A self-powered theranostic DNA nanodevice for amplification of both intracellular microRNA imaging and photodynamic therapy.

While numerous methods exist for diagnosing tumors through the detection of miRNA within tumor cells, few can simultaneously achieve both tumor diagnosis and treatment. In this study, a novel graphene oxide (GO)-based DNA nanodevice (DND), initiated by miRNA, was developed for fluorescence signal amplification imaging and photodynamic therapy in tumor cells. After entering the cells, tumor-associated miRNA drives DND to Catalyzed hairpin self-assembly (CHA). The CHA reaction generated a multitude of DNA Y-type structures, resulting in a substantial amplification of Ce6 fluorescence release and the generation of numerous singlet oxygen (1O2) species induced by laser irradiation, consequently inducing cell apoptosis. In solution, DND exhibited high selectivity and sensitivity to miRNA-21, with a detection limit of 11.47 pM. Furthermore, DND discriminated between normal and tumor cells via fluorescence imaging and specifically generated O21 species in tumor cells upon laser irradiation, resulting in tumor cells apoptosis. The DND offer a new approach for the early diagnosis and timely treatment of malignant tumors.

Emerging role of PES1 in disease: A promising therapeutic target?

Pescadillo ribosomal biogenesis factor 1 (PES1), a nucleolar protein initially identified in zebrafish, plays an important role in embryonic development and ribosomal biogenesis. Notably, PES1 has been found to be overexpressed in a number of cancer types, where it contributes to tumorigenesis and cancer progression by promoting cell proliferation, suppressing cellular senescence, modulating the tumor microenvironment (TME) and promoting drug resistance in cancer cells. Moreover, recent emerging evidence suggests that PES1 expression is significantly elevated in the livers of Type 2 diabetes mellitus (T2DM) and obese patients, indicating its involvement in the pathogenesis of metabolic diseases through lipid metabolism regulation. In this review, we present the structural characteristics and biological functions of PES1, as well as complexes in which PES1 participates. Furthermore, we comprehensively summarize the multifaceted role of PES1 in various diseases and the latest insights into its underlying molecular mechanisms. Finally, we discuss the potential clinical translational perspectives of targeting PES1, highlighting its promising as a therapeutic intervention and treatment target.

Tumor microenvironment ameliorative and adaptive nanoparticles with photothermal-to-photodynamic switch for cancer phototherapy.

The notorious tumor microenvironment (TME) usually becomes more deteriorative during phototherapeutic progress that hampers the antitumor efficacy. To overcome this issue, we herein report the ameliorative and adaptive nanoparticles (TPASIC-PFH@PLGA NPs) that simultaneously reverse hypoxia TME and switch photoactivities from photothermal-dominated state to photodynamic-dominated state to maximize phototherapeutic effect. TPASIC-PFH@PLGA NPs are designed by incorporating oxygen-rich liquid perfluorohexane (PFH) into the intraparticle microenvironment to regulate the intramolecular motions of AIE photosensitizer TPASIC. TPASIC exhibits a unique aggregation-enhanced reactive oxygen species (ROS) generation feature. PFH incorporation affords TPASIC the initially dispersed state, thus promoting active intramolecular motions and photothermal conversion efficiency. While PFH volatilization leads to nanoparticle collapse and the formation of tight TPASIC aggregates with largely enhanced ROS generation efficiency. As a consequence, PFH incorporation not only currently promotes both photothermal and photodynamic efficacies of TPASIC and increases the intratumoral oxygen level, but also enables the smart photothermal-to-photodynamic switch to maximize the phototherapeutic performance. The integration of PFH and AIE photosensitizer eventually delivers more excellent antitumor effect over conventional phototherapeutic agents with fixed photothermal and photodynamic efficacies. This study proposes a new nanoengineering strategy to ameliorate TME and adapt the treatment modality to fit the changed TME for advanced antitumor applications.

Nanoscale octopus guiding telomere entanglement: An innovative strategy for inducing apoptosis in cancer cells.

