Development and validation of machine learning models for diagnosis and prognosis of cancer by urinary proteomics, based on the FLEMENGHO cohort.

The current study aims to develop and validate machine learning (ML) models for the prediction of cancer status by the non-invasive urinary proteomic in a population-based cohort. In this retrospective study, urinary proteome profiles in 804 cases from the FLEMENGHO cohort were measured by mass spectrometry. After feature selection by LASSO on both clinical variables and urinary proteome profile, benchmark models by clinical variables were built with six different ML algorithms. Proteome-based models and combined models were built and compared with the benchmark models. The models' performance, i.e. area under the curve (AUC) was compared by Delong method. The 95% confidence interval was estimated by the bootstrapping method. The best-performing model was explained by Shapley Additive Explanations (SHAP) method. The predictive role of proteome biomarkers in longitudinal cancer diagnosis was also explored. A clinical model, based on age, blood sugar and blood lipid profile, yielded the best AUC of 0.75 (0.68-0.82), with 0.80 (0.72-0.91) for the proteome model based on 13 selected biomarkers and 0.83 (0.77-0.90) for the combined model (P=0.01 for comparison with clinical model). SHAP on the support vector machine in the combined setting showed that except for age, proteome biomarkers contribute to the final prediction of the model. After adjusting with clinical factors, three proteome biomarkers are independent risk factors for longitudinal cancer development. Urinary proteome profiling, together with fine-tuned machine learning algorithms, demonstrates the predictive potential for cancer diagnosis transparently.

Dynamic 3D morphology of chick embryos and allantois depicted nondestructively by 3.0T clinical magnetic resonance imaging.

Driven by a global trend of applying replace-reduce-refine or 3Rs' guidance for experimental animals in life sciences, chick embryo and particularly allantois with its chorioallantoic membrane have been increasingly utilized to substitute laboratory animals, which call for more extensive and updated knowledge about this novel experimental setup. In this study, being noninvasive, nonionizing, and super-contrasting with high spatiotemporal resolutions, magnetic resonance imaging (MRI) was chosen as an imaging modality for in ovo monitoring morphologic evolution of the chick embryo, allantois, and chorioallantoic membrane longitudinally throughout embryonic day (ED) 1 until ED20. Cooled in 0°C ice bath for 60 min to reduce MRI motion artifacts, 3 chick embryos (n = 60 in total) on each ED were scanned by a clinical 3.0T MRI scanner to demonstrate 3D images of both T2- and T1-weighted imaging (T2WI, T1WI) sequences at axial, sagittal, and coronal slices. The volumes of both the entire chick embryo and allantois were semi-automatically segmented based on intensity-based thresholding and region-growing algorithms. The morphometries or quantified 3D structures were achieved by refined segmentation, and confirmed by histological analyses (one for each ED). After MRI, the rest of chick embryos (n = 40) continued for incubation. The images from ED2 to ED4 could demonstrate the structural changes of latebra, suggesting its transition into a nutrient supplying channel of yolk sac. The allantois could be recognized by MRI, and its relative volumes on each ED revealed an evolving profile peaked on ED12, with a statistically significant difference from those of earlier and later EDs (P < 0.01). The hypointensity of the yolk due to the susceptibility effect of its enriched iron content overshadowed the otherwise hyperintensity of its lipid components. The chick embryos survived prior cooling and MRI till hatching on ED21. The results could be further developed into a 3D MRI atlas of chick embryo. Clinical 3.0T MRI proved effective as a noninvasive approach to study in ovo 3D embryonic development across the full period (ED1-ED20), which can complement the present knowhow for poultry industry and biomedical science.

AI-based MRI auto-segmentation of brain tumor in rodents, a multicenter study.

