NCI Workshop on ctDNA in Cancer Treatment and Clinical Care.

Detection of cell-free circulating tumor DNA (ctDNA) from solid tumors is a fast-evolving field with significant potential for improving patient treatment outcomes. The spectrum of applications for ctDNA assays is broad and includes very diverse intended uses that will require different strategies to demonstrate utility. On September 14-15, 2023, the National Cancer Institute held an in-person workshop in Rockville, MD entitled "ctDNA in Cancer Treatment and Clinical Care". The goal of the workshop was to examine what is currently known and what needs to be determined for various ctDNA liquid biopsy use cases related to treatment and management of patients with solid tumors and to explore how the community can best assess the value of ctDNA assays and technology. Additionally, new approaches were presented that may show promise in the future. The information exchanged in this workshop will provide the community with a better understanding of this field and its potential to affect and benefit decision-making in the treatment of patients with solid tumors.

Upregulation of TET2 and Resistance to DNA Methyltransferase (DNMT) Inhibitors in DNMT1-Deleted Cancer Cells.

BACKGROUND: Ten-eleven-translocation (TET) 2 is a member of the TET family of proteins (TET1-3). DNMT1 gene deletion confers resistance to DNA methyltransferase (DNMT) inhibitors in colorectal, breast, and ovarian cancer cells. Currently, the effect of DNMT1 gene status on TET2 phenotype following DNMT inhibitor treatment is unclear in human malignancies. METHODS: Human colorectal carcinoma HCT116 cells (DNMT+/+) and their isogenic DNMT1 knockout (DNMT1-/-) counterpart were treated with DNMT inhibitors. Expression of TET2 and tumor suppressor (p16ink4A and p15ink4B) proteins were examined by Western blot. Apoptosis and CDKN2A promoter demethylation following drug treatment were detected by Annexin-V apoptosis assay and methylation-specific PCR. RESULTS: TET2 expression was robustly increased in DNMT1-/- cells by 0.5 µM and 5 µM decitabine and azacitidine treatment. Augmentation of TET2 expression was accompanied by re-expression of p16ink4A and p15ink4B proteins and CDKN2A promoter demethylation. TET2 upregulation and tumor suppressor re-expression were associated with resistance conferred by DNMT1 deletion. Treatment with 5-aza-4'-thio-2'-deoxycytidine at a low 0.5 µM dose only upregulated TET2 and reduced CDKN2A promoter methylation, and re-expression of p16ink4A in DNMT1-/- cells. DNMT inhibitors showed minimal effects on TET2 upregulation and re-expression of tumor suppressor proteins in cells with intact DNMT1. CONCLUSIONS: DNMT1 gene deletion made cancer cells prone to TET2 upregulation and activation of tumor suppressor expression upon DNMT inhibitor challenge. TET2 augmentation is concomitant with resistance to DNMT inhibitors in a DNMT1-deleted state.

A Pan-Cancer Patient-Derived Xenograft Histology Image Repository with Genomic and Pathologic Annotations Enables Deep Learning Analysis.

Patient-derived xenografts (PDX) model human intra- and intertumoral heterogeneity in the context of the intact tissue of immunocompromised mice. Histologic imaging via hematoxylin and eosin (H&E) staining is routinely performed on PDX samples, which could be harnessed for computational analysis. Prior studies of large clinical H&E image repositories have shown that deep learning analysis can identify intercellular and morphologic signals correlated with disease phenotype and therapeutic response. In this study, we developed an extensive, pan-cancer repository of >1,000 PDX and paired parental tumor H&E images. These images, curated from the PDX Development and Trial Centers Research Network Consortium, had a range of associated genomic and transcriptomic data, clinical metadata, pathologic assessments of cell composition, and, in several cases, detailed pathologic annotations of neoplastic, stromal, and necrotic regions. The amenability of these images to deep learning was highlighted through three applications: (i) development of a classifier for neoplastic, stromal, and necrotic regions; (ii) development of a predictor of xenograft-transplant lymphoproliferative disorder; and (iii) application of a published predictor of microsatellite instability. Together, this PDX Development and Trial Centers Research Network image repository provides a valuable resource for controlled digital pathology analysis, both for the evaluation of technical issues and for the development of computational image-based methods that make clinical predictions based on PDX treatment studies. Significance: A pan-cancer repository of >1,000 patient-derived xenograft hematoxylin and eosin-stained images will facilitate cancer biology investigations through histopathologic analysis and contributes important model system data that expand existing human histology repositories.

