Individualized Ultrasound-Guided Intervention Phantom Development, Fabrication, and Proof of Concept.

Commercial ultrasound vascular phantoms lack the anatomic diversity required for robust pre-clinical interventional device testing. We fabricated individualized phantoms to test an artificial intelligence enabled ultrasound-guided surgical robotic system (AI-GUIDE) which allows novices to cannulate deep vessels. After segmenting vessels on computed tomography scans, vessel cores, bony anatomy, and a mold tailored to the skin contour were 3D-printed. Vessel cores were coated in silicone, surrounded in tissue-mimicking gel tailored for ultrasound and needle insertion, and dissolved with water. One upper arm and four inguinal phantoms were constructed. Operators used AI-GUIDE to deploy needles into phantom vessels. Two groin phantoms were tested due to imaging artifacts in the other two phantoms. Six operators (medical experience: none, 3; 1-5 years, 2; 5+ years, 1) inserted 27 inguinal needles with 81% (22/27) success in a median of 48 seconds. Seven operators performed 24 arm injections, without tuning the AI for arm anatomy, with 71% (17/24) success. After excluding failures due to motor malfunction and a defective needle, success rate was 100% (22/22) in the groin and 85% (17/20) in the arm. Individualized 3D-printed phantoms permit testing of surgical robotics across a large number of operators and different anatomic sites. AI-GUIDE operators rapidly and reliably inserted a needle into target vessels in the upper arm and groin, even without prior medical training. Virtual device trials in individualized 3-D printed phantoms may improve rigor of results and expedite translation.Clinical Relevance- Individualized phantoms enable rigorous and efficient evaluation of interventional devices and reduce the need for animal and human subject testing.

PARP14 inhibition restores PD-1 immune checkpoint inhibitor response following IFNγ-driven acquired resistance in preclinical cancer models.

Resistance mechanisms to immune checkpoint blockade therapy (ICBT) limit its response duration and magnitude. Paradoxically, Interferon γ (IFNγ), a key cytokine for cellular immunity, can promote ICBT resistance. Using syngeneic mouse tumour models, we confirm that chronic IFNγ exposure confers resistance to immunotherapy targeting PD-1 (α-PD-1) in immunocompetent female mice. We observe upregulation of poly-ADP ribosyl polymerase 14 (PARP14) in chronic IFNγ-treated cancer cell models, in patient melanoma with elevated IFNG expression, and in melanoma cell cultures from ICBT-progressing lesions characterised by elevated IFNγ signalling. Effector T cell infiltration is enhanced in tumours derived from cells pre-treated with IFNγ in immunocompetent female mice when PARP14 is pharmacologically inhibited or knocked down, while the presence of regulatory T cells is decreased, leading to restoration of α-PD-1 sensitivity. Finally, we determine that tumours which spontaneously relapse in immunocompetent female mice following α-PD-1 therapy upregulate IFNγ signalling and can also be re-sensitised upon receiving PARP14 inhibitor treatment, establishing PARP14 as an actionable target to reverse IFNγ-driven ICBT resistance.

Noninvasive Monitoring of Simulated Hemorrhage and Whole Blood Resuscitation.

Hemorrhage is the leading cause of preventable death from trauma. Accurate monitoring of hemorrhage and resuscitation can significantly reduce mortality and morbidity but remains a challenge due to the low sensitivity of traditional vital signs in detecting blood loss and possible hemorrhagic shock. Vital signs are not reliable early indicators because of physiological mechanisms that compensate for blood loss and thus do not provide an accurate assessment of volume status. As an alternative, machine learning (ML) algorithms that operate on an arterial blood pressure (ABP) waveform have been shown to provide an effective early indicator. However, these ML approaches lack physiological interpretability. In this paper, we evaluate and compare the performance of ML models trained on nine ABP-derived features that provide physiological insight, using a database of 13 human subjects from a lower-body negative pressure (LBNP) model of progressive central hypovolemia and subsequent progressive restoration to normovolemia (i.e., simulated hemorrhage and whole blood resuscitation). Data were acquired at multiple repressurization rates for each subject to simulate varying resuscitation rates, resulting in 52 total LBNP collections. This work is the first to use a single ABP-based algorithm to monitor both simulated hemorrhage and resuscitation. A gradient-boosted regression tree model trained on only the half-rise to dicrotic notch (HRDN) feature achieved a root-mean-square error (RMSE) of 13%, an R2 of 0.82, and area under the receiver operating characteristic curve of 0.97 for detecting decompensation. This single-feature model's performance compares favorably to previously reported results from more-complex black box machine learning models. This model further provides physiological insight because HRDN represents an approximate measure of the delay between the ABP ejected and reflected wave and therefore is an indication of cardiac and peripheral vascular mechanisms that contribute to the compensatory response to blood loss and replacement.

