Ovarian function in female survivors of high-risk neuroblastoma.

INTRODUCTION: Data on ovarian function in neuroblastoma survivors are limited. We sought to determine the prevalence of ovarian dysfunction in a cohort of high-risk neuroblastoma survivors and compare outcomes among survivors treated with and without autologous stem cell rescue (ASCR) preceded by myeloablative chemotherapy. METHODS: Retrospective review of female survivors of high-risk neuroblastoma ≥5 years from diagnosis, diagnosed between 1982 and 2014, and followed in a tertiary cancer center. Participants were divided into two groups: individuals treated with conventional chemotherapy ± radiation ("non-ASCR") (n = 32) or with chemotherapy ± radiation followed by myeloablative chemotherapy with ASCR ("ASCR") (n = 51). Ovarian dysfunction was defined as follicle-stimulating hormone ≥15 mU/mL, while premature ovarian insufficiency (POI) was defined as persistent ovarian dysfunction requiring hormone replacement therapy. Poisson models were used to determine prevalence ratios of ovarian dysfunction and POI. RESULTS: Among 83 females (median attained age: 19 years [range, 10-36]; median follow-up: 15 years [range, 7-36]), 49 (59%) had ovarian dysfunction, and 34 (41%) developed POI. Survivors treated with ASCR were 3.2-fold more likely to develop ovarian dysfunction (95% CI: 1.8-6.0; p < 0.001) and 4.5-fold more likely to develop POI (95% CI: 1.7-11.7; p = 0.002) when compared with those treated with conventional chemotherapy, after adjusting for attained age. Two participants in the non-ASCR group and six in the ASCR group achieved at least one spontaneous pregnancy. DISCUSSION: Ovarian dysfunction is prevalent in female high-risk neuroblastoma survivors, especially after ASCR. Longitudinal follow-up of larger cohorts is needed to inform counseling about the risk of impaired ovarian function after neuroblastoma therapy.

Long Prime-Boost Interval and Heightened Anti-GD2 Antibody Response to Carbohydrate Cancer Vaccine.

The carbohydrate ganglioside GD2/GD3 cancer vaccine adjuvanted by ß-glucan stimulates anti-GD2 IgG1 antibodies that strongly correlate with improved progression-free survival (PFS) and overall survival (OS) among patients with high-risk neuroblastoma. Thirty-two patients who relapsed on the vaccine (first enrollment) were re-treated on the same vaccine protocol (re-enrollment). Titers during the first enrollment peaked by week 32 at 751 ± 270 ng/mL, which plateaued despite vaccine boosts at 1.2-4.5 month intervals. After a median wash-out interval of 16.1 months from the last vaccine dose during the first enrollment to the first vaccine dose during re-enrollment, the anti-GD2 IgG1 antibody rose to a peak of 4066 ± 813 ng/mL by week 3 following re-enrollment (p < 0.0001 by the Wilcoxon matched-pairs signed-rank test). Yet, these peaks dropped sharply and continually despite repeated boosts at 1.2-4.5 month intervals, before leveling off by week 20 to the first enrollment peak levels. Despite higher antibody titers, patients experienced no pain or neuropathic side effects, which were typically associated with immunotherapy using monoclonal anti-GD2 antibodies. By the Kaplan-Meier method, PFS was estimated to be 51%, and OS was 81%. The association between IgG1 titer during re-enrollment and ß-glucan receptor dectin-1 SNP rs3901533 was significant (p = 0.01). A longer prime-boost interval could significantly improve antibody responses in patients treated with ganglioside conjugate cancer vaccines.

Evolutionary Strategies AI Addresses Multiple Technical Challenges in Deep Learning Deployment: Proof-of-Principle Demonstration for Neuroblastoma Brain Metastasis Detection.

Two significant obstacles hinder the advancement of Radiology AI. The first is the challenge of overfitting, where small training data sets can result in unreliable outcomes. The second challenge is the need for more generalizability, the lack of which creates difficulties in implementing the technology across various institutions and practices. A recent innovation, deep neuroevolution (DNE), has been introduced to tackle the overfitting issue by training on small data sets and producing accurate predictions. However, the generalizability of DNE has yet to be proven. This paper strives to overcome this barrier by demonstrating that DNE can achieve satisfactory results in diverse external validation sets. The main innovation of the work is thus showing that DNE can generalize to varied outside data. Our example use case is predicting brain metastasis from neuroblastoma, emphasizing the importance of AI with limited data sets. Despite image collection and labeling advancements, rare diseases will always constrain data availability. We optimized a convolutional neural network (CNN) with DNE to demonstrate generalizability. We trained the CNN with 60 MRI images and tested it on a separate diverse collection of images from over 50 institutions. For comparison, we also trained with the more traditional stochastic gradient descent (SGD) method, with the two variants of (1) training from scratch and (2) transfer learning. Our results show that DNE demonstrates excellent generalizability with 97% accuracy on the heterogeneous testing set, while neither form of SGD could reach 60% accuracy. DNE's ability to generalize from small training sets to external and diverse testing sets suggests that it or similar approaches may play an integral role in improving the clinical performance of AI.

Intraoperative Radiation Therapy for Relapsed or Refractory High-Risk Neuroblastoma: A 27-Year Experience.

PURPOSE: To evaluate outcomes after intraoperative radiation therapy (IORT) in high-risk neuroblastoma (NB), including local control, overall survival, and toxicity. METHODS AND MATERIALS: This was a single institution retrospective study of 92 pediatric patients with NB treated with IORT from 1995 to 2022. Each IORT application was considered a separate event for a total of 110 sites treated. Local failure was calculated using the cumulative incidence function and survival by Kaplan-Meier method from the day of surgery. RESULTS: All patients had high-risk relapsed or treatment refractory disease. Median age was 6 years (range, 2-34 years). Median follow-up for all patients and surviving patients was 16 months and 4 years, respectively. All patients previously received chemotherapy, 93% had prior external beam radiation therapy to the site of IORT (median dose, 21.6 Gy; range, 10-36 Gy), and 94% had a prior surgery for tumor resection. The median IORT dose was 12 Gy (range, 8-18 Gy) and median area treated was 18 cm2 (range, 2.5-60 cm2). The cumulative incidence of local failure was 23% at 2 years and 29% at 5 years. The overall survival (OS) was 44% at 2 years and 29% at 5 years. Local failure after IORT was associated with worse OS (hazard ratio, 1.74; 95% CI, 1.07-2.84; P = .0267). Toxicity from IORT was rare, with postoperative complications likely related to IORT seen in 7 (8%) patients. CONCLUSIONS: Our study represents the largest, most recent analysis of the efficacy and safety of IORT in patients with relapsed or refractory NB. Less than one-third of patients failed locally at 5 years, and achieving local control affected overall survival. Minimal toxicities directly related to IORT were observed. Overall, IORT is an effective and safe technique to achieve local control in high-risk relapsed or refractory neuroblastoma.