Machine intelligence for radiation science: summary of the Radiation Research Society 67th annual meeting symposium.

The era of high-throughput techniques created big data in the medical field and research disciplines. Machine intelligence (MI) approaches can overcome critical limitations on how those large-scale data sets are processed, analyzed, and interpreted. The 67th Annual Meeting of the Radiation Research Society featured a symposium on MI approaches to highlight recent advancements in the radiation sciences and their clinical applications. This article summarizes three of those presentations regarding recent developments for metadata processing and ontological formalization, data mining for radiation outcomes in pediatric oncology, and imaging in lung cancer.

Treatment-planning approaches to intensity modulated proton therapy and the impact on dose-weighted linear energy transfer.

PURPOSE: We quantified the effect of various forward-based treatment-planning strategies in proton therapy on dose-weighted linear energy transfer (LETd). By maintaining the dosimetric quality at a clinically acceptable level, we aimed to evaluate the differences in LETd among various treatment-planning approaches and their practicality in minimizing biologic uncertainties associated with LETd. METHOD: Eight treatment-planning strategies that are achievable in commercial treatment-planning systems were applied on a cylindrical water phantom and four pediatric brain tumor cases. Each planning strategy was compared to either an opposed lateral plan (phantom study) or original clinical plan (patient study). Deviations in mean and maximum LETd from clinically acceptable dose distributions were compared. RESULTS: In the phantom study, using a range shifter and altering the robust scenarios during optimization had the largest effect on the mean clinical target volume LETd, which was reduced from 4.5 to 3.9 keV/µm in both cases. Variations in the intersection angle between beams had the largest effect on LETd in a ring defined 3 to 5 mm outside the target. When beam intersection angles were reduced from opposed laterals (180°) to 120°, 90°, and 60°, corresponding maximum LETd increased from 7.9 to 8.9, 10.9, and 12.2 keV/µm, respectively. A clear trend in mean and maximum LETd variations in the clinical cases could not be established, though spatial distribution of LETd suggested a strong dependence on patient anatomy and treatment geometry. CONCLUSION: Changes in LETd from treatment-plan setup follow intuitive trends in a controlled phantom experiment. Anatomical and other patient-specific considerations, however, can preclude generalizable strategies in clinical cases. For pediatric cranial radiation therapy, we recommend using opposed lateral treatment fields to treat midline targets.

Image-based data mining applies to data collected from children.

PURPOSE: Image-based data mining (IBDM) is a novel voxel-based method for analyzing radiation dose responses that has been successfully applied in adult data. Because anatomic variability and side effects of interest differ for children compared to adults, we investigated the feasibility of IBDM for pediatric analyses. METHODS: We tested IBDM with CT images and dose distributions collected from 167 children (aged 10 months to 20 years) who received proton radiotherapy for primary brain tumors. We used data from four reference patients to assess IBDM sensitivity to reference selection. We quantified spatial-normalization accuracy via contour distances and deviations of the centers-of-mass of brain substructures. We performed dose comparisons with simplified and modified clinical dose distributions with a simulated effect, assessing their accuracy via sensitivity, positive predictive value (PPV) and Dice similarity coefficient (DSC). RESULTS: Spatial normalizations and dose comparisons were insensitive to reference selection. Normalization discrepancies were small (average contour distance < 2.5 mm, average center-of-mass deviation < 6 mm). Dose comparisons identified differences (p < 0.01) in 81% of simplified and all modified clinical dose distributions. The DSCs for simplified doses were high (peak frequency magnitudes of 0.9-1.0). However, the PPVs and DSCs were low (maximum 0.3 and 0.4, respectively) in the modified clinical tests. CONCLUSIONS: IBDM is feasible for childhood late-effects research. Our findings may inform cohort selection in future studies of pediatric radiotherapy dose responses and facilitate treatment planning to reduce treatment-related toxicities and improve quality of life among childhood cancer survivors.