The DNA Methyltransferase Inhibitor 5-Aza-4'-thio-2'-Deoxycytidine Induces C>G Transversions and Acute Lymphoid Leukemia Development.

DNA methyltransferase inhibitors (DNMTi), most commonly cytidine analogs, are compounds that decrease 5'-cytosine methylation. DNMTi are used clinically based on the hypothesis that cytosine demethylation will lead to re-expression of tumor suppressor genes. 5-Aza-4'-thio-2'-deoxycytidine (Aza-TdCyd or ATC) is a recently described thiol-substituted DNMTi that has been shown to have anti-tumor activity in solid tumor models. In this study, we investigated the therapeutic potential of ATC in a murine transplantation model of myelodysplastic syndrome. ATC treatment led to the transformation of transplanted wild-type bone marrow nucleated cells into lymphoid leukemia, and healthy mice treated with ATC also developed lymphoid leukemia. Whole-exome sequencing revealed 1,000 acquired mutations, almost all of which were C>G transversions in a specific 5'-NCG-3' context. These mutations involved dozens of genes involved in human lymphoid leukemia, such as Notch1, Pten, Pax5, Trp53, and Nf1. Human cells treated in vitro with ATC showed 1,000 acquired C>G transversions in a similar context. Deletion of Dck, the rate-limiting enzyme for the cytidine salvage pathway, eliminated C>G transversions. Taken together, these findings demonstrate a highly penetrant mutagenic and leukemogenic phenotype associated with ATC. Significance: Treatment with a DNA methyltransferase inhibitor generates a distinct mutation signature and triggers leukemic transformation, which has important implications for the research and clinical applications of these inhibitors.

HTS384 NCI60: The Next Phase of the NCI60 Screen.

The NCI60 human tumor cell line screen has been in operation as a service to the cancer research community for more than 30 years. The screen operated with 96-well plates, a 2-day exposure period to test agents, and following cell fixation, a visible absorbance endpoint by the protein-staining dye sulforhodamine B. In this study, we describe the next phase of this important cancer research tool, the HTS384 NCI60 screen. Although the cell lines remain the same, the updated screen is performed with 384-well plates, a 3-day exposure period to test agents, and a luminescent endpoint to measure cell viability based upon cellular ATP content. In this study, a library of 1,003 FDA-approved and investigational small-molecule anticancer agents was screened by the two NCI60 assays. The datasets were compared with a focus on targeted agents with at least six representatives in the library. For many agents, including inhibitors of EGFR, BRAF, MEK, ERK, and PI3K, the patterns of GI50 values were very similar between the screens with strong correlations between those patterns within the dataset from each screen. However, for some groups of targeted agents, including mTOR, BET bromodomain, and NAMPRTase inhibitors, there were limited or no correlations between the two datasets, although the patterns of GI50 values and correlations between those patterns within each dataset were apparent. Beginning in January 2024, the HTS384 NCI60 screen became the free screening service of the NCI to facilitate drug discovery by the cancer research community. Significance: The new NCI60 cell line screen HTS384 shows robust patterns of response to oncology agents and substantial overlap with the classic screen, providing an updated tool for studying therapeutic agents. See related commentary by Colombo and Corsello, p. 2397.

Assessment of Patient-Derived Xenograft Growth and Antitumor Activity: The NCI PDXNet Consensus Recommendations.