Inhibiting ERK5 overcomes breast cancer resistance to anti-HER2 therapy by targeting the G1/S cell cycle transition.

Targeting the human epidermal growth factor receptor 2 (HER2) became a landmark in the treatment of HER2-driven breast cancer. Nonetheless, the clinical efficacy of anti-HER2 therapies can be short-lived and a significant proportion of patients ultimately develop metastatic disease and die. One striking consequence of oncogenic activation of HER2 in breast cancer cells is the constitutive activation of the extracellular-regulated protein kinase 5 (ERK5) through its hyperphosphorylation. In this study, we sought to decipher the significance of this unique molecular signature in promoting therapeutic resistance to anti-HER2 agents. We found that a small-molecule inhibitor of ERK5 suppressed the phosphorylation of the retinoblastoma protein (RB) in HER2 positive breast cancer cells. As a result, ERK5 inhibition enhanced the anti-proliferative activity of single-agent anti-HER2 therapy in resistant breast cancer cell lines by causing a G1 cell cycle arrest. Moreover, ERK5 knockdown restored the anti-tumor activity of the anti-HER2 agent lapatinib in human breast cancer xenografts. Taken together, these findings support the therapeutic potential of ERK5 inhibitors to improve the clinical benefit that patients receive from targeted HER2 therapies.

Feature Importance Analysis for Compensatory Reserve to Predict Hemorrhagic Shock.

Hemorrhage is the leading cause of preventable death from trauma. Traditionally, vital signs have been used to detect blood loss and possible hemorrhagic shock. However, vital signs are not sensitive for early detection because of physiological mechanisms that compensate for blood loss. As an alternative, machine learning algorithms that operate on an arterial blood pressure (ABP) waveform acquired via photoplethysmography have been shown to provide an effective early indicator. However, these machine learning approaches lack physiological interpretability. In this paper, we evaluate the importance of nine ABP-derived features that provide physiological insight, using a database of 40 human subjects from a lower-body negative pressure model of progressive central hypovolemia. One feature was found to be considerably more important than any other. That feature, the half-rise to dicrotic notch (HRDN), measures an approximate time delay between the ABP ejected and reflected wave components. This delay is an indication of compensatory mechanisms such as reduced arterial compliance and vasoconstriction. For a scale of 0% to 100%, with 100% representing normovolemia and 0% representing decompensation, linear regression of the HRDN feature results in root-mean-squared error of 16.9%, R2 of 0.72, and an area under the receiver operating curve for detecting decompensation of 0.88. These results are comparable to previously reported results from the more complex black box machine learning models. Clinical Relevance- A single physiologically interpretable feature measured from an arterial blood pressure waveform is shown to be effective in monitoring for blood loss and impending hemorrhagic shock based on data from a human lower-body negative pressure model of progressive central hypolemia.

Transfer Learning for Automated COVID-19 B-Line Classification in Lung Ultrasound.

Lung ultrasound (LUS) as a diagnostic tool is gaining support for its role in the diagnosis and management of COVID-19 and a number of other lung pathologies. B-lines are a predominant feature in COVID-19, however LUS requires a skilled clinician to interpret findings. To facilitate the interpretation, our main objective was to develop automated methods to classify B-lines as pathologic vs. normal. We developed transfer learning models based on ResNet networks to classify B-lines as pathologic (at least 3 B-lines per lung field) vs. normal using COVID-19 LUS data. Assessment of B-line severity on a 0-4 multi-class scale was also explored. For binary B-line classification, at the frame-level, all ResNet models pretrained with ImageNet yielded higher performance than the baseline nonpretrained ResNet-18. Pretrained ResNet-18 has the best Equal Error Rate (EER) of 9.1% vs the baseline of 11.9%. At the clip-level, all pretrained network models resulted in better Cohen's kappa agreement (linear-weighted) and clip score accuracy, with the pretrained ResNet-18 having the best Cohen's kappa of 0.815 [95% CI: 0.804-0.826], and ResNet-101 the best clip scoring accuracy of 93.6%. Similar results were shown for multi-class scoring, where pretrained network models outperformed the baseline model. A class activation map is also presented to guide clinicians in interpreting LUS findings. Future work aims to further improve the multi-class assessment for severity of B-lines with a more diverse LUS dataset.