Telomere length plays a crucial role in cellular aging and the risk of diseases. Unlike normal cells, cancer cells can extend their own survival by maintaining telomere stability through telomere maintenance mechanism. Therefore, regulating the lengths of telomeres have emerged as a promising approach for anti-cancer treatment. In this study, we introduce a nanoscale octopus-like structure designed to induce physical entangling of telomere, thereby efficiently triggering telomere dysfunction. The nanoscale octopus, composed of eight-armed PEG (8-arm-PEG), are functionalized with cell penetrating peptide (TAT) to facilitate nuclear entry and are covalently bound to N-Methyl Mesoporphyrin IX (NMM) to target G-quadruplexes (G4s) present in telomeres. The multi-armed configuration of the nanoscale octopus enables targeted binding to multiple G4s, physically disrupting and entangling numerous telomeres, thereby triggering telomere dysfunction. Both in vitro and in vivo experiments indicate that the nanoscale octopus significantly inhibits cancer cell proliferation, induces apoptosis through telomere entanglement, and ultimately suppresses tumor growth. This research offers a novel perspective for the development of innovative anti-cancer interventions and provides potential therapeutic options for targeting telomeres.

Conversion of albumin into a BODIPY-like photosensitizer by a flick reaction, tumor accumulation and photodynamic therapy.

The accumulation of photosensitizers (PSs) in lesion sites but not in other organs is an important challenge for efficient image guiding in photodynamic therapy. Cancer cells are known to express a significant number of albumin-binding proteins that take up albumin as a nutrient source. Here, we converted albumin to a novel BODIPY-like PS by generating a tetrahedral boron environment via a flick reaction. The formed albumin PS has almost the same 3-dimensional structural feature as free albumin because binding occurs at Sudlow Site 1, which is located in the interior space of albumin. An i.v. injection experiment in tumor-bearing mice demonstrated that the human serum albumin PS effectively accumulated in cancer tissue and, more surprisingly, albumin PS accumulated much more in the cancer tissue than in the liver and kidneys. The albumin PS was effective at killing tumor cells through the generation of reactive oxygen species under light irradiation. The crystal structure of the albumin PS was fully elucidated by X-ray crystallography; thus, further tuning of the structure will lead to novel physicochemical properties of the albumin PS, suggesting its potential in biological and clinical applications.

Cyclic catalysis of intratumor Fe3+/2+ initiated by a hollow mesoporous iron sesquioxide nanoparticle for ferroptosis therapy of large tumors.

Numerous nanoparticles have been utilized to deliver Fe2+ for tumor ferroptosis therapy, which can be readily converted to Fe3+via Fenton reactions to generate hydroxyl radical (•OH). However, the ferroptosis therapeutic efficacy of large tumors is limited due to the slow conversion of Fe3+ to Fe2+via Fenton reactions. Herein, a strategy of intratumor Fe3+/2+ cyclic catalysis is proposed for ferroptosis therapy of large tumors, which was realized based on our newly developed hollow mesoporous iron sesquioxide nanoparticle (HMISN). Cisplatin (CDDP) and Gd-poly(acrylic acid) macrochelates (GP) were loaded into the hollow core of HMISN, whose surface was modified by laccase (LAC). Fe3+, CDDP, GP, and LAC can be gradually released from CDDP@GP@HMISN@LAC in the acidic tumor microenvironment. The intratumor O2 can be catalyzed into superoxide anion (O2•-) by LAC, and the intratumor NADPH oxidases can be activated by CDDP to generate O2•-. The O2•- can react with Fe3+ to generate Fe2+, and raise H2O2 level via the superoxide dismutase. The generated Fe2+ and H2O2 can be fast converted into Fe3+ and •OH via Fenton reactions. The cyclic catalysis of intratumor Fe3+/2+ initiated by CDDP@GP@HMISN@LAC can be used for ferroptosis therapy of large tumors.

A carrier-free nanovaccine combined with cancer immunotherapy overcomes gemcitabine resistance.