Automatic segmentation of rodent brain tumor on magnetic resonance imaging (MRI) may facilitate biomedical research. The current study aims to prove the feasibility for automatic segmentation by artificial intelligence (AI), and practicability of AI-assisted segmentation. MRI images, including T2WI, T1WI and CE-T1WI, of brain tumor from 57 WAG/Rij rats in KU Leuven and 46 mice from the cancer imaging archive (TCIA) were collected. A 3D U-Net architecture was adopted for segmentation of tumor bearing brain and brain tumor. After training, these models were tested with both datasets after Gaussian noise addition. Reduction of inter-observer disparity by AI-assisted segmentation was also evaluated. The AI model segmented tumor-bearing brain well for both Leuven and TCIA datasets, with Dice similarity coefficients (DSCs) of 0.87 and 0.85 respectively. After noise addition, the performance remained unchanged when the signal-noise ratio (SNR) was higher than two or eight, respectively. For the segmentation of tumor lesions, AI-based model yielded DSCs of 0.70 and 0.61 for Leuven and TCIA datasets respectively. Similarly, the performance is uncompromised when the SNR was over two and eight respectively. AI-assisted segmentation could significantly reduce the inter-observer disparities and segmentation time in both rats and mice. Both AI models for segmenting brain or tumor lesions could improve inter-observer agreement and therefore contributed to the standardization of the following biomedical studies.

Impact of Blood-Brain Barrier to Delivering a Vascular-Disrupting Agent: Predictive Role of Multiparametric MRI in Rodent Craniofacial Metastasis Models.

Vascular-disrupting agents (VDAs) have shown a preliminary anti-cancer effect in extracranial tumors; however, the therapeutic potential of VDAs in intracranial metastatic lesions remains unclear. Simultaneous intracranial and extracranial tumors were induced by the implantation of rhabdomyosarcoma in 15 WAG/Rij rats. Pre-treatment characterizations were performed at a 3.0 T clinical magnet including a T2 relaxation map, T1 relaxation map, diffusion-weighted imaging (DWI), and perfusion-weighted imaging (PWI). Shortly afterward, a VDA was intravenously given and MRI scans at 1 h, 8 h, and 24 h after treatment were performed. In vivo findings were further confirmed by postmortem angiography and histopathology staining with H&E, Ki67, and CD31. Before VDA treatment, better perfusion (AUC30: 0.067 vs. 0.058, p < 0.05) and AUC300 value (0.193 vs. 0.063, p < 0.001) were observed in extracranial lesions, compared with intracranial lesions. After VDA treatment, more significant and persistent perfusion deficiency measured by PWI (AUC30: 0.067 vs. 0.008, p < 0.0001) and a T1 map (T1 ratio: 0.429 vs. 0.587, p < 0.05) were observed in extracranial tumors, in contrast to the intracranial tumor (AUC30: 0.058 vs. 0.049, p > 0.05, T1 ratio: 0.497 vs. 0.625, p < 0.05). Additionally, significant changes in the T2 value and apparent diffusion coefficient (ADC) value were observed in extracranial lesions, instead of intracranial lesions. Postmortem angiography and pathology showed a significantly larger H&E-stained area of necrosis (86.2% vs. 18.3%, p < 0.0001), lower CD31 level (42.7% vs. 54.3%, p < 0.05), and lower Ki67 level (12.2% vs. 32.3%, p < 0.01) in extracranial tumors, compared with intracranial lesions. The BBB functioned as a barrier against the delivery of VDA into intracranial tumors and multiparametric MRI may predict the efficacy of VDAs on craniofacial tumors.

Sequential Administrations of a Vascular-Disrupting Agent, High-Intensity Focused Ultrasound, and a Radioactively labeled Necrosis Avid Compound for Eradicating Solid Malignancies.

Radical treatment of malignant solid tumors should aim to be less traumatic, precise, and effective. OncoCiDia, as a noninvasive, sequential dual-targeting, small-molecule, broad spectrum anticancer theranostic approach, may fulfill these requirements of solid cancer (Onco) treatment with both tumoricidal (Ci) and diagnostic (Dia) effects. However, it is unlikely to cure patients with cancer, especially those with large and irregular tumors and with tumors residing in certain organs, such as the brain and pancreas, because of insufficient necrosis generation. To amplify ablative efficacy, this shortcoming could be overcome by combining high-intensity focused ultrasound (HIFU) with the use of a vascular-disrupting agent (VDA) and a radioactively labeled necrosis avid compound (NAC), such as 131I-Hypericin (131I-Hyp), which are the first and second targeting drugs used in OncoCiDia. This study proposes the combined use of OncoCiDia and HIFU (Onco-HIFU-CiDia) as a synergistic treatment for malignant tumors to achieve a curative multimodality and multidrug regimen for patients with solid cancers, in accordance with the current trend of cancer patient care.

Towards updated understanding of brain metastasis.