Pre- and Posttherapy Risk Factors for Vasculopathy in Pediatric Patients With Craniopharyngioma Treated With Surgery and Proton Radiation Therapy.

PURPOSE: Vasculopathy (VAS) is a significant complication associated with radiation therapy in patients treated for brain tumors. We studied the type, location, severity, timing, and resolution of VAS in children with craniopharyngioma treated with proton radiation therapy (PRT) and evaluated predictors of stenosis (STN) using a novel patient and imaging-based modeling approach. METHODS AND MATERIALS: Children with craniopharyngioma (n = 94) were treated with 54 Gy relative biological effectiveness PRT in a clinical trial, NCT01419067. We evaluated VAS type, location, severity, and resolution. VAS events were segmented and related to their location, operative corridor, PRT dose, and vascular territory to facilitate mixed effect logistic regression modeling of spatial predictors of STN events. RESULTS: Forty-five (47.9%) patients had 111 instances of confirmed VAS (pre-PRT n = 37, 33.3%). The median time to post-PRT VAS was 3.41 years (95% confidence interval, 1.86-6.11). STN events were observed post-PRT in 23.4% (n = 22) of patients. Post-PRT VAS was detected by cerebral angiogram in 9.6% (n = 9), severe in 4.3% (n = 4), and compensated on perfusion in 2.1% (n = 2). Revascularization was required for 5 (5.3%) patients. Postsurgical, pre-PRT VAS, and PRT dose to unperturbed vessels were predictive of STN. The effect of PRT on STN was negligible within the surgical corridor. CONCLUSIONS: VAS often precedes PRT and was the strongest predictor of post-PRT STN. The adverse effect of PRT on STN was only apparent in unperturbed vasculature beyond the operative corridor.

Acknowledging and overcoming barriers to entry into radiation science for women.

PURPOSE: Radiation science is a unique field that brings together various disciplines to understand nature, develop new technologies, and cure diseases. Our field is a prime example of advancement through a diverse pool of competencies. Similarly, studies show that the power of diversity requires proportionate representation of sex and gender, minorities, or other groups. Nevertheless, women are still underrepresented in the radiation sciences, although disparities and underlying mechanisms were first described decades ago. This review summarizes barriers to entry and retention and suggests strategies for overcoming disparities in our field. We also highlight a concerted effort by young professionals to promote the underrepresented and underserved within the radiation science community. CONCLUSION: The radiation science community should avoid losing diverse perspectives among its ranks due to sex bias or gender disparity among others. Through targeted efforts, we can cultivate change and harness the talent of researchers, practitioners, and other professionals for the benefit of scientific progress, health-care improvement, and societal advancement overall.

Generalized approach for radiotherapy treatment planning by optimizing projected health outcome: preliminary results for prostate radiotherapy patients.

Research in cancer care increasingly focuses on survivorship issues, e.g. managing disease- and treatment-related morbidity and mortality occurring during and after treatment. This necessitates innovative approaches that consider treatment side effects in addition to tumor cure. Current treatment-planning methods rely on constrained iterative optimization of dose distributions as a surrogate for health outcomes. The goal of this study was to develop a generally applicable method to directly optimize projected health outcomes. We developed an outcome-based objective function to guide selection of the number, angle, and relative fluence weight of photon and proton radiotherapy beams in a sample of ten prostate-cancer patients by optimizing the projected health outcome. We tested whether outcome-optimized radiotherapy (OORT) improved the projected longitudinal outcome compared to dose-optimized radiotherapy (DORT) first for a statistically significant majority of patients, then for each individual patient. We assessed whether the results were influenced by the selection of treatment modality, late-risk model, or host factors. The results of this study revealed that OORT was superior to DORT. Namely, OORT maintained or improved the projected health outcome of photon- and proton-therapy treatment plans for all ten patients compared to DORT. Furthermore, the results were qualitatively similar across three treatment modalities, six late-risk models, and 10 patients. The major finding of this work was that it is feasible to directly optimize the longitudinal (i.e. long- and short-term) health outcomes associated with the total (i.e. therapeutic and stray) absorbed dose in all of the tissues (i.e. healthy and diseased) in individual patients. This approach enables consideration of arbitrary treatment factors, host factors, health endpoints, and times of relevance to cancer survivorship. It also provides a simpler, more direct approach to realizing the full beneficial potential of cancer radiotherapy.