Drug resistance is a significant challenge in cancer chemotherapy and is a primary factor contributing to poor recovery for cancer patients. Although drug-loaded nanoparticles have shown promise in overcoming chemotherapy resistance, they often carry a combination of drugs and require advanced design and manufacturing processes. Furthermore, they seldom approach chemotherapy-resistant tumors from an immunotherapy perspective. In this study, we developed a therapeutic nanovaccine composed solely of chemotherapy-induced resistant tumor antigens (CIRTAs) and the immune adjuvant Toll-like receptor (TLR) 7/8 agonist R848 (CIRTAs@R848). This nanovaccine does not require additional carriers and has a simple production process. It efficiently delivers antigens and immune stimulants to dendritic cells (DCs) simultaneously, promoting DCs maturation. CIRTAs@R848 demonstrated significant tumor suppression, particularly when used in combination with the immune checkpoint blockade (ICB) anti-PD-1 (αPD-1). The combined therapy increased the infiltration of T cells into the tumor while decreasing the proportion of regulatory T cells (Tregs) and modulating the tumor microenvironment, resulting in long-term immune memory. Overall, this study introduces an innovative strategy for treating chemotherapy-resistant tumors from a novel perspective, with potential applications in personalized immunotherapy and precision medicine.

Liposomes-enabled cancer chemoimmunotherapy.

Chemoimmunotherapy is an emerging paradigm in the clinic for treating several malignant diseases, such as non-small cell lung cancer, breast cancer, and large B-cell lymphoma. However, the efficacy of this strategy is still restricted by serious adverse events and a high therapeutic termination rate, presumably due to the lack of tumor-targeted distribution of both chemotherapeutic and immunotherapeutic agents. Targeted drug delivery has the potential to address this issue. Among the most promising nanocarriers in clinical translation, liposomes have drawn great attention in cancer chemoimmunotherapy in recent years. Liposomes-enabled cancer chemoimmunotherapy has made significant progress in clinics, with impressive therapeutic outcomes. This review summarizes the latest preclinical and clinical progress in liposome-enabled cancer chemoimmunotherapy and discusses the challenges and future directions of this field.

Harnessing nanomaterials for copper-induced cell death.

Copper (Cu), an essential micronutrient with redox properties, plays a pivotal role in a wide array of pathological and physiological processes across virtually all cell types. Maintaining an optimal copper concentration is critical for cellular survival: insufficient copper levels disrupt respiration and metabolism, while excess copper compromises cell viability, potentially leading to cell death. Similarly, in the context of cancer, copper exhibits a dual role: appropriate amount of copper can promote tumor progression and be an accomplice, yet beyond befitting level, copper can bring about multiple types of cell death, including autophagy, apoptosis, ferroptosis, immunogenic cell death, pyroptosis, and cuproptosis. These forms of cell death are beneficial against cancer progression; however, achieving precise copper regulation within tumors remains a significant challenge in the pursuit of effective cancer therapies. The emergence of nanodrug delivery systems, distinguished by their precise targeting, controlled release, high payload capacity, and the ability to co-deliver multiple agents, has revitalized interest in exploiting copper's precise regulatory capabilities. Nevertheless, there remains a dearth of comprehensive review of copper's bidirectional effects on tumorigenesis and the role of copper-based nanomaterials in modulating tumor progression. This paper aims to address this gap by elucidating the complex role in cancer biology and highlighting its potential as a therapeutic target. Through an exploration of copper's dualistic nature and the application of nanotechnology, this review seeks to offer novel insights and guide future research in advancing cancer treatment.

Enhanced anticancer efficacy of TRAIL-conjugated and odanacatib-loaded PLGA nanoparticles in TRAIL resistant cancer.