Brain metastasis (BM) is a common complication in cancer patients with advanced disease and attributes to treatment failure and final mortality. Currently there are several therapeutic options available; however these are only suitable for limited subpopulation: surgical resection or radiosurgery for cases with a limited number of lesions, targeted therapies for approximately 18% of patients, and immune checkpoint inhibitors with a response rate of 20-30%. Thus, there is a pressing need for development of novel diagnostic and therapeutic options. This overview article aims to provide research advances in disease model, targeted therapy, blood brain barrier (BBB) opening strategies, imaging and its incorporation with artificial intelligence, external radiotherapy, and internal targeted radionuclide theragnostics. Finally, a distinct type of BM, leptomeningeal metastasis is also covered.

N6-Methyladenosine (m6A)-Related lncRNAs Are Potential Signatures for Predicting Prognosis and Immune Response in Lung Squamous Cell Carcinoma.

Background: Despite increasing understanding of m6A-related lncRNAs in lung cancer, the role of m6A-related lncRNAs in the prognosis and treatment of lung squamous cell carcinoma is poorly understood to date. Thus, the current study aims to elucidate its role and build a model to predict the prognosis of LUSC patients. Materials and Methods: The data of the current study were accessed from the TCGA database. Pearson correlation analysis was performed to identify lncRNAs correlated to m6A. Next, an m6A-related lncRNAs risk model was built using a single factor, least absolute association, selection operator, and multivariate Cox regression analysis. Results: The relevance between 23 m6A genes and 14,056 lncRNAs is shown by Pearson correlation analysis by Sankey diagram. Multivariate Cox regression analysis determined that 11 m6A-lncRNAs show predictive potential in prognosis, which is confirmed by the consistency index, Kaplan-Meier analysis, principal component analysis, and ROC curve. Additionally, the immune analysis showed that the enrichment of immune cells, major histocompatibility complex molecules, and immune checkpoints in the high and low-risk subgroups were markedly disparate, with the high-risk group showing a stronger immune escape ability and a worse response to immunotherapy. Conclusion: In conclusion, the risk model based on m6A-related lncRNAs showed great promise in predicting the prognosis and the efficacy of immunotherapy.

Development and characterization of a chick embryo chorioallantoic membrane (CAM) based platform for evaluation of vasoactive medications.

Among various anti-cancer therapies, tumor vascular disrupting agents (VDAs) play a crucial role, for which their off-targeting effects on normal vessels need also to be investigated. The purpose of this study was to set up an in-ovo platform that combines a laser speckle contrast imaging (LSCI) modality with chick embryo chorioallantoic membrane (CAM) to real-time monitor vascular diameters and perfusion without and with intravascular injection. Two eggshell windows for both observation or measurement and injection were opened. Dynamic blood perfusion images and corresponding statistic graphs were acquired by using a LSCI unit on CAMs from embryo date (ED) 9 to ED15. A dedicated fine needle catheter was made for slow intravascular administration over 30 min with simultaneous LSCI acquisition. To verify the connectivity between CAM vessels and the embryonic circulations in the egg, contrast-enhanced 3D micro computed tomography (µCT), 2D angiography and histology were executed. This platform was successfully established to acquire, quantify and demonstrate vascular and hemodynamic information from the CAM. Chick embryos even with air cell opened remained alive from ED9 to ED15. Through collecting LSCI derived CAM vascular diameter and perfusion parameters, ED12 was determined as the best time window for vasoactive drug studies. A reverse correlation between CAM vessel diameter and blood perfusion rate was found (p < 0.002). Intravascular infusion and simultaneous LSCI acquisition for 30 min in ovo proved feasible. Contrast-enhanced angiography and histomorphology could characterize the connectivity between CAM vasculature and embryonic circulation. This LSCI-CAM platform was proved effective for investigating the in-ovo hemodynamics, which paves the road for further preclinical research on vasoactive medications including VDAs.

Heterogeneity of Synchronous Lung Metastasis Calls for Risk Stratification and Prognostic Classification: Evidence from a Population-Based Database.