Proton therapy delivery method affects dose-averaged linear energy transfer in patients.

The dosimetric advantages of proton therapy have led to its rapid proliferation in recent decades. This has been accompanied by a shift in technology from older units that deliver protons by passive scattering (PS) to newer units that increasingly use pencil-beam scanning (PBS). The biologic effectiveness of proton physical dose purportedly rises with increasing dose-weighted average linear energy transfer (LETD). The objective of this study was to determine the extent to which proton delivery methods affect LETD. We calculated LETDfrom simple, dosimetrically matched, and clinical treatment plans with TOPAS Monte-Carlo transport code. Simple treatment plans comprised single fields of PS and PBS protons in a water phantom. We performed simulations of matched and clinical treatment plans by using the treatment and anatomic data obtained from a cohort of children with craniopharyngioma who previously received PS or PBS proton therapy. We compared the distributions of LETDfrom PS and PBS delivery methods in clinically relevant ROIs. Wilcoxon signed-rank tests comparing single fields in water revealed that the LETDvalues from PBS were significantly greater than those from PS inside and outside the targeted volume (p < 0.01). Statistical tests comparing LETD-volume histograms from matched and clinical treatment plans showed that LETDwas generally greater for PBS treatment plans than for PS treatment plans (p < 0.05). In conclusion, the proton delivery method affects LETDboth inside and outside of the target volume. These findings suggest that PBS is more biologically effective than PS. Given the rapid expansion of PBS proton therapy, future studies are needed to confirm the applicability of treatment evaluation methods developed for PS proton therapy to those for modern PBS treatments to ensure their safety and effectiveness for the growing population of patients receiving proton therapy. This study uses data from two clinical trials: NCT01419067 and NCT02792582.

Justification and optimization of radiation exposures: a new framework to aggregate arbitrary detriments and benefits.

Myriad radiation effects, including benefits and detriments, complicate justifying and optimizing radiation exposures. The purpose of this study was to develop a comprehensive conceptual framework and corresponding quantitative methods to aggregate the detriments and benefits of radiation exposures to individuals, groups, and populations. In this study, concepts from the ICRP for low dose were integrated with clinical techniques focused on high dose to develop a comprehensive figure of merit (FOM) that takes into account arbitrary host- and exposure-related factors, endpoints, and time points. The study built on existing methods with three new capabilities: application to individuals, groups, and populations; extension to arbitrary numbers and types of endpoints; and inclusion of limitation, where relevant. The FOM was applied to three illustrative exposure situations: emergency response, diagnostic imaging, and cancer radiotherapy, to evaluate its utility in diverse settings. The example application to radiation protection revealed the FOM's utility in optimizing the benefits and risks to a population while keeping individual exposures below applicable regulatory limits. Examples in diagnostic imaging and cancer radiotherapy demonstrated the FOM's utility for guiding population- and patient-specific decisions in medical applications. The major finding of this work is that it is possible to quantitatively combine the benefits and detriments of any radiation exposure situation involving an individual or population to perform cost-effectiveness analyses using the ICRP key principles of radiation protection. This FOM fills a chronic gap in the application of radiation-protection theory, i.e., limitations of generalized frameworks to algorithmically justify and optimize radiation exposures. This new framework potentially enhances objective optimization and justification, especially in complex exposure situations.

Automation of Monte Carlo-based treatment plan verification for proton therapy.

