DNA, indicating potential risk for cervical cancer development. This process not only streamlines the screening procedure but also empowers women to take proactive control over their health. Healthcare experts anticipate that this advancement could revolutionize cervical cancer prevention strategies. By removing logistical and psychological barriers to screening, more women, especially those in underserved or remote communities, may participate in regular testing. Early detection through such screening tools enhances treatment options and improves survival rates. Dr. Emily Johnson, a leading gynecologic oncologist, emphasizes the importance of the FDA's approval: "The ability to screen for cervical cancer at home is a game-changer. It breaks down many obstacles women face when trying to access traditional screening methods. This could lead to a significant increase in early diagnosis and ultimately save lives." While the device offers numerous advantages, experts also advise that it should complement, not completely replace, traditional cervical cancer screening methods. In cases where abnormal results are detected, follow-up examinations by healthcare providers remain crucial for accurate diagnosis and appropriate treatment planning. Public health organizations are planning to launch awareness campaigns to inform women about this new option and how to use it effectively. Educational initiatives will focus on proper sample collection techniques, understanding results, and the importance of regular screening intervals. The FDA's approval of this at-home cervical cancer screening tool signifies a promising shift towards more patient-centered healthcare solutions. As technology and medical innovation evolve, such advancements pave the way for improved health outcomes and greater equity in healthcare access worldwide.

  • Predicting relative efficacy of anthracyclines and taxanes in breast cancer neoadjuvant AC-T chemotherapy using longitudinal MRI radiomic model

    1 Introduction

    Neoadjuvant chemotherapy (NAC) has become a standard treatment for breast cancer, primarily aiming at reducing the tumor stage and enhancing the feasibility of breast-conserving surgery (1, 2). One of the most classic NAC treatment regimens is 4-cycle anthracycline and cyclophosphamide (AC) treatment followed by 4-cycle taxane (T) treatment (3–5). During the NAC treatment process, MRI can evaluate the changes in tumor volume at different stages, providing a basis for assessing the treatment efficacy and adjusting the treatment plan in a timely manner (6–9).

    During the clinical treatment process, significant individual variability in patients’ responses to the same NAC regimen is observed. The variability in response occurs not only among different patients but also within the same patient at different treatment stages. In our previous study, we found that the change rates of tumor volume on MRI during the AC and T treatment stages can respectively reflect the treatment efficacy of AC and T (10). Some tumors shrink more rapidly during the AC treatment stage, while others shrink more rapidly during the T treatment stage. This indicates that stratifying patients based on the relative efficacy of AC and T and individually prolonging the cycle of AC or T treatment may improve the overall treatment efficacy of NAC. However, complete tumor volume change rate data is only available after the completion of NAC treatment, at which point it is no longer possible to adjust the regimen. Therefore, during the mid-term of NAC treatment, predicting the relative efficacy of AC and T can provide a basis for timely adjustment of the NAC treatment plan.

    MRI-based radiomic models have been successfully used to predict the treatment outcome of breast cancer NAC. Radiomics extracts abundant quantitative information from breast MRI imaging, revealing the association between tumor imaging features and clinical outcome, which is crucial for optimizing treatment regimens (11–13). However, currently, no radiomic model has been developed to predict the relative treatment efficacy of AC and T in NAC.

    To achieve this goal, in this study, we included breast cancer patients at our center who received NAC with AC followed by T (AC-T). These patients underwent MRI examinations before, during, and after NAC treatment. We used the ratio of the change in tumor volume during the AC and T treatment stages to measure the relative treatment efficacy of AC and T, and developed a radiomic model based on MRI data before and during the NAC treatment to classify patients, thereby stratifying patients who may benefit from prolonged AC treatment during the NAC treatment process.

    2 Methods

    2.1 Study participants

    This study received approval from the institutional review board of our hospital. Given its retrospective design, written informed consent was waived. Figure 1 shows a flowchart of patient recruitment. Female patients diagnosed with breast cancer at the First Affiliated Hospital with Nanjing Medical University between April 2016 and March 2023 were included retrospectively. The eligibility criteria were as follows: operable invasive breast cancer confirmed by core needle biopsy; Estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and the nuclear protein Ki67 were evaluated by immunohistochemical (IHC) staining; Completing four cycles of AC treatment followed by four cycles of T treatment; MRI data obtained before NAC treatment (pre-NAC), before the fifth cycle (mid-NAC), as well as before surgery (post-NAC).

    Figure 1

    The exclusion criteria were as follows: inadequate MRI quality or lack of MRI data; inability to complete the full cycle of NAC due to chemotherapy-related side effects; previous chemotherapy or targeted therapy; distant metastatic lesions; unmeasurable tumor without discernible boundary; tumor scattered or discontinuous after NAC.

    Clinical data were collected for each patient, including gender, age, menstrual status, clinical T stage, and clinical N stage. Patients were randomly stratified into a training set and a test set in a 7:3 ratio. The study design and workflow are depicted in Figure 2.

    Figure 2

    2.2 Treatment strategies and pathological assessment

    All patients started treatment with four cycles of anthracycline (Pharmorubicin 90 mg/m²) and cyclophosphamide (Endoxan 600 mg/m²), each administered on day 1, every 2 or 3 weeks. All patients subsequently underwent four cycles of T treatment (either nanoparticle albumin-bound paclitaxel [Abraxane 260 mg/m²], solvent-based paclitaxel [Taxol 175 mg/m²], docetaxel [Taxotere 75 mg/m²] or liposomal paclitaxel [Lipusu 175 mg/m²]), administered on day 1, every 2 or 3 weeks. HER2 positive patients received targeted therapy with trastuzumab (Herceptin 6 mg/kg with an 8 mg/kg loading dose) in addition to the T treatment.

    ER, PR, HER2, and Ki67 status were determined using IHC. Tumors were classified as ER/PR-positive if they showed ≥1% nuclear-stained cells. HER2 status was assessed as negative (HER2-) with IHC grades of 0 and 1+, and positive (HER2+) with an IHC grade of 3+. For tumors with an IHC grade of 2+, HER2 gene amplification was determined by fluorescence in situ hybridization (FISH). Ki-67 expression was evaluated using a cutoff index of 30%; expressions below 30% were considered low, while those ≥30% were considered high.

