Search

Genetics of recurrent PDX

Molecular subtype, MGMT methylation, expressed fusions

Patient characteristics of viable PDX lines

Purpose:Glioblastoma is the most frequent and lethal primary brain tumor. Development of novel therapies relies on the availability of relevant preclinical models. We have established a panel of 96 glioblastoma patient-derived xenografts (PDX) and undertaken its genomic and phenotypic characterization.Experimental Design:PDXs were established from glioblastoma, IDH-wildtype (n = 93), glioblastoma, IDH-mutant (n = 2), diffuse midline glioma, H3 K27M-mutant (n = 1), and both primary (n = 60) and recurrent (n = 34) tumors. Tumor growth rates, histopathology, and treatment response were characterized. Integrated molecular profiling was performed by whole-exome sequencing (WES, n = 83), RNA-sequencing (n = 68), and genome-wide methylation profiling (n = 76). WES data from 24 patient tumors was compared with derivative models.Results:PDXs recapitulate many key phenotypic and molecular features of patient tumors. Orthotopic PDXs show characteristic tumor morphology and invasion patterns, but largely lack microvascular proliferation and necrosis. PDXs capture common and rare molecular drivers, including alterations of TERT, EGFR, PTEN, TP53, BRAF, and IDH1, most at frequencies comparable with human glioblastoma. However, PDGFRA amplification was absent. RNA-sequencing and genome-wide methylation profiling demonstrated broad representation of glioblastoma molecular subtypes. MGMT promoter methylation correlated with increased survival in response to temozolomide. WES of 24 matched patient tumors showed preservation of most genetic driver alterations, including EGFR amplification. However, in four patient–PDX pairs, driver alterations were gained or lost on engraftment, consistent with clonal selection.Conclusions:Our PDX panel captures the molecular heterogeneity of glioblastoma and recapitulates many salient genetic and phenotypic features. All models and genomic data are openly available to investigators.

Frequency of genomic alterations in PDX compared to TCGA

Short tandem repeat genotype and PDX growth characteristics

PDX Response to Standard Therapy

Histologic diagnoses for all tumors implanted (WHO grade II-IV)

Figures S1-S8, Supplementary Methods, Supplementary Bibliography

Purpose: Glioblastoma is the most frequent and lethal primary brain tumor. Development of novel therapies relies on the availability of relevant preclinical models. We have established a panel of 96 glioblastoma patient-derived xenografts (PDX) and undertaken its genomic and phenotypic characterization. Experimental Design: PDXs were established from glioblastoma, IDH-wildtype (n = 93), glioblastoma, IDH-mutant (n = 2), diffuse midline glioma, H3 K27M-mutant (n = 1), and both primary (n = 60) and recurrent (n = 34) tumors. Tumor growth rates, histopathology, and treatment response were characterized. Integrated molecular profiling was performed by whole-exome sequencing (WES, n = 83), RNA-sequencing (n = 68), and genome-wide methylation profiling (n = 76). WES data from 24 patient tumors was compared with derivative models. Results: PDXs recapitulate many key phenotypic and molecular features of patient tumors. Orthotopic PDXs show characteristic tumor morphology and invasion patterns, but largely lack microvascular proliferation and necrosis. PDXs capture common and rare molecular drivers, including alterations of TERT, EGFR, PTEN, TP53, BRAF, and IDH1, most at frequencies comparable with human glioblastoma. However, PDGFRA amplification was absent. RNA-sequencing and genome-wide methylation profiling demonstrated broad representation of glioblastoma molecular subtypes. MGMT promoter methylation correlated with increased survival in response to temozolomide. WES of 24 matched patient tumors showed preservation of most genetic driver alterations, including EGFR amplification. However, in four patient–PDX pairs, driver alterations were gained or lost on engraftment, consistent with clonal selection. Conclusions: Our PDX panel captures the molecular heterogeneity of glioblastoma and recapitulates many salient genetic and phenotypic features. All models and genomic data are openly available to investigators.

Multimodal integration between mass spectrometry imaging (MSI) and radiology-established modalities such as magnetic resonance imaging (MRI) would allow the investigations of key questions in complex biological systems such as the central nervous system. Such integration would provide complementary multiscale data to bridge the gap between molecular and anatomical phenotypes, potentially revealing new insights into molecular mechanisms underlying anatomical pathologies presented on MRI. Automatic coregistration between 3D MSI/MRI is a computationally challenging process due to dimensional complexity, MSI data sparsity, lack of direct spatial-correspondences, and nonlinear tissue deformation. Here, we present a new computational approach based on stochastic neighbor embedding to nonlinearly align 3D MSI to MRI data, identify and reconstruct biologically relevant molecular patterns in 3D, and fuse the MSI datacube to the MRI space. We demonstrate our method using multimodal high-spectral resolution matrix-assisted laser desorption ionization (MALDI) 9.4 T MSI and 7 T in vivo MRI data, acquired from a patient-derived, xenograft mouse brain model of glioblastoma following administration of the EGFR inhibitor drug of Erlotinib. Results show the distribution of some identified molecular ions of the EGFR inhibitor erlotinib, a phosphatidylcholine lipid, and cholesterol, which were reconstructed in 3D and mapped to the MRI space. The registration quality was evaluated on two normal mouse brains using the Dice coefficient for the regions of brainstem, hippocampus, and cortex. The method is generic and can therefore be applied to hyperspectral images from different mass spectrometers and integrated with other established in vivo imaging modalities such as computed tomography (CT) and positron emission tomography (PET).