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) demonstrates unique characteristics in anticancer therapies as it selectively induces apoptosis in cancer cells. However, most cancer cells are TRAIL-resistant. Odanacatib (ODN), a cathepsin K inhibitor, is considered a novel sensitizer for cancer treatment. Combination therapy between TRAIL and sensitizers is considered a potent platform that improves TRAIL-based anticancer therapies beyond TRAIL monotherapy. Herein, we developed ODN loaded poly(lactic-co-glycolic) nanoparticles conjugated to GST-TRAIL (TRAIL-ODN-PLGA-NPs) to target and treat TRAIL-resistant cancer. TRAIL-ODN-PLGA-NPs demonstrated a significant increase in cellular uptake via death receptors (DR5 and DR4) on surface of cancer cells. TRAIL-ODN-PLGA-NPs exposure destroyed more TRAIL-resistant cells compared to a single treatment with free drugs. The released ODN decreased the Raptor protein, thereby increasing damage to mitochondria by elevating reactive oxygen species (ROS) generation. Additionally, Bim protein stabilization improved TRAIL-resistant cell sensitization to TRAIL-induced apoptosis. The in vivo biodistribution study revealed that TRAIL-ODN-PLGA-NPs demonstrated high location and retention in tumor sites via the intravenous route. Furthermore, TRAIL-ODN-PLGA-NPs significantly inhibited xenograft tumor models of TRAIL-resistant Caki-1 and TRAIL-sensitive MDA-MB-231 cells.The inhibition was associated with apoptosis activation, Raptor protein stabilizing Bim protein downregulation, Bax accumulation, and mitochondrial ROS generation elevation. Additionally, TRAIL-ODN-PLGA-NPs affected the tumor microenvironment by increasing tumor necrosis factor-α and reducing interleukin-6. In conclusion, we evealed that our formulation demonstrated synergistic effects against TRAIL compared with the combination of free drug in vitro and in vivo models. Therefore, TRAIL-ODN-PLGA-NPs may be a novel candidate for TRAIL-induced apoptosis in cancer treatment.

An ultra-long-acting L-asparaginase synergizes with an immune checkpoint inhibitor in starvation-immunotherapy of metastatic solid tumors.

Metastasis stands as the primary contributor to mortality associated with tumors. Chemotherapy and immunotherapy are frequently utilized in the management of metastatic solid tumors. Nevertheless, these therapeutic modalities are linked to serious adverse effects and limited effectiveness in preventing metastasis. Here, we report a novel therapeutic strategy named starvation-immunotherapy, wherein an immune checkpoint inhibitor is combined with an ultra-long-acting L-asparaginase that is a fusion protein comprising L-asparaginase (ASNase) and an elastin-like polypeptide (ELP), termed ASNase-ELP. ASNase-ELP's thermosensitivity enables it to generate an in-situ depot following an intratumoral injection, yielding increased dose tolerance, improved pharmacokinetics, sustained release, optimized biodistribution, and augmented tumor retention compared to free ASNase. As a result, in murine models of oral cancer, melanoma, and cervical cancer, the antitumor efficacy of ASNase-ELP by selectively and sustainably depleting L-asparagine essential for tumor cell survival was substantially superior to that of ASNase or Cisplatin, a first-line anti-solid tumor medicine, without any observable adverse effects. Furthermore, the combination of ASNase-ELP and an immune checkpoint inhibitor was more effective than either therapy alone in impeding melanoma metastasis. Overall, the synergistic strategy of starvation-immunotherapy holds excellent promise in reshaping the therapeutic landscape of refractory metastatic tumors and offering a new alternative for next-generation oncology treatments.

Tumor microenvironment-modulated nanoparticles with cascade energy transfer as internal light sources for photodynamic therapy of deep-seated tumors.

Photodynamic therapy (PDT) is an appealing modality for cancer treatments. However, the limited tissue penetration depth of external-excitation light makes PDT impossible in treating deep-seated tumors. Meanwhile, tumor hypoxia and intracellular reductive microenvironment restrain the generation of reactive oxygen species (ROS). To overcome these limitations, a tumor-targeted self-illuminating supramolecular nanoparticle T-NPCe6-L-N is proposed by integrating photosensitizer Ce6 with luminol and nitric oxide (NO) for chemiluminescence resonance energy transfer (CRET)-activated PDT. The high H2O2 level in tumor can trigger chemiluminescence of luminol to realize CRET-activated PDT without exposure of external light. Meanwhile, the released NO significantly relieves tumor hypoxia via vascular normalization and reduces intracellular reductive GSH level, further enhancing ROS abundance. Importantly, due to the different ROS levels between cancer cells and normal cells, T-NPCe6-L-N can selectively trigger PDT in cancer cells while sparing normal cells, which ensured low side effect. The combination of CRET-based photosensitizer-activation and tumor microenvironment modulation overcomes the innate challenges of conventional PDT, demonstrating efficient inhibition of orthotopic and metastatic tumors on mice. It also provoked potent immunogenic cell death to ensure long-term suppression effects. The proof-of-concept research proved as a new strategy to solve the dilemma of PDT in treatment of deep-seated tumors.