    2.3 MRI acquisition

    MRI scans were conducted in the prone position using either a 1.5 Tesla scanner (MAGNETOM Aera XJ, Siemens) or a 3.0 Tesla scanner (MAGNETOM Trio, Siemens) at three key points: pre-NAC, mid-NAC and post-NAC. The protocol included at least diffusion weighted imaging (DWI) and fat-suppressed dynamic contrast-enhanced (DCE) sequence.

    For MAGNETOM Trio, the imaging protocol included: axial DWI (repetition time [TR]/echo time [TE], 5200 ms/65 ms; matrix, 220 × 110; field of view, 323 mm × 161 mm; thickness, 5 mm) and DCE sequence (TR/TE, 4.23 ms/1.57 ms; matrix, 448 × 448; field of view, 340 mm × 340 mm; thickness, 1 mm). For MAGNETOM Aera, the imaging protocol included: axial DWI (TR/TE, 7500ms/64 ms; matrix, 180 × 84; field of view, 350 mm × 163 mm; thickness, 5 mm) and DCE sequence (TR/TE, 3.90 ms/1.66 ms; matrix, 320 × 320; field of view, 360 mm × 360 mm; thickness, 1.5 mm). Details on specific imaging parameters are available in our previous study (14).

    The DCE sequence was initially acquired prior to contrast agent administration. Gadolinium-DTPA (Magnevist; Bayer Healthcare) was then administered at a dosage of 0.1 mmol/kg with an infusion rate of 3 mL/s, followed by a 20 mL saline flush. Subsequently, the DCE sequence was repeated five times. Apparent diffusion coefficient (ADC) maps were generated from DWI images using two b values.

    2.4 Image segmentation and feature extraction

    For each patient, image segmentation was performed separately on MRI images acquired at pre-NAC, mid-NAC, and post-NAC time points. Using 3D Slicer software (version 5.2.2, www.slicer.org), intratumoral regions were manually delineated slice-by-slice on the second post-contrast phase image of the DCE sequence. The intratumoral regions were then isotropically expanded by 5 mm in three dimensions to obtain the peritumoral regions using the SimpleITK package (version 2.2.1) in Python 3.9.13. A radiologist with five years of experience in breast imaging performed the segmentation for all cases. The radiologist was blinded to the clinicopathological data and the relative efficacy information. Radiomic features were extracted from both the original and filtered images using PyRadiomics (version 3.0.1, https://github.com/Radiomics/pyradiomics). Filtered images were generated using the Laplacian of Gaussian operator and wavelet filters. The Laplacian of Gaussian filter was applied with kernel sizes of 1, 2, 3, 4, and 5 for the DCE sequences and 2, 3, 4, and 5 for the ADC maps. The wavelet filter decomposed each dimension into eight levels. Radiomic features included various categories, including first-order (intensity-based histogram), shape-based, gray-level co-occurrence matrix (GLCM), gray-level size zone matrix (GLSZM), gray-level run length matrix (GLRLM), and gray-level dependence matrix (GLDM). Prior to feature extraction, the intensity distribution of images was normalized. Voxel size was resampled to achieve isotropic voxels of 1.0 mm × 1.0 mm × 1.0 mm for the DCE images and in-plane isotropic voxels of 2.0 mm × 2.0 mm for the ADC maps with the sitkBSpline interpolator. Additionally, voxel intensity values were discretized with fixed bin widths set at 5 for the DCE sequence and 25 for the ADC maps. A total of 1,218 features were extracted from each region for the second post-contrast phase image of the DCE sequence and 1,132 features for the ADC maps. Considering the two types of regions (intratumoral regions and peritumoral regions), each patient’s imaging data from pre-NAC and mid-NAC images contributed a total of 9,400 features from both the DCE and ADC images. To capture longitudinal tumor changes, delta radiomic features were calculated as the differences between the radiomic feature values from the pre-NAC and mid-NAC images. As a result, each patient yielded a total of 14,100 radiomic features.

    2.5 Characterization of relative efficacy

    Tumor volume was derived from Mesh Volume feature within the shape category of radiomic features. Mesh Volume feature calculates the volume of a structure by estimating the volume enclosed within a 3D mesh model. The efficacy of the four cycles of AC treatment was determined by calculating the relative net reduction in tumor volume from pre-NAC to mid-NAC (δAC). Similarly, the efficacy of the four cycles of T treatment was calculated using the relative net reduction in tumor volume from mid-NAC to post-NAC (δT). Relative efficacy is defined by the ratio of δAC and δT. If the ratio of δAC and δT is greater than 1, the patient is considered more sensitive to AC treatment stage (AC is superior to T). Otherwise, the patients is considered more sensitive to T treatment stage (T is superior to AC). Figure 3 and Table 1 show examples of two cases exhibiting different types of relative efficacy.

    Figure 3

    Table 1

    2.6 Radiomic feature selection

    To ensure inter-observer consistency, a second radiologist with 5 years of experience re-segmented a randomly selected subset of 50 cases. Both radiologists were blinded to the clinicopathological data and the relative efficacy information. To assess inter-observer reproducibility, the intraclass correlation coefficient (ICC) was calculated for each radiomic feature. Only features demonstrating satisfactory inter-observer reproducibility, defined as an ICC of 0.80 or higher, were retained in the model.

    For each radiomic model, the following feature selection process was carried out in four steps (1): Features with a variance greater than 1.0 were selected by variance threshold (2); The Mann-Whitney U test was applied to select features associated with NAC treatment relative efficacy (3); Feature importance was ranked using a random forest model, and the top 100 most important features were selected (4); Least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation was used to select features with non-zero coefficients.

    2.7 Radiomic model construction and validation

    First, A total of 9 basic radiomic models were constructed. These models were built using the intratumoral, peritumoral, and intratumoral + peritumoral image features of DCE images; the intratumoral, peritumoral, and intratumoral + peritumoral image features of ADC images; as well as the intratumoral, peritumoral, and intratumoral + peritumoral image features of both DCE and ADC images.

    The radiomic models were developed using an extreme gradient boosting (XGBoost) algorithm, based on the selected radiomic features. A grid search method and five-fold cross-validation were conducted to determine the optimal hyperparameters for the model. Four folds (80% of the patients) were utilized for training the model, while the remaining fold (20% of the patients) was used to select the optimal hyperparameters. The hyperparameters ‘learning_rate’, ‘n_estimators’, and ‘max_depth’ were used in the grid search for model development. To ensure the robustness of the model, the entire construction process was replicated 1000 times using the bootstrap method. The effectiveness of the models was evaluated by analyzing their Receiver Operating Characteristic (ROC) curves in both training and test sets. An optimal model, named the Original model, was selected from the 9 models based on the AUC value.