Dorsolateral prefrontal cortex (DLPFC) is thought to organize items in working memory and this organizational role may also influence long-term memory. To causally test this hypothesized role of DLPFC in long-term memory formation, we used θ-burst noninvasive stimulation (TBS) to modulate DLPFC involvement in a memory task that assessed the influence of active short-term retrieval on later memory. Human subjects viewed three objects on a grid and then either actively retrieved or passively restudied one object’s location after a brief delay. Long-term memory for the other objects was assessed after a delay to evaluate the beneficial role of active short-term retrieval on subsequent memory for the entire set of object locations. We found that DLPFC TBS had no significant effects on short-term memory. In contrast, DLPFC TBS impaired long-term memory selectively in the active-retrieval condition but not in the passive-restudy condition. These findings are consistent with the hypothesized contribution of DLPFC to the organizational processes operative during active short-term retrieval that influence long-term memory, although other regions that were not stimulated could provide similar contributions. Notably, active-retrieval and passive-restudy conditions were intermixed, and therefore nonspecific influences of stimulation were well controlled. These results suggest that DLPFC is causally involved in organizing event information during active retrieval to support coherent long-term memory formation.

BACKGROUND: Failed trials involving targeted therapies face a daunting task of understanding whether the root cause was inadequate targeting, resistance, or insufficient delivery to the tumor. Previous work with a bio-mathematical model has shown the prognostic value of an imaging based invasion metric, D/ρ (mm(2)/yr), which is linearly correlated with the extent of tumor burden beyond the imaging abnormality. As this extent likely impacts the definition of sufficient drug delivery, we investigated whether this metric is able to predict response to ABT-414, an antibody drug conjugate targeting EGFR. METHODS: Preclinical experiments: After initial screening in vitro and in flank, the efficacy of ABT-414 was evaluated for three patient derived cell lines (PDXs), GBM6, GBM12, and GBM39, implanted orthotopically. All three showed strong response to ABT-414 in the previous experiments. In the orthotopic setting, GBM39 was very sensitive to therapy (> 155 days benefit), GBM12 was moderately sensitive (15–30 days benefit), and GBM6 was resistant to therapy (no benefit). Invasion Index: D/ρ was calculated for each cell line based on the original patients’ pre-treatment T1-weighted with Gd contrast and FLAIR MRIs. We then analyzed whether D/ρ was correlated with the observed response to ABT-414. RESULTS: GBM6 had the highest D/ρ = 4.7 (greatest proportion of tumor cells invaded beyond the imaging abnormality), GBM12 was in the middle with D/ρ = 1.7, and GBM39 had the lowest D/ρ = 0.9. This perfectly (inversely) correlates with the response patterns. GBM6 showed no response, GBM12 had a moderate response, and GBM39 showed the greatest response to ABT-414. CONCLUSION: Our results suggest that the imaging based tumor invasion metric, D/ρ, is inversely correlated with tumor response to ABT-414. This supports our hypothesis that drugs with low BBB permeability will be more beneficial for tumors with little tumor burden beyond the imaging abnormality.

ABT-414 is a novel antibody-drug conjugate (ADC) of monomethyl auristatin F (MMAF), a microtubule destabilizing agent, and an anti-EGFR antibody (ABT-806). ABT-414 has demonstrated promising efficacy in clinical trials against newly diagnosed and recurrent glioblastoma (GBM), but little is known about mechanisms of sensitivity and resistance to the drug. The efficacy of ABT-414 was evaluated in vitro across a panel of nine EGFR-amplified GBM patient-derived xenografts (PDXs). Six lines responded to therapy, with the EGFR vIII mutant lines (GBM6, GBM39, GBM76, GBM108) being more sensitive (IC(50) ~0.01 ug/mL) than point mutant (GBM12) or wild-type (GBM84) amplified lines (IC(50) ~0.5 ug/mL). In contrast, GBM26 (point mutant), GBM46 (vII mutant) and GBM59 (vIII mutant) were non-responsive (IC(50) > 10 ug/mL). Further testing revealed that these latter three lines were also resistant to MMAE, a cell-permeable analog of MMAF, which suggests intrinsic resistance to the toxin. In the responsive lines, ABT-414 was further tested in tumors implanted in heterotopic and orthotopic locations using tumor volume measurements and/or bioluminescent imaging (BLI). In flank, GBM6, GBM39, GBM76 and GBM108 were all highly sensitive to therapy, with complete tumor regression observed in all mice extending beyond 200, 250, 130 and 50 days, respectively. Intracranial GBM39 tumors were also highly responsive, with continuous suppression of BLI signal to background, and a survival benefit greater than 155 days. In contrast, GBM6 was resistant to therapy, without detectable impact on tumor growth and a difference in median survival of 5 ± 10 days. GBM76 was intermediately sensitive, with an extension in median survival of 20 ± 10 days. In summary, the majority of EGFR amplified lines tested are sensitive to ABT-414 in vitro and as flank tumors, but efficacy in treatment of orthotopic tumors is more limited. We hypothesize this effect may be related to heterogeneity of drug delivery.