In-situ activation of biomimetic single-site bioorthogonal nanozyme for tumor-specific combination therapy.

Copper-catalyzed click chemistry offers creative strategies for activation of therapeutics without disrupting biological processes. Despite tremendous efforts, current copper catalysts face fundamental challenges in achieving high efficiency, atom economy, and tissue-specific selectivity. Herein, we develop a facile "mix-and-match synthetic strategy" to fabricate a biomimetic single-site copper-bipyridine-based cerium metal-organic framework (Cu/Ce-MOF@M) for efficient and tumor cell-specific bioorthogonal catalysis. This elegant methodology achieves isolated single-Cu-site within the MOF architecture, resulting in exceptionally high catalytic performance. Cu/Ce-MOF@M favors a 32.1-fold higher catalytic activity than the widely used MOF-supported copper nanoparticles at single-particle level, as first evidenced by single-molecule fluorescence microscopy. Furthermore, with cancer cell-membrane camouflage, Cu/Ce-MOF@M demonstrates preferential tropism for its parent cells. Simultaneously, the single-site CuII species within Cu/Ce-MOF@M are reduced by upregulated glutathione in cancerous cells to CuI for catalyzing the click reaction, enabling homotypic cancer cell-activated in situ drug synthesis. Additionally, Cu/Ce-MOF@M exhibits oxidase and peroxidase mimicking activities, further enhancing catalytic cancer therapy. This study guides the reasonable design of highly active heterogeneous transition-metal catalysts for targeted bioorthogonal reactions.

Identifying mitigating strategies for endothelial cell dysfunction and hypertension in response to VEGF receptor inhibitors.

Vascular endothelial growth factor receptor inhibitors (VEGFRis) improve cancer survival but are associated with treatment-limiting hypertension, often attributed to endothelial cell (EC) dysfunction. Using phosphoproteomic profiling of VEGFRi-treated ECs, drugs were screened for mitigators of VEGFRi-induced EC dysfunction and validated in primary aortic ECs, mice, and canine cancer patients. VEGFRi treatment significantly raised systolic blood pressure (SBP) and increased markers of endothelial and renal dysfunction in mice and canine cancer patients. α-Adrenergic-antagonists were identified as drugs that most oppose the VEGFRi proteomic signature. Doxazosin, one such α-antagonist, prevented EC dysfunction in murine, canine, and human aortic ECs. In mice with sorafenib-induced-hypertension, doxazosin mitigated EC dysfunction but not hypertension or glomerular endotheliosis, while lisinopril mitigated hypertension and glomerular endotheliosis without impacting EC function. Hence, reversing EC dysfunction was insufficient to mitigate VEGFRi-induced-hypertension in this mouse model. Canine cancer patients with VEGFRi-induced-hypertension were randomized to doxazosin or lisinopril and both agents significantly decreased SBP. The canine clinical trial supports safety and efficacy of doxazosin and lisinopril as antihypertensives for VEGFRi-induced-hypertension and the potential of trials in canines with spontaneous cancer to accelerate translation. The overall findings demonstrate the utility of phosphoproteomics to identify EC-protective agents to mitigate cardio-oncology side effects.

The incidence and trends of proteinuria, azotemia and hypertension in cats receiving toceranib phosphate.