    For the construction of the Delta model, the differences in the features of Original model between the pre-NAC and mid-NAC stages were calculated. Specifically, for each feature, the value at the mid-NAC stage was subtracted from the value at the pre-NAC stage. These calculated differences served as the new feature set for the Delta model.

    For the Fusion model, we employed a union of features from both the Original model and the Delta model. Specifically, the Fusion model incorporates both the feature values from the pre-NAC and mid-NAC stages themselves and the differences in feature values between these stages.

    The XGBoost algorithm, which was the same as that used in the construction of the Original model, was employed to train the Delta model and the Fusion model.

    2.8 Hybrid model construction and validation

    To construct hybrid models, the outputs of the Original, Delta, and Fusion radiomic models were used as radiomic signatures. Important clinicopathological variables are separately selected for each radiomic signature in the training set. Specifically, individual logistic regression models were established for each radiomic signature, incorporating all clinicopathological variables. These models were used to evaluate the association of the combined variables with the relative efficacy in NAC. Backward stepwise selection based on the Akaike information criterion was then performed to identify important clinicopathological variables for each radiomic signature.

    Hybrid models were constructed using logistic regression, which combined each radiomic signature with corresponding important clinicopathological variables. To select the optimal hyperparameters for the model, a grid search method combined with five-fold cross-validation was implemented. The hyperparameters ‘solver’, ‘penalty’, and ‘C’ were used in the grid search for model development. Additionally, for comparative purposes, a clinicopathological model that exclusively contained all clinicopathological variables was also established.

    2.9 Statistical analysis

    Chi-square tests were used to compare clinicopathological characteristics between patients in different groups or sets using categorical variables. Model performance was evaluated by the area under the receiver operating characteristic curve (AUC). Sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated for both the training and test sets. The 95% confidence intervals (CI) for each metric were computed using the bootstrap method with 1000 intervals. The optimal cutoff value for the radiomic score in the training set was determined by maximizing the Youden index, and these fixed cutoff values were subsequently applied to the test set. All statistical tests were two-sided, with statistical significance indicated by a P value <0.05. All statistical analyses were performed using R 4.2.3 or Python 3.9.13.

    3 Results

    3.1 Baseline characteristics of patients

    A total of 303 patients were excluded from the study due to insufficient MRI data (n=292) or inadequate image quality (n=11). Consequently, 190 patients were included in the study. Table 2 summarizes the baseline characteristics of all patients. The proportion of patients for whom AC was superior to T was 48.1% (64 out of 133) in the training set and 47.4% (27 out of 57) in the test set. No significant differences in clinicopathological characteristics were observed between patients for whom AC was superior to T and those for whom T was superior to AC in either set (all p-values > 0.05).

    Table 2

    3.2 Development and performance of radiomic models

    First, nine radiomic models were constructed by integrating features from specific MRI sequences and regions. Details of the selected features for the nine radiomic models are provided in Supplementary Table S1. Table 3 provides the performance metrics of each radiomic model within the training and test sets. In the training set, the DCE+ADC-tumor+peri model yielded the best prediction with an AUC of 0.864 (95% CI: 0.792-0.911). In the test set, models utilizing the combined DCE+ADC sequences demonstrated relatively higher and more stable performance compared to models using either the DCE or ADC sequence alone. Specifically, the AUCs were 0.663 (95% CI: 0.549-0.767) for the DCE+ADC-tumor model, 0.598 (95% CI: 0.502-0.690) for the DCE+ADC-peri model, and 0.585 (95% CI: 0.480-0.677) for the DCE+ADC-tumor+peri model. Given the significance of peritumoral features, which reflect the tumor microenvironment and offer valuable insights into tumor behavior, the multi-sequence, multi-region DCE+ADC-tumor+peri model was consequently chosen as the Original model.

    Table 3

    Subsequently, three radiomic models were developed, each based on different aspects of radiomic features: the Original model, the Delta model, and the Fusion model. Details of the selected features for the Original, Delta, and Fusion models are provided in Supplementary Table S2. Table 4 summarizes the AUC, accuracy, sensitivity, specificity, PPV, and NPV for each radiomic model within the training and test sets. The Delta model outperformed the Original model with an AUC of 0.887 (95% CI: 0.816-0.930) in the training set and 0.757 (95% CI: 0.683-0.817) in the test set. The Fusion model did not show improved performance over the Delta model, with an AUC of 0.887 (95% CI: 0.822-0.931) in the training set and 0.749 (95% CI: 0.644-0.837) in the test set.

    Table 4

    3.3 Development and performance of hybrid models

    Hybrid models were constructed by combining radiomic features with clinical indicators to predict the relative efficacy in NAC. Table 4 provides a summary of the AUC, accuracy, sensitivity, specificity, PPV, and NPV for each hybrid model in both the training and test sets. After separately incorporating radiomic signatures and corresponding important clinicopathological features into three hybrid models, the performance of Delta+clinicopath model continued to outperform the Original+clinicopath and Fusion+clinicopath models, achieving an AUC of 0.887 (95% CI: 0.873-0.892) in the training set and 0.772 (95% CI: 0.744-0.786) in the test set.

    During the development of hybrid models, alongside the Original radiomic signature, menstrual status, clinical T stage, and HER2 status were identified as independent predictors of the relative efficacy in NAC. For the Delta radiomic signature, clinical T stage was identified as an independent predictor of the relative efficacy in NAC (Table 5). Similarly, the clinical T stage also independently predicted the relative efficacy in NAC for the Fusion radiomic signature.

    Table 5

    Figure 4 displays the ROC curves for the most effective radiomic model (the Delta model), the most effective hybrid model (the Delta+clinicopath model) and the clinicopathological model alone. In the training set, both the Delta model and the Delta+clinicopath model achieved an AUC of 0.887. However, the Delta+clinicopath model exhibited a more concentrated performance, with a 95% CI ranging from 0.873 to 0.892, compared to the Delta model’s wider range of 0.816 to 0.930. In the test set, the Delta+clinicopath model produced a higher AUC of 0.772 in contrast to the Delta model’s 0.757. The combined performance on the training and test sets underscores the advantages of the Delta+clinicopath model in stratifying patients by predicting their relative efficacy in NAC.