Although patient-derived xenografts (PDX) are commonly used for preclinical modeling in cancer research, a standard approach to in vivo tumor growth analysis and assessment of antitumor activity is lacking, complicating the comparison of different studies and determination of whether a PDX experiment has produced evidence needed to consider a new therapy promising. We present consensus recommendations for assessment of PDX growth and antitumor activity, providing public access to a suite of tools for in vivo growth analyses. We expect that harmonizing PDX study design and analysis and assessing a suite of analytical tools will enhance information exchange and facilitate identification of promising novel therapies and biomarkers for guiding cancer therapy.

The effect of liver dysfunction on the pharmacokinetic disposition of belinostat and its five metabolites in patients with advanced cancers.

Belinostat was approved in 2014 for the treatment of relapsed or refractory peripheral T-cell lymphoma, however, there was insufficient data to recommend a dose in patients with moderate to severe hepatic impairment. The purpose of this analysis was to characterize the pharmacokinetic disposition of belinostat and its five metabolites in patients with advanced cancers and varying degrees of liver dysfunction. A population pharmacokinetic model was therefore developed to describe the parent-metabolite system. The final model was then implemented to assess the effect of liver impairment on each metabolic pathway of belinostat. It was determined that significant pharmacokinetic differences could only be demonstrated in patients with severe hepatic impairment. The final model estimated a 35%-47% reduction in metabolic clearance attributed to UGT1A1/2B7 glucuronidation, CYP2A6/3A4/2C9 metabolism, and ß-oxidation. These hepatic impairment effects reduced between-subject variability by only 5%-8% for their respective parameter, with a large amount of remaining unexplained variability. With further validation, this model can be leveraged to assess the need for dose adjustments in this patient population.

Association of changes in expression of HDAC and SIRT genes after drug treatment with cancer cell line sensitivity to kinase inhibitors.

Histone deacetylases (HDACs) and sirtuins (SIRTs) are important epigenetic regulators of cancer pathways. There is a limited understanding of how transcriptional regulation of their genes is affected by chemotherapeutic agents, and how such transcriptional changes affect tumour sensitivity to drug treatment. We investigated the concerted transcriptional response of HDAC and SIRT genes to 15 approved antitumor agents in the NCI-60 cancer cell line panel. Antitumor agents with diverse mechanisms of action induced upregulation or downregulation of multiple HDAC and SIRT genes. HDAC5 was upregulated by dasatinib and erlotinib in the majority of the cell lines. Tumour cell line sensitivity to kinase inhibitors was associated with upregulation of HDAC5, HDAC1, and several SIRT genes. We confirmed changes in HDAC and SIRT expression in independent datasets. We also experimentally validated the upregulation of HDAC5 mRNA and protein expression by dasatinib in the highly sensitive IGROV1 cell line. HDAC5 was not upregulated in the UACC-257 cell line resistant to dasatinib. The effects of cancer drug treatment on expression of HDAC and SIRT genes may influence chemosensitivity and may need to be considered during chemotherapy.

Pharmacodynamic effects of the PARP inhibitor talazoparib (MDV3800, BMN 673) in patients with BRCA-mutated advanced solid tumors.

PURPOSE: Talazoparib is an inhibitor of the poly (ADP-ribose) polymerase (PARP) family of enzymes and is FDA-approved for patients with (suspected) deleterious germline BRCA1/2-mutated, HER2­negative, locally advanced or metastatic breast cancer. Because knowledge of the pharmacodynamic (PD) effects of talazoparib in patients has been limited to studies of PARP enzymatic activity (PARylation) in peripheral blood mononuclear cells, we developed a study to assess tumoral PD response to talazoparib treatment (NCT01989546). METHODS: We administered single-agent talazoparib (1 mg/day) orally in 28-day cycles to adult patients with advanced solid tumors harboring (suspected) deleterious BRCA1 or BRCA2 mutations. The primary objective was to examine the PD effects of talazoparib; the secondary objective was to determine overall response rate (ORR). Tumor biopsies were mandatory at baseline and post-treatment on day 8 (optional at disease progression). Biopsies were analyzed for PARylation, DNA damage response (γH2AX), and epithelial‒mesenchymal transition. RESULTS: Nine patients enrolled in this trial. Four of six patients (67%) evaluable for the primary PD endpoint exhibited a nuclear γH2AX response on day 8 of treatment, and five of six (83%) also exhibited strong suppression of PARylation. A transition towards a more mesenchymal phenotype was seen in 4 of 6 carcinoma patients, but this biological change did not affect γH2AX or PAR responses. The ORR was 55% with the five partial responses lasting a median of six cycles. CONCLUSION: Intra-tumoral DNA damage response and inhibition of PARP enzymatic activity were confirmed in patients with advanced solid tumors harboring BRCA1/2 mutations after 8 days of talazoparib treatment.