PURPOSE: Independent calculations of proton therapy plans are an important quality control procedure in treatment planning. When using custom Monte Carlo (MC) models of the beamline, deploying the calculations can be laborious, time consuming, and require in-depth knowledge of the computational environment. We developed an automated framework to remove these barriers and integrate our MC model into the clinical workflow. MATERIALS AND METHODS: The Eclipse Scripting Application Programming Interface was used to initiate the automation process. A series of MATLAB scripts were then used for preprocessing of input data and postprocessing of results. Additional scripts were used to monitor the calculation process and appropriately deploy calculations to an institutional high-performance computing facility. The automated framework and beamline models were validated against 160 patient specific QA measurements from an ionization chamber array and using a ±3%/3 mm gamma criteria. RESULTS: The automation reduced the human-hours required to initiate and run a calculation to 1-2 min without leaving the treatment planning system environment. Validation comparisons had an average passing rate of 99.4% and were performed at depths ranging from 1 to 15 cm. CONCLUSION: An automated framework for running MC calculations was developed which enables the calculation of dose and linear energy transfer within a clinically relevant workflow and timeline. The models and framework were validated against patient specific QA measurements and exhibited excellent agreement. Before this implementation, execution was prohibitively complex for an untrained individual and its use restricted to a research environment.

Method to quickly and accurately calculate absorbed dose from therapeutic and stray photon exposures throughout the entire body in individual patients.

PURPOSE: Photon radiotherapy techniques typically devote considerable attention to limiting the exposure of healthy tissues outside of the target volume. Numerous studies have shown, however, that commercial treatment planning systems (TPSs) significantly underestimate the absorbed dose outside of the treatment field. The purpose of this study was to test the feasibility of quickly and accurately calculating the total absorbed dose to the whole body from photon radiotherapy in individual patients. METHODS: We created an extended TPS by implementing a physics-based analytical model for the absorbed dose from stray photons during photon therapy into a research TPS. We configured and validated the extended TPS using measurements of 6- and 15-MV photon beams in water-box and anthropomorphic phantoms. We characterized the additional computation time required for therapeutic and stray dose calculations in a 44 × 30 × 180 cm3 water-box phantom. RESULTS: The extended TPS achieved superior dosimetric accuracy compared to the research TPS in both water and anthropomorphic phantoms, especially outside of the primary treatment field. In the anthropomorphic phantom, the extended TPS increased the generalized gamma index passing rate by a factor of 10 and decreased the median dosimetric discrepancy in the out-of-field region by a factor of 26. The extended TPS achieved an average discrepancy <1% in and near the treatment field and <1 mGy/Gy far from the treatment field in the anthropomorphic phantom. Characterization of computation time revealed that on average, the extended TPS only required 7% longer than the research TPS to calculate the total absorbed dose. CONCLUSIONS: The results of this work suggest that it is feasible to quickly and accurately calculate whole-body doses inside and outside of the therapeutic treatment field in individual patients on a routine basis using physics-based analytical dose models. This additional capability enables a more personalized approach to minimizing the risk of radiogenic late effects, such as second cancer and cardiac toxicity, as part of the treatment planning process.

A physics-based analytical model of absorbed dose from primary, leakage, and scattered photons from megavoltage radiotherapy with MLCs.