    Figure 4

    For the optimal hybrid model, calibration curve analysis showed reasonable consistency between the predicted probabilities and actual outcomes regarding the relative efficacy in NAC in both the training and test sets (Figure 5). The Hosmer-Lemeshow test yielded non-significant results in both sets, with a p-value of 0.592 in the training set and 0.295 in the test set, indicating no significant deviation from a perfect model fit. DCA revealed that the optimal hybrid model delivered a substantial clinical net benefit at all threshold probabilities in the training set (as shown in Figure 6A) and between 0 and 0.74 in the test set (as shown in Figure 6B).

    Figure 5

    Figure 6

    4 Discussion

    Longitudinal monitoring of tumor response during NAC using imaging methods and adjusting subsequent regimens based on the treatment response is essential for personalized treatment of breast cancer and maximizing the potential of NAC (15, 16). Accurate prediction of the relative efficacy of regimens is crucial for optimizing NAC treatment plans. In the AC-T regimen, if the initial four cycles of AC demonstrate superior efficacy, extending the AC treatment may provide greater benefits to the patient. Conversely, if the initial four cycles of AC prove less effective, timely switching to T aligns with the patient’s best interest. However, few studies have explored how to predict the relative efficacy during the mid-term of NAC. In this study, we developed a multi-sequence, multi-region MRI radiomic model to predict the relative efficacy of four cycles of AC treatment followed by four cycles of T treatment, demonstrating robust performance. Additionally, integrating clinicopathological factors with radiomics significantly enhanced predictive accuracy, suggesting that our model could inform treatment adjustments during the mid-term of NAC.

    The best model for predicting the relative efficacy of AC and T during NAC was the Delta + clinicopath model. A similar prospective trial by Guo et al. (17) demonstrated the predictive power of combining delta radiomic features with clinical indicators in predicting pathologic complete response after NAC. They developed a model combining delta radiomic features with clinical indicators, which achieved AUCs of 0.934 and 0.864 in the training and test sets, respectively, outperforming the model based solely on delta radiomic features. This result aligns with our study, highlighting the valuable contribution of breast cancer-related clinical or pathological factors in improving the accuracy of delta radiomics.

    To capture the complex tumor microenvironment changes induced by NAC, we utilized a multi-sequence strategy with DCE and ADC. DCE is widely used in breast MRI radiomics studies for visualizing tumor vascularity, while ADC reflects tissue microstructure, cellular density, and membrane integrity (18). Prior studies have shown that combining DCE with ADC or DWI in radiomic models predicts NAC treatment response more accurately (19–22), and our results are consistent with these findings, underscoring the value of such combinations for evaluating NAC efficacy. Additionally, because peritumoral features such as lymphovascular invasion and angiogenesis are key prognostic factors (23, 24), and the efficacy of peritumoral radiomics has been preliminarily demonstrated in NAC response evaluation (25), we included both intratumoral and peritumoral regions in our multi-region strategy. For this multi-region strategy, our results are in line with previous studies that combine imaging features from multiple regions (26–29), showing improved predictive performance compared to using features from a single region. The role of multi-sequence and multi-region strategies can be observed by examining the features incorporated into the models. In the radiomic models (Supplementary Table S1), the DCE+ADC-tumor model included 8 DCE features and 6 ADC features, the DCE+ADC-peri model included 5 DCE features and 7 ADC features, and the DCE+ADC-tumor+peri model included 8 DCE features and 7 ADC features. This even distribution of DCE and ADC features across the models indicates the enhancement of predictive performance through multi-sequence combination. From a multi-region perspective, the DCE+ADC-tumor+peri model included 8 intratumoral and 7 peritumoral features, suggesting that the combination of multi-sequence and multi-region features captured richer tumor-related information, making it particularly effective for predicting relative efficacy in NAC.

    However, this study has several limitations. First, while our model predicts relative efficacy between NAC regimens based on tumor volume changes, it is unclear whether adjusting treatment based on this prediction will improve pathological complete response or overall outcomes. This requires further validation through prospective trials. Second, although this study utilized the second post-contrast phase images of the DCE sequence for segmentation to improve reproducibility, it should be acknowledged that volumetric assessment of non-concentric regression tumors poses unique challenges, with MRI potentially overestimating or underestimating tumor volume in such cases. Third, the study’s retrospective design, small sample size, and single-center data limit the generalizability of the findings. Larger, multicenter studies are needed for broader validation. Fourth, the focus on the AC-T regimen may introduce biases, including the underrepresentation of HER2 positive cases, which will be addressed in future studies.

    5 Conclusions

    In this study, we developed a hybrid radiomic model that integrates clinical and biopsy pathology data with pre-NAC and mid-NAC breast MRI to predict the relative efficacy of AC and T treatments. This model demonstrated strong predictive performance and could serve as a valuable tool for guiding treatment decisions during NAC. By enabling early prediction of treatment response, the model holds potential for patient stratification and personalized adjustment of NAC regimens.

    Data availability statement

    The original contributions presented in the study are included in the article/- Supplementary Material. Further inquiries can be directed to the corresponding authors.

    Ethics statement

    The studies involving humans were approved by the First Affiliated Hospital with Nanjing Medical University. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because it is a retrospective study.

    Author contributions

    KL: Data curation, Methodology, Project administration, Writing – original draft. RZ: Data curation, Investigation, Methodology, Software, Writing – original draft. JZ: Data curation, Methodology, Software, Writing – review & editing. SQW: Formal analysis, Resources, Validation, Visualization, Writing – review & editing. YJ: Formal analysis, Investigation, Validation, Writing – review & editing. FW: Data curation, Resources, Supervision, Validation, Writing – review & editing. JW: Conceptualization, Supervision, Validation, Writing – review & editing. SJW: Conceptualization, Funding acquisition, Supervision, Writing – review & editing. XZ: Conceptualization, Supervision, Writing – review & editing. YT: Conceptualization, Supervision, Writing – review & editing.

    Funding

    The author(s) declare that financial support was received for the research and/or publication of this article. This work was supported by Jiangsu Province Hospital High-level Talent Cultivation Program (Phase I, CZ0121002010039 to Shouju Wang), the Major Basic Research Fund of Jiangsu Province Hospital (TS202401 to Shouju Wang), the National Natural Science Foundation of China (No. 82372019 to Shouju Wang), and the Double First-Class Foundation of China Pharmaceutical University (No. CPUQNJC22_03 to Shouju Wang).