The epidemiology and associated potential heterogeneity of synchronous lung metastasis (sLM) have not been reported at a population-based level. Cancer patients with valid information about sLM status in the Surveillance, Epidemiology, and End Results database were enrolled. The prevalence of sLM, with a 95% confidential interval, and median survival of sLM, with interquartile range, were calculated and compared by Chi-square analyses and log-rank tests by primary cancer type and clinicopathological factors. Furthermore, the risk factors of sLM development were identified by multivariate logistic regression. Among 1,672,265 enrolled cases, 3.3% cases were identified with sLM, with a median survival of 7 months. Heterogeneity in prevalence and prognosis in sLM was observed among different primary cancers, with the highest prevalence in main bronchus cancer and best survival in testis cancer. Higher prevalence and poorer prognosis were observed in the older population, male population, African American, patients with lower socioeconomic status, and cases with advanced T stage, N stage, or more malignant pathological characteristics. Race, age, T stage, N stage, metastasis to other sites, insurance status and marital status were associated with sLM development (p < 0.001). The current study highlights the heterogeneity of the prevalence and prognosis in patients with sLM.

Development and characterization of a rat brain metastatic tumor model by multiparametric magnetic resonance imaging and histomorphology.

To facilitate the development of new brain metastasis (BM) treatment, an easy-to-use and clinically relevant animal model with imaging platform is needed. Rhabdomyosarcoma BM was induced in WAG/Rij rats. Post-implantation surveillance and characterizations were systematically performed with multiparametric MRI including 3D T1 and T2 weighted imaging, diffusion-weighted imaging (DWI), T1 and T2 mapping, and perfusion-weighted imaging (PWI), which were validated by postmortem digital radiography (DR), µCT angiography and histopathology. The translational potential was exemplified by the application of a vascular disrupting agent (VDA). BM was successfully induced in most rats of both genders (18/20). Multiparametric MRI revealed significantly higher T2 value, pre-contrast-enhanced (preCE) T1 value, DWI-derived apparent diffusion coefficient (ADC) and CE ratio, but a lower post-contrast-enhanced (postCE) T1 value in BM lesions than in adjacent brain (p < 0.01). PWI showed the dynamic and higher contrast agent uptake in the BM compared with the adjacent brain. DR, µCT and histopathology characterized the BM as hypervascular tumors. After VDA treatment, the BM showed drug-related perfusion changes and partial necrosis as evidenced by anatomical, functional MRI parameters and postmortem findings. The present BM model and imaging modalities represent a feasible and translational platform for developing BM-targeting therapeutics.

Utilisation of Chick Embryo Chorioallantoic Membrane as a Model Platform for Imaging-Navigated Biomedical Research.

The fertilised chick egg and particularly its chorioallantoic membrane (CAM) have drawn continuing interest in biomedicine and bioengineering fields, especially for research on vascular study, cancer, drug screening and development, cell factors, stem cells, etc. This literature review systemically introduces the CAM's structural evolution, functions, vascular features and the circulation system, and cell regulatory factors. It also presents the major and updated applications of the CAM in assays for pharmacokinetics and biodistribution, drug efficacy and toxicology testing/screening in preclinical pharmacological research. The time course of CAM applications for different assays and their advantages and limitations are summarised. Among these applications, two aspects are emphasised: (1) potential utility of the CAM for preclinical studies on vascular-disrupting agents (VDAs), promising for anti-cancer vascular-targeted therapy, and (2) modern imaging technologies, including modalities and their applications for real-time visualisation, monitoring and evaluation of the changes in CAM vasculature as well as the interactions occurring after introducing the tested medical, pharmaceutical and biological agents into the system. The aim of this article is to help those working in the biomedical field to familiarise themselves with the chick embryo CAM as an alternative platform and to utilise it to design and optimise experimental settings for their specific research topics.

Preparation and validation of cyclodextrin-based excipients for radioiodinated hypericin applied in a targeted cancer radiotherapy.