OBJECTIVES: This retrospective study aimed to determine the incidence and trends of proteinuria, elevations in serum creatinine and urea, and systolic blood pressure in cats undergoing treatment with toceranib. METHODS: In total, 32 cats treated with toceranib for malignancies were analyzed. Cats were included if urinalysis and urine protein:creatinine ratio (UPC) measurements were available at 28 days (T1) and 56 days (T2) after starting the treatment. Cats with concurrent lower urinary tract disease, including urinary tract malignancy, were excluded. Friedman's ANOVA compared variables between time points, and the Spearman test assessed the correlation between treatment duration and UPC. RESULTS: The median starting dose of toceranib was 2.68 mg/kg (range 1.7-3.9). In total, 15 (46.9%) cats received concurrent non-steroidal anti-inflammatory drugs. The most commonly treated tumors were oral squamous cell carcinoma (n = 10) and mast cell tumor (n = 5). None of the 32 cats developed progressive proteinuria or azotemia during the follow-up period (median 56 days; range 56-336). Notably, UPC and serum creatinine were significantly lower at T2 compared with baseline (P = 0.012 and 0.001, respectively). Among the four cats with baseline proteinuria, UPC decreased over time with or without concurrent telmisartan treatment (n = 2). All four of these cats experienced a reduction in tumor size with toceranib concurrently with their decreased UPC. There was no significant correlation between UPC and the duration of toceranib treatment (P = 0.089). Blood pressure was not significantly different over the assessed time points. CONCLUSIONS AND RELEVANCE: The incidence of proteinuria, renal azotemia and hypertension in cats treated with toceranib for neoplasia appears to be low. Toceranib may be a viable treatment option even in cats with pre-existing proteinuria or renal disease, with careful monitoring of trends recommended.

Predictive value of near-term prediction models for severe immune-related adverse events in malignant tumor PD-1 inhibitor therapy.

Immune-related adverse events (irAEs) impact outcomes, with most research focusing on early prediction (baseline data), rather than near-term prediction (one cycle before the occurrence of irAEs and the current cycle). We aimed to explore the near-term predictive value of neutrophil/lymphocyte ratio (NLR), platelet/lymphocyte ratio (PLR), absolute eosinophil count (AEC) for severe irAEs induced by PD-1 inhibitors. Data were collected from tumor patients treated with PD-1 inhibitors. NLR, PLR, and AEC data were obtained from both the previous and the current cycles of irAEs occurrence. A predictive model was developed using elastic net logistic regression Cutoff values were determined using Youden's Index. The predicted results were compared with actual data using Bayesian survival analysis. A total of 138 patients were included, of whom 47 experienced grade 1-2 irAEs and 18 experienced grade 3-5 irAEs. The predictive model identified optimal α and λ through 10-fold cross-validation. The Shapiro-Wilk test, Kruskal-Wallis test and logistic regression showed that only current cycle data were meaningful. The NLR was statistically significant in predicting irAEs in the previous cycle. Both NLR and AEC were significant predictors of irAEs in the current cycle. The model achieved an area under the ROC curve (AUC) of 0.783, with a sensitivity of 77.8% and a specificity of 80.8%. A probability ≥ 0.1345 predicted severe irAEs. The model comprising NLR, AEC, and sex may predict the irAEs classification in the current cycle, offering a near-term predictive advantage over baseline models and potentially extending the duration of immunotherapy for patients.

Therapeutic types and advantages of functionalized nanoparticles in inducing ferroptosis in cancer therapy.

BACKGROUND: The clinical efficacy of cancer treatment protocols remains unsatisfactory; however, the emergence of ferroptosis-driven therapy strategies has renewed hope for tumor treatment, owing to their remarkable tumor suppression effects. Biologically based small-molecule inducers are used in conventional method to induce ferroptosis. Nevertheless, some molecular drugs have limited solubility, poor ability to target cells, and fast metabolism, which hinder their ability to induce ferroptosis over a prolonged period. Fortunately, further investigations of ferroptosis and the development of nanotechnology have demonstrated that nanoparticles (NPs) are more efficient in inducing ferroptosis than drugs alone, which opens up new perspectives for cancer therapy. OBJECTIVE: In order to organize a profile of recent advance in NPs for inducing ferroptosis in cancer therapy, and NPs were comprehensively classified in a new light.Materials and methods: We comprehensively searched the databases such as PubMed and Embase. The time limit for searching was from the establishment of the database to 2023.11. All literatures were related to "ferroptosis", "nanoparticles", "nanodelivery systems", "tumors", "cancer". RESULTS: We summarized and classified the available NPs from a new perspective. The NPs were classified into six categories based on their properties: (1) iron oxide NPs (2) iron - based conversion NPs (3) core-shell structure (4) organic framework (5) silica NPs (6) lipoprotein NPs. According to the therapeutic types of NPs, they can be divided into categories: (1) NPs induced ferroptosis-related immunotherapy (2) NPs loaded with drugs (3) targeted therapy of NPs (4) multidrug resistance therapy (5) gene therapy with NPs (6) energy conversion therapy. CONCLUSIONS: The insights gained from this review can provide ideas for the development of original NPs and nanodelivery systems, pave the way for related nanomaterials application in clinical cancer therapy, and advance the application and development of nanotechnology in the medical field.