    Conflict of interest

    The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

    The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.

    Generative AI statement

    The author(s) declare that no Generative AI was used in the creation of this manuscript.

    Publisher’s note

    All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

    Supplementary material

    The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fonc.2025.1544833/full#supplementary-material

    Supplementary Table 1 | Details of selected features in 9 basic radiomic models.

    Supplementary Table 2 | Details of selected features in the Original model, the Delta model, and the Fusion model.

    References

  • Combination of Significant Weight Gain and Late Motherhood Greatly Increases Risk of Breast Cancer, Study Finds

    A recent study has found that women who experience significant weight gain after the age of 20, combined with either having their first child after the age of 30 or not having children at all, face a markedly increased risk of developing breast cancer. According to the research, these women are almost three times more likely to develop breast cancer than those who give birth at a younger age and maintain a stable weight throughout their adult years. The findings shed light on how lifestyle and reproductive factors together can influence breast cancer risk and underscore the importance of maintaining a healthy weight and understanding reproductive health timelines. Breast cancer remains one of the most common cancers globally, affecting millions of women each year. Identifying risk factors plays a crucial role in prevention, early detection, and management. While age, genetics, and family history have long been recognized as non-modifiable risk factors, modifiable factors such as weight and reproductive history are receiving growing attention for their significant impact. The study highlights two key risk components: post-20 weight gain and reproductive timing. Women who gain a substantial amount of weight after their early adulthood seem to be at elevated risk. This risk is exacerbated when combined with late childbirth, specifically having the first child after 30 years of age, or choosing not to have children at all. The research suggests that the combination of hormonal changes associated with pregnancy and the effects of carrying excess weight later in life may create a physiological environment more conducive to the development of breast cancer cells. Reproductive history has long been tied to breast cancer risk. Women who have children earlier tend to have a reduced risk compared to those who have children later or not at all. Pregnancy induces breast cell differentiation and hormonal changes that are believed to provide a protective effect against certain types of breast cancer. Conversely, delayed pregnancy or nulliparity (never having given birth) means longer exposure to estrogen and progesterone, hormones linked to breast cancer progression. Weight gain, particularly after young adulthood, is another important factor. Excess weight, especially in the form of adipose tissue, can lead to increased estrogen production and chronic inflammation, both of which have been implicated in cancer development. The accumulation of body fat after age 20 may influence hormone levels and metabolic processes over time, further increasing breast cancer risk for some women. The interplay between weight gain and reproductive timing suggests that maintaining a stable and healthy weight throughout adulthood may be an important strategy for reducing breast cancer risk, particularly for women who have children later in life or choose not to have them. These findings support public health initiatives aimed at promoting healthy lifestyles, including diet, physical activity, and weight management as part of broader cancer prevention efforts. Healthcare providers are encouraged to consider these combined risk factors when assessing breast cancer risk profiles in women. This approach can help tailor screening recommendations and preventive advice. For example, women who have had significant weight gain after age 20 and delayed childbirth or no childbirth might benefit from more vigilant breast cancer screening and lifestyle modification counseling. While this study adds valuable insight into breast cancer risk, researchers emphasize that it is one piece of the complex risk puzzle. Genetics, environmental exposures, lifestyle factors, and access to healthcare all interact to influence individual risk. Ongoing research is needed to further unravel how these factors combine and how best to implement effective prevention strategies. In conclusion, the study underscores the importance of healthy weight maintenance and reproductive health planning, demonstrating their significant roles in breast cancer risk. Women are advised to discuss their individual risks with healthcare professionals and consider lifestyle choices that can contribute to long-term health benefits. Early prevention and awareness remain key in the fight against breast cancer.

  • Novel Cancer Immunotherapy Offers New Hope for Late-Stage Cancer Patients

    Researchers from the A*STAR Institute of Molecular and Cell Biology (A*STAR IMCB), in collaboration with the local biotechnology firm Intra-ImmuSG, have recently announced remarkable findings from a Phase II clinical trial evaluating a novel cancer immunotherapy. This innovative treatment demonstrated significant effectiveness in patients suffering from late-stage cancers—a demographic that has traditionally faced scarce and often ineffective therapeutic options. The encouraging results not only highlight the potential for this therapy to transform the landscape of cancer care but also provide renewed hope to both patients and healthcare professionals grappling with the immense challenges posed by advanced malignancies.

    Cancer continues to be one of the leading causes of illness and death on a global scale, with late-stage cancers presenting some of the toughest hurdles in treatment. Conventional methods such as chemotherapy, radiation, and surgery, while valuable, frequently fall short once the disease has progressed extensively. This predicament has driven scientists to explore novel directions, with immunotherapy emerging at the forefront. By harnessing the body's own immune system to identify and attack cancer cells, immunotherapy represents a promising paradigm shift in oncology. The collaborative efforts between A*STAR IMCB and Intra-ImmuSG exemplify the power of blending academic research prowess with the agility and innovation of the biotech industry to create cutting-edge therapeutic solutions.

    The Phase II clinical trial was meticulously designed to assess not only the safety and tolerability but also the therapeutic efficacy of this new immunotherapy across a diverse group of patients afflicted with various types of advanced-stage cancers. Many of these patients had already undergone multiple standard treatments with diminishing returns. The trial's results were nothing short of impressive: a significant proportion of participants experienced meaningful tumor shrinkage accompanied by manageable side effects. This is especially noteworthy because late-stage cancer patients typically have limited prospects and endure harsh side effects with conventional therapies. The underlying mechanism of this novel treatment involves activating and empowering the body’s natural immune defenses to recognize and eliminate tumor cells more efficiently, opening the door to potentially longer-lasting remission and improved quality of life.

    One of the most exciting aspects highlighted by the research team is the novelty of the immunological targets and the therapeutic platform that form the foundation of this therapy. Unlike current checkpoint inhibitors—another class of immunotherapies that have revolutionized cancer treatment in recent years—this new approach engenders a more potent and durable immune response tailored intricately to the tumor’s unique microenvironment. This precision targeting is a significant leap forward in the field and underscores the essential role of translational research; it bridges the gap between the laboratory bench and bedside application, validating complex biological concepts through tangible patient outcomes. The promising results from this study pave the way for the upcoming Phase III clinical trials, which will aim to refine dosing strategies, confirm safety and efficacy, and bring this therapy closer to regulatory approval and widespread clinical adoption.