A burgeoning population of cancer survivors is at risk of late health effects following radiation therapy including second cancers, with the majority of these cancers occurring outside of the treatment volume of the primary cancer. Commercial radiotherapy treatment planning systems underestimate the out-of-field dose. Previous analytical models of out-of-field dose have assumed radial symmetry and/or approximated the dimensions of collimators as semi-infinite planes. The purpose of this work was to develop a physics-based analytical model of total absorbed dose from primary, scattered, and leakage radiation for square fields from a 6 MV beam at any arbitrary point in a phantom, including in-plane, cross-plane, and out-of-plane locations. The model includes the essential physics of radiation transport through beam-limiting-devices including rounded edges of MLC leaves. The model agreed well with measurements and Monte Carlo simulations of absorbed dose in a water-box phantom and was validated for field-sizes ranging from 2 [Formula: see text] 2 to 20 [Formula: see text] 20 cm2. The signed and unsigned average percent differences were [Formula: see text] and 15.9%, respectively, for all points and field-sizes considered. An extended gamma index analysis reveals a 92% pass rate with criteria of 3 mm distance-to-agreement, 3% relative dose difference in-field, and 3 mGy Gy-1 absolute dose difference out-of-field. The average wall-clock time to calculate dose to one million points was 3.3 min. These results suggest that it is feasible to calculate absorbed dose from both therapeutic and stray radiation using physics-based analytical models with good accuracy, thus overcoming a major obstacle to routinely assessing exposures. Additionally, this work demonstrates the importance of relaxing certain simplifications such as assuming a radially symmetric stray-dose distribution. Because the model is physics-based and may be reconfigured according to the dimensions of beam-limiting-devices, adapting it to other treatment units should be straight forward. Uses for such a model include clinical and research applications, such as clinical trials and epidemiological studies.

An objective method to evaluate radiation dose distributions varying by three orders of magnitude.

PURPOSE: Modern radiotherapy practices typically report the absorbed dose (D) within the 5% relative isodose volume (i.e., the therapeutic dose region) to an accuracy of 3%-5%. Gamma-index analysis, the most commonly used method to evaluate dosimetric accuracy, has low sensitivity to discrepancies that occur outside of this region. The objective of this study was to develop an evaluation method with high sensitivity across dose distributions spanning three orders of magnitude. METHODS: We generalized the gamma index to include an additional criterion, the absolute absorbed dose difference, specifically for the low-dose region (i.e., D ≤ 5%). We also proposed a method to objectively select the appropriate magnitudes for relative-dose-difference, absolute-dose-difference, and distance-to-agreement criteria. We demonstrated the generalized gamma-index method by first finding the appropriate generalized gamma-index agreement criteria at an interval of specified passing rates. Next, we used the generalized gamma index to evaluate one-, two-, and three-dimensional absorbed dose distributions in a water-box phantom and voxelized patient geometry. RESULTS: Generalized gamma-index passing rates for one-, two-, and three-dimensional dose distributions were 55.4%, 44.5%, and 8.9%, respectively. Traditional gamma-index passing rates were 100%, 97.8%, and 96.4%, respectively. These results reveal that the generalized method has adequate sensitivity in all regions (i.e., therapeutic and low dose). Additionally, the algorithmic determination of triplets of agreement criteria revealed that they are strong functions of the specified passing rate. CONCLUSIONS: The major finding of this work is that the proposed method provides an objective evaluation of the agreement of dose distributions spanning three orders of magnitude. In particular, this generalized method correctly characterized dosimetric agreement in the low-dose region, which was not possible by traditional methods. The proposed algorithmic selection of agreement criteria decreased subjectivity and requirements of user judgment and skill. This method could find utility in a variety of applications including dose-algorithm development and translation.

Common basis for orofacial clefting and cortical interneuronopathy.