Prevalence and risks of intravenous chemotherapy-induced severe neutropenia in solid cancers: a multicenter retrospective cohort study on real-life data.

PURPOSE: We aimed at identifying prevalence, clinical outcomes and prognostic factors in cancer patients with intravenous chemotherapy-induced severe neutropenia (ICISN). METHODS: In this multicenter retrospective cohort study on the clinical data warehouse of Greater Paris University Hospitals (AP-HP), we included all adult patients with solid cancer hospitalized between 2016 and 2021 with intravenous chemotherapy within 30 days prior to severe neutropenia (D70 or D611 ICD-10 codes AND a neutrophil count < 500/mm3). The primary endpoint was referral to intensive care unit (ICU) or death within 30 days. We collected cancer, patient, and treatment characteristics. RESULTS: Among 141,586 cancer inpatients, 40,660 received chemotherapy among whom 661 (1.6%) had ICISN. Median age was 63 years (interquartile range (IQR), 54-70) and 330 patients (49%) were female. The median Charlson score was 10 (IQR, 8-11). Main primary cancers were lung (n = 204, 31%) and breast (n = 87, 13%). Advanced cancers were found in 551 patients (83%), 331 (50%) were in 1st line of chemotherapy, 284 (42%) in the 1st cycle of the current line and 149 (22%) had primary G-CSF. Documented bacterial (mostly gram-negative bacilli) and fungal infections were observed in 113 (17%) and 19 (3%) patients; 58 (9%) were transferred to ICU and 82 (12%) died within 30 days, 372 (56%) patients received subsequent chemotherapy. Independent prognostic factors were the level of monocyte, lymphocyte counts or albuminemia and a documented bacterial infection, while Charlson index and primary prophylactic G-CSF were not associated with patient clinical outcomes. CONCLUSION: Despite the use of primary G-CSF, ICISN remains a frequent event, which leads to ICU death in one on five cases Some prognostic factors of severity have been highlighted and could help clinicians to prevent severe complications.

Multi-task deep latent spaces for cancer survival and drug sensitivity prediction.

MOTIVATION: Cancer is a very heterogeneous disease that can be difficult to treat without addressing the specific mechanisms driving tumour progression in a given patient. High-throughput screening and sequencing data from cancer cell-lines has driven many developments in drug development, however, there are important aspects crucial to precision medicine that are often overlooked, namely the inherent differences between tumours in patients and the cell-lines used to model them in vitro. Recent developments in transfer learning methods for patient and cell-line data have shown progress in translating results from cell-lines to individual patients in silico. However, transfer learning can be forceful and there is a risk that clinically relevant patterns in the omics profiles of patients are lost in the process. RESULTS: We present MODAE, a novel deep learning algorithm to integrate omics profiles from cell-lines and patients for the purposes of exploring precision medicine opportunities. MODAE implements patient survival prediction as an additional task in a drug-sensitivity transfer learning schema and aims to balance autoencoding, domain adaptation, drug-sensitivity prediction, and survival prediction objectives in order to better preserve the heterogeneity between patients that is relevant to survival. While burdened with these additional tasks, MODAE performed on par with baseline survival models, but struggled in the drug-sensitivity prediction task. Nevertheless, these preliminary results were promising and show that MODAE provides a novel AI-based method for prioritizing drug treatments for high-risk patients. AVAILABILITY AND IMPLEMENTATION: https://github.com/UEFBiomedicalInformaticsLab/MODAE.