    Beyond the immediate clinical impact, this breakthrough underscores the vital contribution of local research institutions and biotechnology companies in driving innovation. The partnership between A*STAR IMCB and Intra-ImmuSG serves as a powerful model of how interdisciplinary collaboration can accelerate the development of life-changing medical therapies. It also highlights the importance of sustained investment in biomedical research, which is essential for confronting formidable diseases like cancer. While the path from promising clinical trial data to routine clinical use is complex and fraught with challenges, the progress achieved here marks a significant milestone. This novel immunotherapy holds the promise of becoming a new standard treatment specifically tailored for advanced cancer patients with limited options. Moreover, it may lay the groundwork for personalized medicine strategies that customize treatments according to the distinct molecular characteristics of individual tumors, potentially revolutionizing how we approach cancer therapy in the future.

    In conclusion, the Phase II clinical trial results from A*STAR IMCB and Intra-ImmuSG herald a breakthrough moment in the realm of cancer immunotherapy. The therapy’s notable efficacy and manageable safety profile offer fresh optimism for patients facing late-stage cancers, a group that often confronts bleak prognoses. Continued research efforts, strategic investment, and collaborative initiatives will be key to transforming these encouraging results into accessible, effective treatments. These advancements have the potential not only to improve patient survival but also to enhance the quality of life for cancer sufferers worldwide. As the fight against cancer continues, this innovative immunotherapy shines as a beacon of hope and a testament to the power of science and collaboration.

    #CancerImmunotherapy #MedicalBreakthrough #AstarIMCB #IntraImmuSG #CancerResearch #ImmunotherapyInnovation #HopeForCancerPatients

    Source

  • MoonRISe-1: A Targeted Approach to Intermediate-Risk NMIBC

    Roger Li, MD, a genitourinary oncologist at Moffitt Cancer Center in Tampa, Florida, discusses the MoonRISe-1 study (NCT06319820) evaluating TAR-210 in FGFR-altered intermediate-risk non-muscle-invasive bladder cancer (NMIBC). Here, he highlights how this targeted approach compares with the current standard of care of intravesical chemotherapy.

    “Not only in my practice, but also just for treatment of intermediate-risk NMIBC in general, there has not traditionally been a targeted approach,” he explains.

    The randomized, phase 3 MoonRISe-1 study is evaluating the efficacy of TAR-210 to prevent recurrence within the bladder by inserting the TAR-210 in the bladder every 3 months for up to 1 year, vs intravesical chemotherapy using an induction course of 4 to 6 weekly induction, followed by a 6 to 12 monthly maintenance therapy. The study’s primary end point is disease-free survival.1

    Li continues to discuss how erdafitinib (Balversa) has been approved by the FDA in the metastatic setting after 1 line of systemic treatment with progression.2 The approval of erdafitinib in January 2024 was specifically for the treatment of patients with locally advanced or metastatic urothelial carcinoma with FGFR3 alterations with disease progression following a prior line of treatment. The approval, which was supported by findings from the THOR study (NCT03390504), showed that erdafitinib led to statistically significant improvements in overall survival, progression-free survival, and overall response rate in this patient population.

    “That has really been the only targeted agent that has been approved for use in all of bladder cancer. It is very exciting for us to move to a targeted therapy paradigm, particularly using a drug that is so well tolerated,” he adds.

    “Along with the trial, we will also be able to understand potential mechanisms of resistance and understand the mechanisms that are kind of underpinning either resistance to treatment or just progression in general,” Li shares.

    REFERENCES:

    1. A study to evaluate TAR-210 versus single agent intravesical cancer treatment in participants with bladder cancer (MoonRISe-1). ClinicalTrials.gov. Updated April 27, 2025. Accessed May 14, 2025. https://clinicaltrials.gov/study/NCT06319820

  • Differentiating Bispecific Administration Between Institutions for Multiple Myeloma

    During a Satellite Sessions program that was hosted by CancerNetwork®, a panel of experts in hematologic cancers discussed their institutions’ practices for administering bispecific antibodies, specifically B-cell maturation antigen (BCMA) or GPRC5D, to patients with multiple myeloma. Factors such as chimeric antigen receptor (CAR) T-cell eligibility, distance, age, and agent availability were all considered relevant and worthy of consideration when assessing potential treatment options.

    The panel was led by Thomas G. Martin, MD, associate director of the myeloma program at the University of California, San Francisco (UCSF) Helen Diller Family Comprehensive Cancer Center, director of the Unrelated Donor Transplantation Programs for Adults at UCSF, and research director of hematologic malignancies. He was joined by Alfred Chung, MD, a hematologist-oncologist at UCSF; Katrina Anne Fischer, MD, MBA, an oncologist at Pacific Cancer Care in Monterey, California; Robert W. Weber, MD, a medical oncologist at Sutter Health in San Francisco, California; Samantha Shenoy, MSN, NP, a nurse practitioner at UCSF; Aditi Choudhry, MD, a medical oncologist at John Muir Health in Walnut Creek, California; Gigi Chen, MD, a hematologist at John Muir Health in Walnut Creek, California; Derek A. Wong, MD, an associate clinical professor and medical director of the MS in Genetic Counseling Program, Pediatrics, at UCLA David Geffen School of Medicine; and Ram K. Chillar, MD, a hematologist-oncologist in California.

    Candidates for Bispecific Antibodies

    Martin / Who are the candidates we should be [administering bispecifics in an outpatient setting]?

    Chung / It depends on [whether] they’re eligible for CAR T because there are data that were presented here that [show], after a prior T-cell redirection therapy, the duration of responses is much shorter. If a patient is CAR T–eligible and we have a plan for that, I would do that first. If they’re not going to be eligible for social reasons or health reasons, a lot of patients can tolerate the bispecific antibodies. The cytokine release syndrome [CRS] is generally manageable. Then, as long as we’re careful with the intravenous immunoglobulin prophylaxis and infection monitoring, BCMA bispecifics can be done. With talquetamab-tgvs [Talvey], the toxicity profile is different… I usually go for BCMA first for targeting and then GPRC5D. That’s my practice.