Orofacial clefts (OFCs) of the lip and/or palate are among the most common human birth defects. Current treatment strategies focus on functional and cosmetic repair but even when this care is available, individuals born with OFCs are at high risk for persistent neurobehavioral problems. In addition to learning disabilities and reduced academic achievement, recent evidence associates OFCs with elevated risk for a constellation of psychiatric outcomes including anxiety disorders, autism spectrum disorder, and schizophrenia. The relationship between these outcomes and OFCs is poorly understood and controversial. Recent neuroimaging studies in humans and mice demonstrate subtle morphological brain abnormalities that co-occur with OFCs but specific molecular and cellular mechanisms have not been investigated. Here, we provide the first evidence directly linking OFC pathogenesis to abnormal development of GABAergic cortical interneurons (cINs). Lineage tracing revealed that the structures that form the upper lip and palate develop in molecular synchrony and spatiotemporal proximity to cINs, suggesting these populations may have shared sensitivity to genetic and/or teratogenic insult. Examination of cIN development in a mouse model of nonsyndromic OFCs revealed significant disruptions in cIN proliferation and migration, culminating in misspecification of the somatostatin-expressing subgroup. These findings reveal a unified developmental basis for orofacial clefting and disrupted cIN development, and may explain the significant overlap in neurobehavioral and psychiatric outcomes associated with OFCs and cIN dysfunction. This emerging mechanistic understanding for increased prevalence of adverse neurobehavioral outcomes in OFC patients is the entry-point for developing evidence-based therapies to improve patient outcomes.

Sonic hedgehog regulation of Foxf2 promotes cranial neural crest mesenchyme proliferation and is disrupted in cleft lip morphogenesis.

Cleft lip is one of the most common human birth defects, yet our understanding of the mechanisms that regulate lip morphogenesis is limited. Here, we show in mice that sonic hedgehog (Shh)-induced proliferation of cranial neural crest cell (cNCC) mesenchyme is required for upper lip closure. Gene expression profiling revealed a subset of Forkhead box (Fox) genes that are regulated by Shh signaling during lip morphogenesis. During cleft pathogenesis, reduced proliferation in the medial nasal process mesenchyme paralleled the domain of reduced Foxf2 and Gli1 expression. SHH ligand induction of Foxf2 expression was dependent upon Shh pathway effectors in cNCCs, while a functional GLI-binding site was identified downstream of Foxf2 Consistent with the cellular mechanism demonstrated for cleft lip pathogenesis, we found that either SHH ligand addition or FOXF2 overexpression is sufficient to induce cNCC proliferation. Finally, analysis of a large multi-ethnic human population with cleft lip identified clusters of single-nucleotide polymorphisms in FOXF2 These data suggest that direct targeting of Foxf2 by Shh signaling drives cNCC mesenchyme proliferation during upper lip morphogenesis, and that disruption of this sequence results in cleft lip.

A descriptive and broadly applicable model of therapeutic and stray absorbed dose from 6 to 25 MV photon beams.

PURPOSE: To develop a simple model of therapeutic and stray absorbed dose for a variety of treatment machines and techniques without relying on proprietary machine-specific parameters. METHODS: Dosimetry measurements conducted in this study and from the literature were used to develop an analytical model of absorbed dose from a variety of treatment machines and techniques in the 6 to 25 MV interval. A modified one-dimensional gamma-index analysis was performed to evaluate dosimetric accuracy of the model on an independent dataset consisting of measured dose profiles from seven treatment units spanning four manufacturers. RESULTS: The average difference between the calculated and measured absorbed dose values was 9.9% for those datasets on which the model was trained. Additionally, these results indicate that the model can provide accurate calculations of both therapeutic and stray radiation dose from a wide variety of radiotherapy units and techniques. CONCLUSIONS: We have developed a simple analytical model of absorbed dose from external beam radiotherapy treatments in the 6 to 25 MV beam energy range. The model has been tested on measured data from multiple treatment machines and techniques, and is broadly applicable to contemporary external beam radiation therapy.

Gene-environment interactions in cortical interneuron development and dysfunction: A review of preclinical studies.