Atezolizumab for Advanced Alveolar Soft Part Sarcoma.

BACKGROUND: Alveolar soft part sarcoma (ASPS) is a rare soft-tissue sarcoma with a poor prognosis and no established therapy. Recently, encouraging responses to immune checkpoint inhibitors have been reported. METHODS: We conducted an investigator-initiated, multicenter, single-group, phase 2 study of the anti-programmed death ligand 1 (PD-L1) agent atezolizumab in adult and pediatric patients with advanced ASPS. Atezolizumab was administered intravenously at a dose of 1200 mg (in patients ≥18 years of age) or 15 mg per kilogram of body weight with a 1200-mg cap (in patients <18 years of age) once every 21 days. Study end points included objective response, duration of response, and progression-free survival according to Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1, as well as pharmacodynamic biomarkers of multistep drug action. RESULTS: A total of 52 patients were evaluated. An objective response was observed in 19 of 52 patients (37%), with 1 complete response and 18 partial responses. The median time to response was 3.6 months (range, 2.1 to 19.1), the median duration of response was 24.7 months (range, 4.1 to 55.8), and the median progression-free survival was 20.8 months. Seven patients took a treatment break after 2 years of treatment, and their responses were maintained through the data-cutoff date. No treatment-related grade 4 or 5 adverse events were recorded. Responses were noted despite variable baseline expression of programmed death 1 and PD-L1. CONCLUSIONS: Atezolizumab was effective at inducing sustained responses in approximately one third of patients with advanced ASPS. (Funded by the National Cancer Institute and others; ClinicalTrials.gov number, NCT03141684.).

Multicellular Complex Tumor Spheroid Response to DNA Repair Inhibitors in Combination with DNA-damaging Drugs.

Multicellular spheroids comprised of malignant cells, endothelial cells, and mesenchymal stem cells served as an in vitro model of human solid tumors to investigate the potentiation of DNA-damaging drugs by pharmacologic modulation of DNA repair pathways. The DNA-damaging drugs, topotecan, trabectedin, and temozolomide were combined with varied inhibitors of DNA damage response enzymes including PARP (olaparib or talazoparib), ATM (ataxia telangiectasia mutated; AZD-1390), ATR (ataxia telangiectasia and Rad3-related protein; berzosertib or elimusertib), and DNA-PK (DNA-dependent protein kinase; nedisertib or VX-984). A range of clinically achievable concentrations were tested up to the clinical Cmax, if known. Mechanistically, the types of DNA damage induced by temozolomide, topotecan, and trabectedin are distinct, which was apparent from the response of spheroids to combinations with various DNA repair inhibitors. Although most combinations resulted in additive cytotoxicity, synergistic activity was observed for temozolomide combined with PARP inhibitors as well as combinations of the ATM inhibitor AZD-1390 with either topotecan or trabectedin. These findings might provide guidance for the selection of anticancer agent combinations worthy of further investigation. Significance: Clinical efficacy of DNA-damaging anticancer drugs can be influenced by the DNA damage response in tumor cells. The potentiation of DNA-damaging drugs by pharmacologic modulation of DNA repair pathways was assessed in multicellular tumor spheroids. Although most combinations demonstrated additive cytotoxicity, synergistic cytotoxicity was observed for several drug combinations.

Targeted Investigational Oncology Agents in the NCI-60: A Phenotypic Systems-based Resource.