Oxidative stress from DGAT1 oncoprotein inhibition in melanoma suppresses tumor growth when ROS defenses are also breached.

Dysregulated cellular metabolism is a cancer hallmark for which few druggable oncoprotein targets have been identified. Increased fatty acid (FA) acquisition allows cancer cells to meet their heightened membrane biogenesis, bioenergy, and signaling needs. Excess FAs are toxic to non-transformed cells but surprisingly not to cancer cells. Molecules underlying this cancer adaptation may provide alternative drug targets. Here, we demonstrate that diacylglycerol O-acyltransferase 1 (DGAT1), an enzyme integral to triacylglyceride synthesis and lipid droplet formation, is frequently up-regulated in melanoma, allowing melanoma cells to tolerate excess FA. DGAT1 over-expression alone transforms p53-mutant zebrafish melanocytes and co-operates with oncogenic BRAF or NRAS for more rapid melanoma formation. Antagonism of DGAT1 induces oxidative stress in melanoma cells, which adapt by up-regulating cellular reactive oxygen species defenses. We show that inhibiting both DGAT1 and superoxide dismutase 1 profoundly suppress tumor growth through eliciting intolerable oxidative stress.

Detection of COVID-19 using multimodal data from a wearable device: results from the first TemPredict Study.

Early detection of diseases such as COVID-19 could be a critical tool in reducing disease transmission by helping individuals recognize when they should self-isolate, seek testing, and obtain early medical intervention. Consumer wearable devices that continuously measure physiological metrics hold promise as tools for early illness detection. We gathered daily questionnaire data and physiological data using a consumer wearable (Oura Ring) from 63,153 participants, of whom 704 self-reported possible COVID-19 disease. We selected 73 of these 704 participants with reliable confirmation of COVID-19 by PCR testing and high-quality physiological data for algorithm training to identify onset of COVID-19 using machine learning classification. The algorithm identified COVID-19 an average of 2.75 days before participants sought diagnostic testing with a sensitivity of 82% and specificity of 63%. The receiving operating characteristic (ROC) area under the curve (AUC) was 0.819 (95% CI [0.809, 0.830]). Including continuous temperature yielded an AUC 4.9% higher than without this feature. For further validation, we obtained SARS CoV-2 antibody in a subset of participants and identified 10 additional participants who self-reported COVID-19 disease with antibody confirmation. The algorithm had an overall ROC AUC of 0.819 (95% CI [0.809, 0.830]), with a sensitivity of 90% and specificity of 80% in these additional participants. Finally, we observed substantial variation in accuracy based on age and biological sex. Findings highlight the importance of including temperature assessment, using continuous physiological features for alignment, and including diverse populations in algorithm development to optimize accuracy in COVID-19 detection from wearables.

AI-Enabled, Ultrasound-Guided Handheld Robotic Device for Femoral Vascular Access.

Hemorrhage is a leading cause of trauma death, particularly in prehospital environments when evacuation is delayed. Obtaining central vascular access to a deep artery or vein is important for administration of emergency drugs and analgesics, and rapid replacement of blood volume, as well as invasive sensing and emerging life-saving interventions. However, central access is normally performed by highly experienced critical care physicians in a hospital setting. We developed a handheld AI-enabled interventional device, AI-GUIDE (Artificial Intelligence Guided Ultrasound Interventional Device), capable of directing users with no ultrasound or interventional expertise to catheterize a deep blood vessel, with an initial focus on the femoral vein. AI-GUIDE integrates with widely available commercial portable ultrasound systems and guides a user in ultrasound probe localization, venous puncture-point localization, and needle insertion. The system performs vascular puncture robotically and incorporates a preloaded guidewire to facilitate the Seldinger technique of catheter insertion. Results from tissue-mimicking phantom and porcine studies under normotensive and hypotensive conditions provide evidence of the technique's robustness, with key performance metrics in a live porcine model including: a mean time to acquire femoral vein insertion point of 53 ± 36 s (5 users with varying experience, in 20 trials), a total time to insert catheter of 80 ± 30 s (1 user, in 6 trials), and a mean number of 1.1 (normotensive, 39 trials) and 1.3 (hypotensive, 55 trials) needle insertion attempts (1 user). These performance metrics in a porcine model are consistent with those for experienced medical providers performing central vascular access on humans in a hospital.

