Radiation nanomedicines for cancer treatment: a scientific journey and view of the landscape

BACKGROUND

Radiation nanomedicines are nanoparticles labeled with radionuclides that emit α- or β-particles or Auger electrons for cancer treatment. We describe here our 15 years scientific journey studying locally-administered radiation nanomedicines for cancer treatment. We further present a view of the radiation nanomedicine landscape by reviewing research reported by other groups.

MAIN BODY

Gold nanoparticles were studied initially for radiosensitization of breast cancer to X-radiation therapy. These nanoparticles were labeled with 111In to assess their biodistribution after intratumoural vs. intravenous injection. Intravenous injection was limited by high liver and spleen uptake and low tumour uptake, while intratumoural injection provided high tumour uptake but low normal tissue uptake. Further, [111In]In-labeled gold nanoparticles modified with trastuzumab and injected iintratumourally exhibited strong tumour growth inhibition in mice with subcutaneous HER2-positive human breast cancer xenografts. In subsequent studies, strong tumour growth inhibition in mice was achieved without normal tissue toxicity in mice with human breast cancer xenografts injected intratumourally with gold nanoparticles labeled with β-particle emitting 177Lu and modified with panitumumab or trastuzumab to specifically bind EGFR or HER2, respectively. A nanoparticle depot (nanodepot) was designed to incorporate and deliver radiolabeled gold nanoparticles to tumours using brachytherapy needle insertion techniques. Treatment of mice with s.c. 4T1 murine mammary carcinoma tumours with a nanodepot incorporating [90Y]Y-labeled gold nanoparticles inserted into one tumour arrested tumour growth and caused an abscopal growth-inhibitory effect on a distant second tumour. Convection-enhanced delivery of [177Lu]Lu-AuNPs to orthotopic human glioblastoma multiforme (GBM) tumours in mice arrested tumour growth without normal tissue toxicity. Other groups have explored radiation nanomedicines for cancer treatment in preclinical animal tumour xenograft models using gold nanoparticles, liposomes, block copolymer micelles, dendrimers, carbon nanotubes, cellulose nanocrystals or iron oxide nanoparticles. These nanoparticles were labeled with radionuclides emitting Auger electrons (111In, 99mTc, 125I, 103Pd, 193mPt, 195mPt), β-particles (177Lu, 186Re, 188Re, 90Y, 198Au, 131I) or α-particles (225Ac, 213Bi, 212Pb, 211At, 223Ra). These studies employed intravenous or intratumoural injection or convection enhanced delivery. Local administration of these radiation nanomedicines was most effective and minimized normal tissue toxicity.

CONCLUSIONS

Radiation nanomedicines have shown great promise for treating cancer in preclinical studies. Local intratumoural administration avoids sequestration by the liver and spleen and is most effective for treating tumours, while minimizing normal tissue toxicity.

HAVOC: Small-scale histomic mapping of cancer biodiversity across large tissue distances using deep neural networks

Intratumoral heterogeneity can wreak havoc on current precision medicine strategies because of challenges in sufficient sampling of geographically separated areas of biodiversity distributed across centimeter-scale tumor distances. To address this gap, we developed a deep learning pipeline that leverages histomorphologic fingerprints of tissue to create "Histomic Atlases of Variation Of Cancers" (HAVOC). Using a number of objective molecular readouts, we demonstrate that HAVOC can define regional cancer boundaries with distinct biology. Using larger tumor specimens, we show that HAVOC can map biodiversity even across multiple tissue sections. By guiding profiling of 19 partitions across six high-grade gliomas, HAVOC revealed that distinct differentiation states can often coexist and be regionally distributed within these tumors. Last, to highlight generalizability, we benchmark HAVOC on additional tumor types. Together, we establish HAVOC as a versatile tool to generate small-scale maps of tissue heterogeneity and guide regional deployment of molecular resources to relevant biodiverse niches.

Remote collection of electrophysiological data with brain wearables: opportunities and challenges

Collection of electroencephalographic (EEG) data provides an opportunity to non-invasively study human brain plasticity, learning and the evolution of various neuropsychiatric disorders. Traditionally, due to sophisticated hardware, EEG studies have been largely limited to research centers which restrict both testing contexts and repeated longitudinal measures. The emergence of low-cost "wearable" EEG devices now provides the prospect of frequent and remote monitoring of the human brain for a variety of physiological and pathological brain states. In this manuscript, we survey evidence that EEG wearables provide high-quality data and review various software used for remote data collection. We then discuss the growing body of evidence supporting the feasibility of remote and longitudinal EEG data collection using wearables including a discussion of potential biomedical applications of these protocols. Lastly, we discuss some additional challenges needed for EEG wearable research to gain further widespread adoption.

The Incidence and Prevalence of Primary Central Nervous System (CNS) Tumours in Canada (2010-2017), and the Survival of Patients Diagnosed with CNS Tumours (2008-2017)

Primary central nervous system (CNS) tumours are heterogeneous, with different treatment pathways and prognoses depending on their histological and molecular classification. Due to their anatomical location, all CNS tumours, regardless of malignancy, can be debilitating. We used vital statistics linked to Canadian Cancer Registry data to estimate the age-standardized incidence rates (ASIR), Kaplan-Meier survival rates (SR), and limited-duration prevalence proportions (PP) of 25 histology-specific CNS tumour groups that were classified based on site and histology. During 2010-2017, 45,115 patients were diagnosed with 47,085 primary CNS tumours, of which 19.0% were unclassified. The average annual ASIR was 21.48/100,000 person-years and did not vary by sex. The ASIR increased with age, particularly for meningioma, unclassified tumours, and glioblastoma. The eight-year PP was 102.1/100,000 persons (index date 1 January 2018). The most common histology was meningioma (ASIR: 5.19; PP: 31.6). The overall five-year SR among 51,310 patients diagnosed during 2008-2017 was 57.2% (95% CI: 56.8-57.7%). SRs varied by tumour behaviour, histology, and patient age, with the lowest SR among glioblastoma patients (5-year SRs ranged from 1.3-25.7%). For non-malignant tumours, the 5-year SRs ranged from 37.4-100%. We provide the most up-to-date histology-specific surveillance estimates for primary CNS tumours in Canada.