The NCI-60 human tumor cell line panel has proved to be a useful tool for the global cancer research community in the search for novel chemotherapeutics. The publicly available cell line characterization and compound screening data from the NCI-60 assay have significantly contributed to the understanding of cellular mechanisms targeted by new oncology agents. Signature sensitivity/resistance patterns generated for a given chemotherapeutic agent against the NCI-60 panel have long served as fingerprint presentations that encompass target information and the mechanism of action associated with the tested agent. We report the establishment of a new public NCI-60 resource based on the cell line screening of a large and growing set of 175 FDA-approved oncology drugs (AOD) plus >825 clinical and investigational oncology agents (IOA), representing a diverse set (>250) of therapeutic targets and mechanisms. This data resource is available to the public (https://ioa.cancer.gov) and includes the raw data from the screening of the IOA and AOD collection along with an extensive set of visualization and analysis tools to allow for comparative study of individual test compounds and multiple compound sets.

The New NCI Precision Medicine Trials.

Basket, umbrella, and platform trial designs (master protocols) have emerged over the last decade to study precision medicine approaches in oncology. First-generation trials like NCI-MATCH (Molecular Analysis for Therapy Choice) have proven the principle that studying targeted therapies on a large scale is feasible both from the laboratory and clinical perspectives. However, single-agent targeted therapies have shown limited ability to control metastatic disease, despite careful matching of drug to target. As such, newer approaches employing combinations of targeted therapy, or targeted therapy with standard therapies, need to be considered. The NCI has recently embarked on three second-generation precision medicine trials to address this need: ComboMATCH, iMATCH, and myeloMATCH. The design of these trials and necessary infrastructure are discussed in the following perspective.

Exogenous DNA enhances DUOX2 expression and function in human pancreatic cancer cells by activating the cGAS-STING signaling pathway.

Pro-inflammatory cytokines upregulate the expression of the H2O2-producing NADPH oxidase dual oxidase 2 (DUOX2)2 which, when elevated, adversely affects survival from pancreatic ductal adenocarcinoma (PDAC). Because the cGAS-STING pathway is known to initiate pro-inflammatory cytokine expression following uptake of exogenous DNA, we examined whether activation of cGAS-STING could play a role in the generation of reactive oxygen species by PDAC cells. Here, we found that a variety of exogenous DNA species markedly increased the production of cGAMP, the phosphorylation of TBK1 and IRF3, and the translocation of phosphorylated IRF3 into the nucleus, leading to a significant, IRF3-dependent enhancement of DUOX2 expression, and a significant flux of H2O2 in PDAC cells. However, unlike the canonical cGAS-STING pathway, DNA-related DUOX2 upregulation was not mediated by NF-κB. Although exogenous IFN-ß significantly increased Stat1/2-associated DUOX2 expression, intracellular IFN-ß signaling that followed cGAMP or DNA exposure did not itself increase DUOX2 levels. Finally, DUOX2 upregulation subsequent to cGAS-STING activation was accompanied by the enhanced, normoxic expression of HIF-1α and VEGF-A as well as DNA double strand cleavage, suggesting that cGAS-STING signaling may support the development of an oxidative, pro-angiogenic microenvironment that could contribute to the inflammation-related genetic instability of pancreatic cancer.

The NCI-MATCH trial: lessons for precision oncology.

The NCI-MATCH (Molecular Analysis for Therapy Choice) trial ( NCT02465060 ) was launched in 2015 as a genomically driven, signal-seeking precision medicine platform trial-largely for patients with treatment-refractory, malignant solid tumors. Having completed in 2023, it remains one of the largest tumor-agnostic, precision oncology trials undertaken to date. Nearly 6,000 patients underwent screening and molecular testing, with a total of 1,593 patients (inclusive of continued accrual from standard next-generation sequencing) being assigned to one of 38 substudies. Each substudy was a phase 2 trial of a therapy matched to a genomic alteration, with a primary endpoint of objective tumor response by RECIST criteria. In this Perspective, we summarize the outcomes of the initial 27 substudies in NCI-MATCH, which met its signal-seeking objective with 7/27 positive substudies (25.9%). We discuss key aspects of the design and operational conduct of the trial, highlighting important lessons for future precision medicine studies.