The extracellular-regulated protein kinase 5 (ERK5) enhances metastatic burden in triple-negative breast cancer through focal adhesion protein kinase (FAK)-mediated regulation of cell adhesion.

There is overwhelming clinical evidence that the extracellular-regulated protein kinase 5 (ERK5) is significantly dysregulated in human breast cancer. However, there is no definite understanding of the requirement of ERK5 in tumor growth and metastasis due to very limited characterization of the pathway in disease models. In this study, we report that a high level of ERK5 is a predictive marker of metastatic breast cancer. Mechanistically, our in vitro data revealed that ERK5 was critical for maintaining the invasive capability of triple-negative breast cancer (TNBC) cells through focal adhesion protein kinase (FAK) activation. Specifically, we found that phosphorylation of FAK at Tyr397 was controlled by a kinase-independent function of ERK5. Accordingly, silencing ERK5 in mammary tumor grafts impaired FAK phosphorylation at Tyr397 and suppressed TNBC cell metastasis to the lung without preventing tumor growth. Collectively, these results establish a functional relationship between ERK5 and FAK signaling in promoting malignancy. Thus, targeting the oncogenic ERK5-FAK axis represents a promising therapeutic strategy for breast cancer exhibiting aggressive clinical behavior.

Estimating Sedentary Breathing Rate from Chest-Worn Accelerometry From Free-Living Data.

Breathing rate was estimated from chest-worn accelerometry collected from 1,522 servicemembers during training by a wearable physiological monitor. A total of 29,189 hours of training and sleep data were analyzed. The primary purpose of the monitor was to assess thermal-work strain and avoid heat injuries. The monitor design was thus not optimized to estimate breathing rate. Since breathing rate cannot be accurately estimated during periods of high activity, a qualifier was applied to identify sedentary time periods, totaling 8,867 hours. Breathing rate was estimated for a total of 4,179 hours, or 14% of the total collection and 47% of the sedentary total, primarily during periods of sleep. The breathing rate estimation method was compared to an FDA 510(K)-cleared criterion breathing rate sensor (Zephyr, Annapolis MD, USA) in a controlled laboratory experiment, which showed good agreement between the two techniques. Contributions of this paper are to: 1) provide the first analysis of accelerometry-derived breathing rate on free-living data including periods of high activity as well as sleep, along with a qualifier that effectively identifies sedentary periods appropriate for estimating breathing rate; 2) test breathing rate estimation on a data set with a total duration that is more than 60 times longer than that of the largest previously reported study, 3) test breathing rate estimation on data from a physiological monitor that has not been expressly designed for that purpose.

Image Processing Pipeline for Liver Fibrosis Classification Using Ultrasound Shear Wave Elastography.

The purpose of this study was to develop an automated method for classifying liver fibrosis stage ≥F2 based on ultrasound shear wave elastography (SWE) and to assess the system's performance in comparison with a reference manual approach. The reference approach consists of manually selecting a region of interest from each of eight or more SWE images, computing the mean tissue stiffness within each of the regions of interest and computing a resulting stiffness value as the median of the means. The 527-subject database consisted of 5526 SWE images and pathologist-scored biopsies, with data collected from a single system at a single site. The automated method integrates three modules that assess SWE image quality, select a region of interest from each SWE measurement and perform machine learning-based, multi-image SWE classification for fibrosis stage ≥F2. Several classification methods were developed and tested using fivefold cross-validation with training, validation and test sets partitioned by subject. Performance metrics were area under receiver operating characteristic curve (AUROC), specificity at 95% sensitivity and number of SWE images required. The final automated method yielded an AUROC of 0.93 (95% confidence interval: 0.90-0.94) versus 0.69 (95% confidence interval: 0.65-0.72) for the reference method, 71% specificity with 95% sensitivity versus 5% and four images per decision versus eight or more. In conclusion, the automated method reported in this study significantly improved the accuracy for ≥F2 classification of SWE measurements as well as reduced the number of measurements needed, which has the potential to reduce clinical workflow.

Predicting Influenza A Tropism with End-to-End Learning of Deep Networks.