BACKGROUND: Iodine-131 labeled hypericin (131I-Hyp) has been utilized as a necrosis-avid theragnostic tracer in a dual targeting pan-anticancer strategy called OncoCiDia. Widespread use of previously-tested solvent dimethyl sulfoxide (DMSO) is limited by safety concerns. To tackle this, the present study was designed to explore a clinically feasible excipient for the formulation of the hydrophobic 131I-Hyp for intravenous administration. METHOD: Solubility of Hyp in serial solutions of already-approved hydroxypropyl-ß-cyclodextrin (HP-ß-CD) was evaluated by UVspectrophotometry and 50% HP-ß-CD was chosen for further experiments. Two novel HP-ß-CD-based formulations of 131I-Hyp were compared with previous DMSO-based formulation, with regards to necrosis-targetability and biodistribution, by magnetic resonance imaging, single-photon emission computed tomography (SPECT), gamma counting, autoradiography, fluorescence microscopy and histopathology. RESULTS: Hyp solubility was enhanced with increasing HP-ß-CD concentrations. The radiochemical purity of 131I-Hyp was higher than 90% in all formulations. The necrosis-targetability of 131I-Hyp in the novel formulations was confirmed in vivo by SPECT and in vitro by autoradiography, fluorescence microscopy and histopathology. The plasma clearance of radioactivity was faster in the novel formulations. CONCLUSION: The novel 131I-Hyp formulations with HP-ß-CD could be a suitable pharmaceutical excipient for 131I-Hyp for intravenous administration.

Hypothesis: What is the Best We Can Do with Hydroxychloroquine for COVID-19?

There are widespread anecdotal reports of seemingly successful treatment among the early (three to seven days from symptoms) stage coronavirus disease 2019 (COVID-19) patients with the drug hydroxychloroquine (HCQ), and randomized placebo-controlled trials of HCQ in outpatient settings are underway. In this note, we (1) report observational evidence and present scientific reasoning as to why early treatment with HCQ may succeed while treatment later in the disease progression is likely to fail and (2) hypothesize a public health regime under which HCQ may be used to mitigate the impact of the current pandemic.

Pictorial Imaging-Histopathology Correlation in a Rabbit with Hepatic VX2 Tumor Treated by Transarterial Vascular Disrupting Agent Administration.

Cancer vasculature is immature, disorganized and hyperpermeable and can serve as a target for anti-cancer therapies. Vascular disrupting agents (VDAs) are tubulin protein binding and depolymerizing agents that induce rapid tumoral vascular shutdown and subsequent cancer necrosis. However, two clinical problems exist with all VDAs, i.e. 1) incomplete anticancer effect and 2) dose-dependent toxicity. To tackle these problems, in our ongoing research, a novel VDA C118P is applied by transarterial administration of half the intravenous dose in rabbits with implanted VX2 liver tumor to assess its therapeutic efficacy. Nearly complete tumor necrosis was achieved by only a single arterial dose of C118P at 5 mg/kg, which was documented in a representative case by in vivo digital subtraction arteriogram (DSA) and magnetic resonance imaging (MRI), and further confirmed by ex vivo microangiogram and histopathology. This convincing and promising preliminary outcome would warrant further comprehensive studies to explore the potentials of VDAs by transarterial administration either in mono-drug or in combination for management of solid cancers.

Incidence and prognosis of liver metastasis at diagnosis: a pan-cancer population-based study.

Metastasis is a major cause of cancer-related death and liver metastasis (LM) is a distinct type for its relatively good prognosis after timely treatment for selected patients. However, a generalizable estimation of incidence and prognosis of LM is lacking. Cancer patients with known LM status in the Surveillance, Epidemiology and End Results database were enrolled in the present study. The incidence and prognosis of LM were calculated by primary cancer type and clinicopathological factors. Among 1,630,725 cases, 105,329 (6.46%) cases present LM at diagnosis, with a median survival of 4 months. LM presents at diagnosis in 39.96% of pancreatic cancer, 16.00% of colorectal cancer (CRC) and 12.68% of lung cancer. Of all LM cases, 25.58% originated from lung cancer, with 24.76% from CRC and 17.55% from pancreatic cancer. LM originated from small intestine cancer shows the best prognosis (median survival: 30 months), followed by testis cancer (25 months) and breast cancer (15 months). Subgroup analyses demonstrated disparities in incidence and prognosis of LM, with higher incidence and poorer prognosis in the older population, African American, male, and patients with inferior socioeconomic status. The current study provides a generalizable data resource for the epidemiology of LM, which may help tailor screening protocol, design clinical trials and estimate disease burden.

Predicting Clinical Efficacy of Vascular Disrupting Agents in Rodent Models of Primary and Secondary Liver Cancers: An Overview with Imaging-Histopathology Correlation.