The Childhood Cancer Data Initiative: Using the Power of Data to Learn From and Improve Outcomes for Every Child and Young Adult With Pediatric Cancer.

Data-driven basic, translational, and clinical research has resulted in improved outcomes for children, adolescents, and young adults (AYAs) with pediatric cancers. However, challenges in sharing data between institutions, particularly in research, prevent addressing substantial unmet needs in children and AYA patients diagnosed with certain pediatric cancers. Systematically collecting and sharing data from every child and AYA can enable greater understanding of pediatric cancers, improve survivorship, and accelerate development of new and more effective therapies. To accomplish this goal, the Childhood Cancer Data Initiative (CCDI) was launched in 2019 at the National Cancer Institute. CCDI is a collaborative community endeavor supported by a 10-year, $50-million (in US dollars) annual federal investment. CCDI aims to learn from every patient diagnosed with a pediatric cancer by designing and building a data ecosystem that facilitates data collection, sharing, and analysis for researchers, clinicians, and patients across the cancer community. For example, CCDI's Molecular Characterization Initiative provides comprehensive clinical molecular characterization for children and AYAs with newly diagnosed cancers. Through these efforts, the CCDI strives to provide clinical benefit to patients and improvements in diagnosis and care through data-focused research support and to build expandable, sustainable data resources and workflows to advance research well past the planned 10 years of the initiative. Importantly, if CCDI demonstrates the success of this model for pediatric cancers, similar approaches can be applied to adults, transforming both clinical research and treatment to improve outcomes for all patients with cancer.

DNA damage, demethylation and anticancer activity of DNA methyltransferase (DNMT) inhibitors.

Role of DNA damage and demethylation on anticancer activity of DNA methyltransferase inhibitors (DNMTi) remains undefined. We report the effects of DNMT1 gene deletion/disruption (DNMT1-/-) on anticancer activity of a class of DNMTi in vitro, in vivo and in human cancers. The gene deletion markedly attenuated cytotoxicity and growth inhibition mediated by decitabine, azacitidine and 5-aza-4'-thio-2'-deoxycytidine (aza-T-dCyd) in colon and breast cancer cells. The drugs induced DNA damage that concurred with DNMT1 inhibition, subsequent G2/M cell-cycle arrest and apoptosis, and upregulated p21 in DNMT1+/+ versus DNMT1-/- status, with aza-T-dCyd the most potent. Tumor growth and DNMT1 were significantly inhibited, and p21 was upmodulated in mice bearing HCT116 DNMT1+/+ xenograft and bladder PDX tumors. DNMT1 gene deletion occurred in ~ 9% human colon cancers and other cancer types at varying degrees. Decitabine and azacitidine demethylated CDKN2A/CDKN2B genes in DNMT1+/+ and DNMT1-/- conditions and increased histone-H3 acetylation with re-expression of p16INK4A/p15INK4B in DNMT1-/- state. Thus, DNMT1 deletion confers resistance to DNMTi, and their anti-cancer activity is determined by DNA damage effects. Patients with DNMT1 gene deletions may not respond to DNMTi treatment.

Redox phospholipidomics discovers pro-ferroptotic death signals in A375 melanoma cells in vitro and in vivo.