    Martin / How about [you], Katrina? Are there patients [who] you refer to [UCSF] and say, “That guy is going to need a bispecific” or somebody [who] you don’t refer, and say, “I’m not going to send that guy [to UCSF]?”

    Fischer / We comanage a lot of our patients with multiple myeloma because they’ve seen you [at UCSF and are] ready for transplant, and they’ve been doing maintenance therapy for us. I did recently have somebody on the [GPRC5D], and I referred him up for CAR T. This wasn’t on my radar too much because we’re not giving it, and we haven’t given any bispecific antibodies in Monterey other than with this one patient. It was a big deal to get it all started in our clinic, and our hospital is not willing to do the ramp-up [dosing]. [Jeffrey L. Wolf, MD] had said—the patient’s [older]—and he said, “Let’s do the GPRC5D first because if I want to do [something] BCMA targeted later, I haven’t targeted that yet.”

    Weber / No, I refer everybody [to UCSF], and we’re so close that they don’t stay [inpatient].

    Choosing the Right Bispecific Antibody

    Martin / Sam, do you use BCMA or GPRC5D first?

    Shenoy / Like Alfred said, if you can do ciltacabtagene autoleucel [Carvykti], you should do that first, but if you can’t do CAR T, I don’t think it matters if you do BCMA vs GPRC5D first. If you’re talking about bispecific to bispecific, and we’ve done that before, [I prefer] GPRC5D first and then BCMA. I don’t feel like it matters.

    Martin / How about you, Gigi and Aditi? What’s your choice at John Muir?

    Choudhry / We have both bispecifics available. I don’t have a lot of experience [with them]. I have one patient on teclistamab-cqyv [Tecvayli] right now, but that was [started] before talquetamab got approved. He’s been on it for amyloidosis for the past 3 years now in a complete response.

    Now, if I have a patient, I’d probably use talquetamab because it’s GPRC5D. If they need CAR T later, then [I would administer it] because I don’t want to give them a BCMA-directed bispecific. I just haven’t had a patient. One of my partners has—it was 3 weeks ago—I was there for the ramp-up dose. It went just fine, but he’s not my patient, so I don’t know how he’s doing. We can give both.

    Chen / We usually first refer to [UCSF] for a CAR T evaluation to see if they’re a CAR T candidate. A lot of it is a discussion to decide on which [bispecific] to go with—the BCMA vs the GPRC5D. Recently, I was on service last week, and we started somebody with talquetamab. That was quite an exciting ramp-up…. We do it on day 1 and day 4, and it sounds like [at UCSF you] do it on day 1 and day 3. [That’s a] faster ramp-up than what we did. There was CRS, and there was a concern for [immune effector cell–associated neurotoxicity syndrome (ICANS)], but I think it was all manageable. I don’t think anybody who rolls in can get it. At least for the ICANS, you have to have them be able to write things and see their handwriting. You have to be able to go through all those tests.

    Martin / That’s a good point. Some of our patients didn’t do so well on the ICANS before they got the [medicine]. Derek, how about your practice?

    Wong /I refer them all to you…. The first question is, do they get the CAR T? Mainly, we have not done it yet because our hospital is [El Camino Health], and it’s not part of [Sutter Health]. We can’t get it, so it’s been hard to start.

    Considering Age When Administering Bispecifics

    Martin / For older patients [who] you look at and say, “OK, that guy can’t get a CAR [T-cell therapy],” what do you think about a bispecific [antibody]? Are you thinking that [with] bispecific antibodies, we can accept [patients who are] a little bit less fit?

    Wong / I’m looking for the right opportunity to start. Sooner or later, we’re going to have to start giving [bispecific antibodies to them].

    Weber / My patient population is very old, and I never thought of it as something they would be eligible for. I thought, “How will they live long enough to get through 4 lines?”

    Martin / Well, that is the attrition of multiple myeloma, which is the craziest thing. So, 100% of patients can get line 1 [of treatment], in general. Then, when you get to line 2, we’re down to 60% of patients—the [other] 40% either have died, gotten some bad toxicity, or just quit. You lose that many. [In the] third line of therapy, you lose another 20%, and [in the] fourth line of therapy, you lose another 20%. That’s why we’re trying to move [bispecific antibodies further].

    Weber / By that time, you have a battle-tested patient.

    Martin / Correct. If they can make [it through] that many lines [of therapy], they can get [bispecific antibodies].

    Chen / Is there an age cutoff for CAR T? Anytime I want to give a bispecific antibody, the patient gets taken for CAR T.

    Martin / For CAR T, there is not an age cutoff. Over age 85, it starts to get challenging… Under 80, [there is] no problem. Over 80, [it’s] tough. They have to want it.

    Discussing Sequencing and Safety

    Wong / With the quadruplet therapy [option] now in the first line, and I know it’s relatively new, it’s hard to get to the fourth line. What do you use in the second and third lines when you’ve used all triple-class [therapies]?

    Martin / The real answer to your question is based on [the phase 2 GRIFFIN trial (NCT02874742)] because GRIFFIN started way before all this stuff. GRIFFIN was the first [quadruplet therapy] that we used. I think [at UCSF] we put 9 or 10 patients on GRIFFIN. They’re all still in remission. That was [around] 7 years ago.

    [The question is], what are we going to use when they come out of remission? The majority of those patients, they’re off treatment. I’ve taken them off as of years 3, 4, or 5, and they’ve been off therapy for 2 or 3 years. Honestly, when they relapse, anything’s possible. Whatever’s available at that point in time, to tell you the truth, because they’re not going to be refractory to anything…. We’re going to be able to use whatever the best thing is at that point in time.

    Chillar / Likewise, I have these patients from the Bay Area who are shifting to the Carson Valley [location], and sometimes it becomes an issue. Most of them are in [their] 80s. I’m strictly advising them that it is available there. The one we tried [this] with, this gentleman is 93 [years old] already. He says, “What are my options now? I’ve run out of everything.” I said, “This is a possibility, but you may not have the ability to withstand some of the [adverse] effects. [You] might be done with the CRS, even though we are getting better at handling it.” I’m troubled in that narrow channel.

    Martin / In the real world, we have all these toxicities: skin, nails, and dysgeusia. Sam, how do we manage those?

    Shenoy / The biggest thing…is setting expectations from the very beginning. Giving [the patient] a handout [helps]; it gives them tips about how to manage the [adverse] effects. [It’s] so important to give patients those tools so they go in feeling like they can do something about some of the [adverse] effects that they’re going to experience.