Incidence and survival of primary central nervous system tumors diagnosed in 4 Canadian provinces from 2010 to 2015

Background

The Brain Tumor Registry of Canada was established in 2016 to enhance infrastructure for surveillance and clinical research on Central Nervous System (CNS) tumors. We present information on primary CNS tumors diagnosed among residents of Canada from 2010 to 2015.

Methods

Data from 4 provincial cancer registries were analyzed representing approximately 67% of the Canadian population. Age-standardized incidence rates (ASIR) and 95% confidence intervals (CI) were calculated using the 2011 Canadian population age distribution. Net survival was estimated using the Pohar-Perme method.

Results

A total of 31 644 primary tumors were identified for an ASIR of 22.8 per 100 000 person-years. Nonmalignant tumors made up 47.1% of all classified tumors, with mixed behaviors present in over half of histology groupings. Unclassified were 19.5% of all tumors. The most common histological subtypes are meningiomas (ASIR = 5.5 per 100 000 person-years); followed by glioblastomas (ASIR 4.0 per 100 000 person-years). The overall 5-year net survival rate for CNS tumors was 65.5%; females 70.2% and males 60.4%. GBMs continue to be the most lethal CNS tumors for all sex and age groups.

Conclusions

The low annual frequency of most CNS tumor subtypes emphasizes the value of population-based data on all primary CNS tumors diagnosed among Canadians. The large number of histological categories including mixed behaviors and the proportion of unclassified tumors emphasizes the need for complete reporting. Variation in incidence and survival across histological groups by sex and age highlights the need for comprehensive and histology-specific reporting. These data can be used to better inform research and health system planning.

Mechanical nanosurgery of chemoresistant glioblastoma using magnetically controlled carbon nanotubes

Glioblastoma (GBM) is the most common and aggressive primary brain cancer. Despite multimodal treatment including surgery, radiotherapy, and chemotherapy, median patient survival has remained at ~15 months for decades. This situation demands an outside-the-box treatment approach. Using magnetic carbon nanotubes (mCNTs) and precision magnetic field control, we report a mechanical approach to treat chemoresistant GBM. We show that GBM cells internalize mCNTs, the mobilization of which by rotating magnetic field results in cell death. Spatiotemporally controlled mobilization of intratumorally delivered mCNTs suppresses GBM growth in vivo. Functionalization of mCNTs with anti-CD44 antibody, which recognizes GBM cell surface-enriched antigen CD44, increases mCNT recognition of cancer cells, prolongs mCNT enrichment within the tumor, and enhances therapeutic efficacy. Using mouse models of GBM with upfront or therapy-induced resistance to temozolomide, we show that mCNT treatment is effective in treating chemoresistant GBM. Together, we establish mCNT-based mechanical nanosurgery as a treatment option for GBM.

Neurotrophin Signaling in Medulloblastoma

Neurotrophins are a family of secreted proteins that act by binding to tropomyosin receptor kinase (Trk) or p75NTR receptors to regulate nervous system development and plasticity. Increasing evidence indicates that neurotrophins and their receptors in cancer cells play a role in tumor growth and resistance to treatment. In this review, we summarize evidence indicating that neurotrophin signaling influences medulloblastoma (MB), the most common type of malignant brain cancer afflicting children. We discuss the potential of neurotrophin receptors as new therapeutic targets for the treatment of MB. Overall, activation of TrkA and TrkC types of receptors seem to promote cell death, whereas TrkB might stimulate MB growth, and TrkB inhibition displays antitumor effects. Importantly, we show analyses of the gene expression profile of neurotrophins and their receptors in MB primary tumors, which indicate, among other findings, that higher levels of NTRK1 or NTRK2 are associated with reduced overall survival (OS) of patients with SHH MB tumors.

Visualizing histopathologic deep learning classification and anomaly detection using nonlinear feature space dimensionality reduction

BACKGROUND

There is growing interest in utilizing artificial intelligence, and particularly deep learning, for computer vision in histopathology. While accumulating studies highlight expert-level performance of convolutional neural networks (CNNs) on focused classification tasks, most studies rely on probability distribution scores with empirically defined cutoff values based on post-hoc analysis. More generalizable tools that allow humans to visualize histology-based deep learning inferences and decision making are scarce.

RESULTS

Here, we leverage t-distributed Stochastic Neighbor Embedding (t-SNE) to reduce dimensionality and depict how CNNs organize histomorphologic information. Unique to our workflow, we develop a quantitative and transparent approach to visualizing classification decisions prior to softmax compression. By discretizing the relationships between classes on the t-SNE plot, we show we can super-impose randomly sampled regions of test images and use their distribution to render statistically-driven classifications. Therefore, in addition to providing intuitive outputs for human review, this visual approach can carry out automated and objective multi-class classifications similar to more traditional and less-transparent categorical probability distribution scores. Importantly, this novel classification approach is driven by a priori statistically defined cutoffs. It therefore serves as a generalizable classification and anomaly detection tool less reliant on post-hoc tuning.

CONCLUSION

Routine incorporation of this convenient approach for quantitative visualization and error reduction in histopathology aims to accelerate early adoption of CNNs into generalized real-world applications where unanticipated and previously untrained classes are often encountered.