Growing cancer cells effectively evade most programs of regulated cell death, particularly apoptosis. This necessitates a search for alternative therapeutic modalities to cause cancer cell's demise, among them - ferroptosis. One of the obstacles to using pro-ferroptotic agents to treat cancer is the lack of adequate biomarkers of ferroptosis. Ferroptosis is accompanied by peroxidation of polyunsaturated species of phosphatidylethanolamine (PE) to hydroperoxy- (-OOH) derivatives, which act as death signals. We demonstrate that RSL3-induced death of A375 melanoma cells in vitro was fully preventable by ferrostatin-1, suggesting their high susceptibility to ferroptosis. Treatment of A375 cells with RSL3 caused a significant accumulation of PE-(18:0/20:4-OOH) and PE-(18:0/22:4-OOH), the biomarkers of ferroptosis, as well as oxidatively truncated products - PE-(18:0/hydroxy-8-oxo-oct-6-enoic acid (HOOA) and PC-(18:0/HOOA). A significant suppressive effect of RSL3 on melanoma growth was observed in vivo (utilizing a xenograft model of inoculation of GFP-labeled A375 cells into immune-deficient athymic nude mice). Redox phospholipidomics revealed elevated levels of 18:0/20:4-OOH in RSL3-treated group vs controls. In addition, PE-(18:0/20:4-OOH) species were identified as major contributors to the separation of control and RSL3-treated groups, with the highest variable importance in projection predictive score. Pearson correlation analysis revealed an association between tumor weight and contents of PE-(18:0/20:4-OOH) (r = -0.505), PE-18:0/HOOA (r = -0.547) and PE 16:0-HOOA (r = -0.503). Thus, LC-MS/MS based redox lipidomics is a sensitive and precise approach for the detection and characterization of phospholipid biomarkers of ferroptosis induced in cancer cells by radio- and chemotherapy.

Data augmentation and multimodal learning for predicting drug response in patient-derived xenografts from gene expressions and histology images.

Patient-derived xenografts (PDXs) are an appealing platform for preclinical drug studies. A primary challenge in modeling drug response prediction (DRP) with PDXs and neural networks (NNs) is the limited number of drug response samples. We investigate multimodal neural network (MM-Net) and data augmentation for DRP in PDXs. The MM-Net learns to predict response using drug descriptors, gene expressions (GE), and histology whole-slide images (WSIs). We explore whether combining WSIs with GE improves predictions as compared with models that use GE alone. We propose two data augmentation methods which allow us training multimodal and unimodal NNs without changing architectures with a single larger dataset: 1) combine single-drug and drug-pair treatments by homogenizing drug representations, and 2) augment drug-pairs which doubles the sample size of all drug-pair samples. Unimodal NNs which use GE are compared to assess the contribution of data augmentation. The NN that uses the original and the augmented drug-pair treatments as well as single-drug treatments outperforms NNs that ignore either the augmented drug-pairs or the single-drug treatments. In assessing the multimodal learning based on the MCC metric, MM-Net outperforms all the baselines. Our results show that data augmentation and integration of histology images with GE can improve prediction performance of drug response in PDXs.

Evaluating the indotecan-neutropenia relationship in patients with solid tumors by population pharmacokinetic modeling and sigmoidal Emax regressions.

PURPOSE: This study aimed at characterizing indotecan population pharmacokinetics and explore the indotecan-neutropenia relationship in patients with solid tumors. METHODS: Population pharmacokinetics were assessed using nonlinear mixed-effects modeling of concentration data from two first-in-human phase 1 trials evaluating different dosing schedules of indotecan. Covariates were assessed in a stepwise manner. Final model qualification included bootstrap simulation, visual and quantitative predictive checks, and goodness-of-fit. A sigmoidal Emax model was developed to describe the relationship between average concentration and maximum percent neutrophil reduction. Simulations at fixed doses were conducted to determine the mean predicted decrease in neutrophil count for each schedule. RESULTS: 518 concentrations from 41 patients supported a three-compartment pharmacokinetic model. Body weight and body surface area accounted for inter-individual variability of central/peripheral distribution volume and intercompartmental clearance, respectively. Estimated typical population values were CL 2.75 L/h, Q3 46.0 L/h, and V3 37.9 L. The estimated value of Q2 for a typical patient (BSA = 1.96 m2) was 17.3 L/h, while V1 and V2 for a typical patient (WT = 80 kg) was 33.9 L and 132 L. The final sigmoidal Emax model estimated that half-maximal ANC reduction occurs at an average concentration of 1416 µg/L and 1041 µg/L for the daily and weekly regimens, respectively. Simulations of the weekly regimen demonstrated lower percent reduction in ANC compared to the daily regimen at equivalent cumulative fixed doses. CONCLUSION: The final PK model adequately describes indotecan population pharmacokinetics. Fixed dosing may be justified based on covariate analysis and the weekly dosing regimen may have a reduced neutropenic effect.