    For dysgeusia, which is the most bothersome symptom, dose reduction, which once they’ve achieved a [partial response], usually by cycle 3 or 4, it’s safe to go from dosing every 2 weeks to monthly. That is probably one of the best things you can do…. Honestly, the best thing you can do is dose reduction when they’ve achieved a partial response. After that, just [give them] the best supportive care that you can.

    Reference

    Voorhees PM, Sborov DW, Laubach J, et al. Addition of daratumumab to lenalidomide, bortezomib, and dexamethasone for transplantation-eligible patients with newly diagnosed multiple myeloma (GRIFFIN): final analysis of an open-label, randomised, phase 2 trial. Lancet Haematol. 2023;10(10):e825-e837. doi:10.1016/S2352-3026(23)00217-X

  • Five Under 5: Top Oncology Videos for the Week of 5/11

    Welcome to The Five Under 5, your go-to roundup of the top 5 videos of the week.

    These short videos are designed for busy oncologists to view on the go, and feature expert insights on breaking news, regulatory updates, practice-changing data shared at medical meetings, and other key topics in the realm of oncology.

    Here’s what you may have missed:

    David S. Hong, MD, of The University of Texas MD Anderson Cancer Center, discusses the clinical significance of the FDA’s April 2025 full approval of larotrectinib (Vitrakvi) for NTRK fusion–positive solid tumors. He explained that pooled data from the LOXO-TRK-14001 (NCT02122913), SCOUT (NCT02637687), and NAVIGATE (NCT02576431) trials showed a 60% overall response rate, including a 24% complete response (CR) rate; the median duration of response was 43.3 months. Hong emphasized that these results confirm the durable, tumor-agnostic efficacy of larotrectinib in both adult and pediatric patients, irrespective of tumor type. He also highlighted its favorable toxicity profile, noting that adverse effects such as dizziness and elevated liver function tests are typically manageable. No new safety concerns were identified with the drug.

    Deena M. Atieh Graham, MD, of John Theurer Cancer Center at Hackensack Meridian Health, discusses the FDA’s March 2024 regular approval of mirvetuximab soravtansine-gynx (Elahere) for patients with folate receptor α–positive, platinum-resistant ovarian, fallopian tube, or primary peritoneal cancer. The decision was based on results from the phase 3 MIRASOL trial (NCT04209855), in which patients treated with the antibody-drug conjugate (ADC) achieved a median overall survival of 16.5 months vs 12.7 months with investigator’s choice of chemotherapy. Graham emphasized that ADCs represent a durable shift in ovarian cancer treatment, and their efficacy is determined by the target, linker, and payload—elements that can be further optimized. She also noted that mirvetuximab soravtansine is now being evaluated in the maintenance setting in the phase 3 GLORIOSA trial (NCT05445778), reflecting the broader move toward personalized, targeted therapies in this disease.

    Tanya B. Dorff, MD, of City of Hope, discussed early insights from the real-world OPTYX study (NCT05467176) evaluating relugolix (Orgovyx) in patients with advanced prostate cancer. This prospective, non-interventional registry is assessing how relugolix is used across diverse US clinical settings, including both academic and community practices. Initial data from 999 patients showed that relugolix is frequently used in localized and locally advanced disease, as well as in those with biochemical recurrence or metastatic disease. More than half of patients received the agent in combination with other systemic therapies, such as androgen receptor pathway inhibitors. Dorff noted that the inclusion of patient-reported outcomes, including quality of life and treatment expectations, will provide critical context for real-world treatment decisions.

    Seema A. Bhat, MD, of The Ohio State University, discusses the limited treatment options and poor prognosis for patients with double-refractory chronic lymphocytic leukemia (CLL). These patients often develop mutations that render current therapies ineffective and experience rapid disease progression, highlighting the need for prompt treatment. Two FDA-approved options for this population include the noncovalent BTK inhibitor pirtobrutinib (Jaypirca), which received accelerated approval in 2023 based on phase 1/2 BRUIN trial (NCT03740529) data showing a 72% ORR, and the CD19-directed CAR T-cell therapy lisocabtagene maraleucel (Breyanzi), approved in 2024 with a 20% CR rate in the phase 1/2 TRANSCEND CLL 004 study (NCT03331198). Bhat also notes that PI3K inhibitors have limited efficacy and tolerability, which restricts their use in certain patients. Overall, these advances provide important new options but underscore the continued unmet need in double-refractory CLL.

  • Fukui named chief of clinical breast oncology at UH Cancer Center

    Reading time: 2 minutes

    Jami Fukui, a researcher at the University of Hawaiʻi Cancer Center for the past eight years, has been appointed chief of clinical breast oncology for the center’s newly launched Ka ʻUmeke Lama academic oncology and workforce development program. The initiative is designed to cultivate the next generation of homegrown oncologists and expand equitable cancer care throughout Hawaiʻi.

    Related UH News story: Ka ʻUmeke Lama to transform cancer care in Hawaiʻi, Pacific, December 2024

    In this new role, Fukui will lead the development of clinical oncology programs and oversee clinical trials to help establish new standards of care for managing breast cancer. All clinical and research efforts related to breast cancer at the UH Cancer Center will now be consolidated under her leadership to improve patient care and outcomes.

    “Hawaiʻi has one of the highest incidences of breast cancer cases in the U.S.,” Fukui said. “It’s a privilege to learn from my patients and our community. They guide my research to find ways of overcoming health disparities in cancer treatment, and help to focus our collective efforts to find better ways to care for those diagnosed with breast cancer in Hawaiʻi and the Pacific.”

    Prevention, earlier detection, better treatment

    Fukui has focused her research on various aspects of cancer care, with an emphasis on reducing health disparities. She served as principal investigator for a project that engaged community health educators to support Native Hawaiian and Pacific Islander cancer patients in rural communities.

    “Dr. Fukui’s genuine compassion and a wealth of experience enables her to marry science and the social-emotional aspects of patient care,” said Naoto Ueno, director of the UH Cancer Center. “Her oversight will further advance groundbreaking research that translates into prevention, earlier detection and better treatment or management of the disease, especially for Native Hawaiians, Pacific Islanders and Filipinos, who have a higher prevalence of certain types of cancer. Dr. Fukui is an inspiration to up-and-coming oncologists who will be able to care for those in their own communities.”