The type of host that a virus can infect, referred to as host specificity or tropism, influences infectivity and thus is important for disease diagnosis, epidemic response, and prevention. Advances in DNA sequencing technology have enabled rapid metagenomic analyses of viruses, but the prediction of virus phenotype from genome sequences is an active area of research. As such, automatic prediction of host tropism from analysis of genomic information is of considerable utility. Previous research has applied machine learning methods to accomplish this task, although deep learning (particularly deep convolutional neural network, CNN) techniques have not yet been applied. These techniques have the ability to learn how to recognize critical hierarchical structures within the genome in a data-driven manner. We designed deep CNN models to identify host tropism for human and avian influenza A viruses based on protein sequences and performed a detailed analysis of the results. Our findings show that deep CNN techniques work as well as existing approaches (with 99% mean accuracy on the binary prediction task) while performing end-to-end learning of the prediction model (without the need to specify handcrafted features). The findings also show that these models, combined with standard principal component analysis, can be used to quantify and visualize viral strain similarity.

Deep Learning-Based Automatic Endometrium Segmentation and Thickness Measurement for 2D Transvaginal Ultrasound.

Endometrial thickness is closely related to gyneco-logical function and is an important biomarker in transvaginal ultrasound (TVUS) examinations for assessing female reproductive health. Manual measurement is time-consuming and subject to high inter- and intra- observer variability. In this paper, we present a fully automated endometrial thickness measurement method using deep learning. Our pipeline consists of: 1) endometrium segmentation using a VGG-based U-Net, and 2) endometrial thickness estimation using medial axis transformation. We conducted experimental studies on 137 2D TVUS cases (74/63 secretory phase/proliferative phase). On a test set of 27 cases/277 images, the segmentation Dice score is 0.83. For thickness measurement, we achieved mean absolute error of 1.23/1.38 mm and root mean squared error of 1.79/1.85 mm on two different test sets. The results are considered well within the clinically acceptable range of ±2 mm. Furthermore, our phase-stratified analysis shows that the measurement variance from the secretory phase is higher than that from the proliferative phase, largely due to the high variability of the endometrium appearance in the secretory phase. Future work will extend our current algorithm toward different clinical outcomes for a broader spectrum of clinical applications.

Statin-induced metabolic reprogramming in head and neck cancer: a biomarker for targeting monocarboxylate transporters.

Prognosis of HPV negative head and neck squamous cell carcinoma (HNSCC) patients remains poor despite surgical and medical advances and inadequacy of predictive and prognostic biomarkers in this type of cancer highlights one of the challenges to successful therapy. Statins, widely used for the treatment of hyperlipidaemia, have been shown to possess anti-tumour effects which were partly attributed to their ability to interfere with metabolic pathways essential in the survival of cancer cells. Here, we have investigated the effect of statins on the metabolic modulation of HNSCC cancers with a vision to predict a personalised anticancer therapy. Although, treatment of tumour-bearing mice with simvastatin did not affect tumour growth, pre-treatment for 2 weeks prior to tumour injection, inhibited tumour growth resulting in strongly increased survival. This was associated with increased expression of the monocarboxylate transporter 1 (MCT1) and a significant reduction in tumour lactate content, suggesting a possible reliance of these tumours on oxidative phosphorylation for survival. Since MCT1 is responsible for the uptake of mitochondrial fuels into the cells, we reasoned that inhibiting it would be beneficial. Interestingly, combination of simvastatin with AZD3965 (MCT1 inhibitor) led to further tumour growth delay as compared to monotherapies, without signs of toxicity. In clinical biopsies, prediagnostic statin therapy was associated with a significantly higher MCT1 expression and was not of prognostic value following conventional chemo-radiotherapy. These findings provide a rationale to investigate the clinical effectiveness of MCT1 inhibition in patients with HNSCC who have been taking lipophilic statins prior to diagnosis.

Objective Liver Fibrosis Estimation from Shear Wave Elastography.

Diffuse liver disease is common, primarily driven by high prevalence of non-alcoholic fatty liver disease (NAFLD). It is currently assessed by liver biopsy to determine fibrosis, often staged as F0 (normal) - F4 (cirrhosis). A noninvasive assessment method will allow a broader population to be monitored longitudinally, facilitating risk stratification and treatment efficacy assessment. Ultrasound shear wave elastography (SWE) is a promising noninvasive technique for measuring tissue stiffness that has been shown to correlate with fibrosis stage. However, this approach has been limited by variability in stiffness measurements. In this work, we developed and evaluated an automated framework, called SWE-Assist, that checks SWE image quality, selects a region of interest (ROI), and classifies the ROI to determine whether the fibrosis stage is at or exceeds F2, which is important for clinical decisionmaking. Our database consists of 3,392 images from 328 cases. Several classifiers, including random forest, support vector machine, and convolutional neural network (CNN)) were evaluated. The best approach utilized a CNN and yielded an area under the receiver operating curve (AUROC) of 0.89, compared to the conventional stiffness only based AUROC of 0.74. Moreover, the new method is based on single image per decision, vs. 10 images per decision for the baseline. A larger dataset is needed to further validate this approach, which has the potential to improve the accuracy and efficiency of non-invasive liver fibrosis staging.