Vascular disrupting agents (VDAs) have entered clinical trials for over 15 years. As the leading VDA, combretastatin A4 phosphate (CA4P) has been evaluated in combination with chemotherapy and molecular targeting agents among patients with ovarian cancer, lung cancer and thyroid cancer, but still remains rarely explored in human liver cancers. To overcome tumor residues and regrowth after CA4P monotherapy, a novel dual targeting pan-anticancer theragnostic strategy, i.e., OncoCiDia, has been developed and shown promise previously in secondary liver tumor models. Animal model of primary liver cancer is time consuming to induce, but of value for more closely mimicking human liver cancers in terms of tumor angiogenesis, histopathological heterogeneity, cellular differentiation, tumor components, cancer progression and therapeutic response. Being increasingly adopted in VDA researches, multiparametric magnetic resonance imaging (MRI) provides imaging biomarkers to reflect in vivo tumor responses to drugs. In this article as a chapter of a doctoral thesis, we overview the construction and clinical relevance of primary and secondary liver cancer models in rodents. Target selection for CA4P therapy assisted by enhanced MRI using hepatobiliary contrast agents (CAs), and therapeutic efficacy evaluated by using MRI with a non-specific contrast agent, dynamic contrast enhanced (DCE) imaging, diffusion weighted imaging (DWI) are also described. We then summarize diverse responses among primary hepatocellular carcinomas (HCCs), secondary liver and pancreatic tumors to CA4P, which appeared to be related to tumor size, vascularity, and cellular differentiation. In general, imaging-histopathology correlation studies allow to conclude that CA4P tends to be more effective in secondary liver tumors and in more differentiated HCCs, but less effective in less differentiated HCCs and implanted pancreatic tumor. Notably, cirrhotic liver may be responsive to CA4P as well. All these could be instructive for future clinical trials of VDAs.

DNASE1L3 as an indicator of favorable survival in hepatocellular carcinoma patients following resection.

Hepatocellular carcinoma (HCC) is a common malignancy with a dismal prognosis. It is of great importance to identify biomarkers for the prediction of patients' survival.The mRNA expression level of deoxyribonuclease 1 like 3 (DNASE1L3) and its correlation with survival were accessed in 424 samples from The Cancer Genome Atlas database. Its expression level was confirmed by real-time quantitative polymerase chain reaction and western blotting in 20 pairs of postsurgical specimens. In addition, immunohistochemistry staining of DNASE1L3 was also performed in 113 postoperative samples, using a histochemistry score system. The relationship between patients' survival and DNASE1L3 expression level was evaluated by the Kaplan-Meier method.DNASE1L3 is downregulated in both mRNA and protein levels in HCC tissues, compared with adjacent normal tissues. 52 of 113 HCC specimens showed positive DNASE1L3 protein expression. Patients with positive DNASE1L3 expression had significantly longer overall survival, compared with patients with negative expression (p = 0.023). However, the DNASE1L3 fails to discriminate progression-free survival (p = 0.134). Multivariate COX analysis revealed that positive DNASE1L3 expression and higher differentiation were significantly associated with better overall survival.This study demonstrated that positive DNASE1L3 expression is an independent prognostic factor for better survival in HCC patients following radical resection.

A Model In Vitro Study Using Hypericin: Tumor-Versus Necrosis-Targeting Property and Possible Mechanisms.

Hypericin (Hyp) had been explored as a tumor-seeking agent for years; however, more recent studies showed its necrosis-avidity rather than cancer-seeking property. To further look into this discrepancy, we conducted an in vitro study on Hyp retention in vital and dead cancerous HepG2 and normal LO2 cell lines by measuring the fluorescence intensity and concentration of Hyp in cells. To question the DNA binding theory for its necrosis-avidity, the subcellular distribution of Hyp was also investigated to explore the possible mechanisms of the necrosis avidity. The fluorescence intensity and concentration are significantly higher in dead cells than those in vital cells, and this difference did not differ between HepG2 and LO2 cell lines. Hyp was taken up in vital cells in the early phase and excreted within hours, whereas it was retained in dead cells for more than two days. Confocal microscopy showed that Hyp selectively accumulated in lysosomes rather than cell membrane or nuclei. Hyp showed a necrosis-avid property rather than cancer-targetability. The long-lasting retention of Hyp in dead cells may be associated with halted energy metabolism and/or binding with certain degraded cellular substrates. Necrosis-avidity of Hyp was confirmed, which may be associated with halted energy metabolism in dead LO2 or HepG2 cells.