Cortical interneurons (cINs) are a diverse group of locally projecting neurons essential to the organization and regulation of neural networks. Though they comprise only ∼20% of neurons in the neocortex, their dynamic modulation of cortical activity is requisite for normal cognition and underlies multiple aspects of learning and memory. While displaying significant morphological, molecular, and electrophysiological variability, cINs collectively function to maintain the excitatory-inhibitory balance in the cortex by dampening hyperexcitability and synchronizing activity of projection neurons, primarily through use of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA). Disruption of the excitatory-inhibitory balance is a common pathophysiological feature of multiple seizure and neuropsychiatric disorders, including epilepsy, schizophrenia, and autism. While most studies have focused on genetic disruption of cIN development in these conditions, emerging evidence indicates that cIN development is exquisitely sensitive to teratogenic disruption. Here, we review key aspects of cIN development, including specification, migration, and integration into neural circuits. Additionally, we examine the mechanisms by which prenatal exposure to common chemical and environmental agents disrupt these events in preclinical models. Understanding how genetic and environmental factors interact to disrupt cIN development and function has tremendous potential to advance prevention and treatment of prevalent seizure and neuropsychiatric illnesses.

Gli2 gene-environment interactions contribute to the etiological complexity of holoprosencephaly: evidence from a mouse model.

Holoprosencephaly (HPE) is a common and severe human developmental abnormality marked by malformations of the forebrain and face. Although several genetic mutations have been linked to HPE, phenotypic outcomes range dramatically, and most cases cannot be attributed to a specific cause. Gene-environment interaction has been invoked as a premise to explain the etiological complexity of HPE, but identification of interacting factors has been extremely limited. Here, we demonstrate that mutations in Gli2, which encodes a Hedgehog pathway transcription factor, can cause or predispose to HPE depending upon gene dosage. On the C57BL/6J background, homozygous GLI2 loss of function results in the characteristic brain and facial features seen in severe human HPE, including midfacial hypoplasia, hypotelorism and medial forebrain deficiency with loss of ventral neurospecification. Although normally indistinguishable from wild-type littermates, we demonstrate that mice with single-allele Gli2 mutations exhibit increased penetrance and severity of HPE in response to low-dose teratogen exposure. This genetic predisposition is associated with a Gli2 dosage-dependent attenuation of Hedgehog ligand responsiveness at the cellular level. In addition to revealing a causative role for GLI2 in HPE genesis, these studies demonstrate a mechanism by which normally silent genetic and environmental factors can interact to produce severe outcomes. Taken together, these findings provide a framework for the understanding of the extreme phenotypic variability observed in humans carrying GLI2 mutations and a paradigm for reducing the incidence of this morbid birth defect.

Definition of critical periods for Hedgehog pathway antagonist-induced holoprosencephaly, cleft lip, and cleft palate.

The Hedgehog (Hh) signaling pathway mediates multiple spatiotemporally-specific aspects of brain and face development. Genetic and chemical disruptions of the pathway are known to result in an array of structural malformations, including holoprosencephaly (HPE), clefts of the lip with or without cleft palate (CL/P), and clefts of the secondary palate only (CPO). Here, we examined patterns of dysmorphology caused by acute, stage-specific Hh signaling inhibition. Timed-pregnant wildtype C57BL/6J mice were administered a single dose of the potent pathway antagonist vismodegib at discrete time points between gestational day (GD) 7.0 and 10.0, an interval approximately corresponding to the 15th to 24th days of human gestation. The resultant pattern of facial and brain dysmorphology was dependent upon stage of exposure. Insult between GD7.0 and GD8.25 resulted in HPE, with peak incidence following exposure at GD7.5. Unilateral clefts of the lip extending into the primary palate were also observed, with peak incidence following exposure at GD8.875. Insult between GD9.0 and GD10.0 resulted in CPO and forelimb abnormalities. We have previously demonstrated that Hh antagonist-induced cleft lip results from deficiency of the medial nasal process and show here that CPO is associated with reduced growth of the maxillary-derived palatal shelves. By defining the critical periods for the induction of HPE, CL/P, and CPO with fine temporal resolution, these results provide a mechanism by which Hh pathway disruption can result in "non-syndromic" orofacial clefting, or HPE with or without co-occurring clefts. This study also establishes a novel and tractable mouse model of human craniofacial malformations using a single dose of a commercially available and pathway-specific drug.