Machine learning for medical ultrasound: status, methods, and future opportunities.

Ultrasound (US) imaging is the most commonly performed cross-sectional diagnostic imaging modality in the practice of medicine. It is low-cost, non-ionizing, portable, and capable of real-time image acquisition and display. US is a rapidly evolving technology with significant challenges and opportunities. Challenges include high inter- and intra-operator variability and limited image quality control. Tremendous opportunities have arisen in the last decade as a result of exponential growth in available computational power coupled with progressive miniaturization of US devices. As US devices become smaller, enhanced computational capability can contribute significantly to decreasing variability through advanced image processing. In this paper, we review leading machine learning (ML) approaches and research directions in US, with an emphasis on recent ML advances. We also present our outlook on future opportunities for ML techniques to further improve clinical workflow and US-based disease diagnosis and characterization.

High content screening of patient-derived cell lines highlights the potential of non-standard chemotherapeutic agents for the treatment of glioblastoma.

BACKGROUND: Glioblastoma (GBM) is the most common primary brain malignancy in adults, yet survival outcomes remain poor. First line treatment is well established, however disease invariably recurs and improving prognosis is challenging. With the aim of personalizing therapy at recurrence, we have established a high content screening (HCS) platform to analyze the sensitivity profile of seven patient-derived cancer stem cell lines to 83 FDA-approved chemotherapy drugs, with and without irradiation. METHODS: Seven cancer stem cell lines were derived from patients with GBM and, along with the established cell line U87-MG, each patient-derived line was cultured in tandem in serum-free conditions as adherent monolayers and three-dimensional neurospheres. Chemotherapeutics were screened at multiple concentrations and cells double-stained to observe their effect on both cell death and proliferation. Sensitivity was classified using high-throughput algorithmic image analysis. RESULTS: Cell line specific drug responses were observed across the seven patient-derived cell lines. Few agents were seen to have radio-sensitizing effects, yet some drug classes showed a marked difference in efficacy between monolayers and neurospheres. In vivo validation of six drugs suggested that cell death readout in a three-dimensional culture scenario is a more physiologically relevant screening model and could be used effectively to assess the chemosensitivity of patient-derived GBM lines. CONCLUSION: The study puts forward a number of non-standard chemotherapeutics that could be useful in the treatment of recurrent GBM, namely mitoxantrone, bortezomib and actinomycin D, whilst demonstrating the potential of HCS to be used for personalized treatment based on the chemosensitivity profile of patient tumor cells.

Novel carbonic anhydrase IX-targeted therapy enhances the anti-tumour effects of cisplatin in small cell lung cancer.

Small cell lung cancer (SCLC) has an extremely poor prognosis and methods of improving chemotherapeutic intervention are much sought after. A promising approach lies in inhibiting the tumour-associated enzyme, carbonic anhydrase IX (CA IX), which supports tumour cell survival. The aim of this study was to assess the potential of CA IX inhibition using 4-(3'-(3″,5″-dimethylphenyl)ureido)phenyl sulfamate (S4), for the treatment of human SCLC alone and in combination with cisplatin chemotherapy. Treating SCLC cell lines (DMS 79 and COR-L24) with 100 µM S4 reduced viability in vitro and enhanced cell death when combined with 7 µM cisplatin, most prominently under hypoxic conditions (0.1% O2 ). When either cell line was grown as a xenograft tumour in nude mice, intraperitoneal injection of 50 mg/kg S4 alone and in combination with 3 mg/kg cisplatin led to significantly reduced tumour growth. Combination therapy was superior to single agents and response was greatly accentuated when administering repeated doses of cisplatin in DMS 79 tumours. The mechanism of therapeutic response was investigated in vitro, where S4 treatment increased apoptosis under hypoxic conditions in both DMS 79 and COR-L24 cells. DMS 79 tumours receiving S4 in vivo also displayed increased apoptosis and necrosis. Combining S4 with cisplatin reduced both the area of hypoxia and CA IX-positive cells within tumours and increased necrosis, suggesting hypoxia-specific targeting. This study presents a novel, targeted approach to improving current SCLC therapy via inhibition of CA IX, which enhances apoptosis and significantly inhibits xenograft tumour growth when administered alone and in combination with cisplatin chemotherapy.