Despite improvements in cancer survival rates, metastatic and surgery-resistant cancers, such as pancreatic cancer, remain challenging, with poor prognoses and limited treatment options. Enhancing drug bioavailability in tumors, while minimizing off-target effects, is crucial. Metal-organic frameworks (MOFs) have emerged as promising drug delivery vehicles owing to their high loading capacity, biocompatibility, and functional tunability. However, the vast chemical diversity of MOFs complicates the rational design of biocompatible materials. This study employed machine learning and molecular simulations to identify MOFs suitable for encapsulating gemcitabine, paclitaxel, and SN-38, and identified PCN-222 as an optimal candidate. Following drug loading, MOF formulations are improved for colloidal stability and biocompatibility. In vitro studies on pancreatic cancer cell lines have shown high biocompatibility, cellular internalization, and delayed drug release. Long-term stability tests demonstrated a consistent performance over 12 months. In vivo studies in pancreatic tumor-bearing mice revealed that paclitaxel-loaded PCN-222, particularly with a hydrogel for local administration, significantly reduced metastatic spread and tumor growth compared to the free drug. These findings underscore the potential of PCN-222 as an effective drug delivery system for the treatment of hard-to-treat cancers.

Sexually transmitted infections and urogenital-perinatal infections are significant health challenges owing to their asymptomatic nature, multidrug-resistant pathogens, and lack of effective vaccines. Surfactants are under investigation as potential antimicrobial agents and alternatives to traditional antibiotics. Here, we discovered that N -dodecylpyridinium bromide (C 12 PB), a cationic quaternary ammonium surfactant, has very low potential to induce antimicrobial resistance with no antibiotic cross -resistance or inflammation in vitro . Therefore, we developed a preclinical antibiotic -free cationic surfactant -based cellulose hydrogel for treating sexually transmitted infections. The C 12 PB-hydrogels provided sustained surfactant release, enhancing their biocompatibility and antibacterial activity without inflammation or epithelial disruption of the vaginal tract. In a preclinical model of Neisseria gonorrhoeae infection, a single application of the C 12 PB-hydrogel showed a 2- to 3 -fold reduction in infection. This lays the foundation for the future development of C 12 PB-hydrogels for sexually transmitted infections, demonstrating potent antibacterial activity and minimal risk of antimicrobial resistance or inflammation.

Photodynamic therapy (PDT), an emergent noninvasive cancer treatment, is largely dependent on the presence of efficient photosensitizers (PSs) and a sufficient oxygen supply. However, the therapeutic efficacy of PSs is greatly compromised by poor solubility, aggregation tendency, and oxygen depletion within solid tumors during PDT in hypoxic microenvironments. Despite the potential of PS-based metal-organic frameworks (MOFs), addressing hypoxia remains challenging. Boron dipyrromethene (BODIPY) chromophores, with excellent photostability, have exhibited great potential in PDT and bioimaging. However, their practical application suffers from limited chemical stability under harsh MOF synthesis conditions. Herein, we report the synthesis of the first example of a Zr-based MOF, namely, 69-L2, exclusively constructed from the BODIPY-derived ligands via a single-crystal to single-crystal post-synthetic exchange, where a direct solvothermal method is not applicable. To increase the PDT performance in hypoxia, we modify 69-L2 with fluorinated phosphate-functionalized methoxy poly(ethylene glycol). The resulting 69-L2@F is an oxygen carrier, enabling tumor oxygenation and simultaneously acting as a PS for reactive oxygen species (ROS) generation under LED irradiation. We demonstrate that 69-L2@F has an enhanced PDT effect in triple-negative breast cancer MDA-MB-231 cells under both normoxia and hypoxia. Following positive results, we evaluated the in vivo activity of 69-L2@F with a hydrogel, enabling local therapy in a triple-negative breast cancer mice model and achieving exceptional antitumor efficacy in only 2 days. We envision BODIPY-based Zr-MOFs to provide a solution for hypoxia relief and maximize efficacy during in vivo PDT, offering new insights into the design of promising MOF-based PSs for hypoxic tumors.

Owing to their distinct physical and chemical properties, inorganic nanoparticles (NPs) have shown promising results in preclinical cancer therapy, but designing and engineering them for effective therapeutic purposes remains a challenge. Although a comprehensive database of inorganic NP research is not currently available, it is crucial for developing effective cancer therapies. In this context, machine learning (ML) has emerged as a transformative tool, but its adaptation to nanomedicine is hindered by inexistent or small datasets. Here we assembled a large database of inorganic NPs, comprising experimental datasets from 745 preclinical studies in cancer nanomedicine. Using descriptive statistics and explainable ML models we mined this database to gain knowledge of inorganic NP design patterns and inform future NP research for cancer treatment. Our analyses suggest that NP shape and therapy type are prominent features in determining in vivo efficacy, measured as a percentage of tumour reduction. Moreover, our database provides a large-scale open-access resource for discriminative ML that the broader nanotechnology community can utilize. Our work blueprints data mining for translational cancer research and offers evidence for standardizing NP reporting to accelerate and de-risk inorganic NP-based drug delivery, which may help to improve patient outcomes in clinical settings. This analysis leverages a large-scale literature review, text mining, statistics and machine learning to identify trends, shortcomings and future opportunities in developing and deploying inorganic nanoparticles for cancer diagnosis and therapy.

Breast cancer is the primary cause of cancer-related death in women worldwide. Breast cancer subtypes are characterized by different gene expression patterns, which drive their prognostic factors and therapeutic options. Among them, triple-negative breast cancer (TNBC) is one of the deadliest due to its aggressiveness, high rate of early recurrence and distant metastases, and limited therapeutic options. Despite the recent approval of monoclonal antibodies targeting programmed cell death protein 1 (PD-1) or its ligand (PD-L1) for the treatment of TNBC patients with a locally recurrent unresectable or metastatic tumor expressing PD-L1, their response rate is very modest. It is reported that polymeric nanoparticle (NP)-based cancer vaccines, co-entrapping tumor-associated antigens, Toll-like receptor ligands and small interfering RNA (siRNA) targeting the expression of the immunosuppressive cytokine transforming growth factor (TGF)-β1 by dendritic cells, sensitized TNBC to the agonist immune checkpoint OX40, inhibiting tumor growth and increasing overall survival. This anti-tumor immune-mediated effect is also observed in a luminal type of mammary cancer similar to human disease. Therefore, these synergistic anticancer effects of αOX40 and the antigen-specific adaptive immunity induced by nanovaccine-mediated TGF-β silencing may guide the development of novel combination regimens able to improve the response rate to this aggressive tumor.

Glioblastoma is a lethal brain cancer with treatment resistance stemming from its interactions with the surrounding microenvironment and obstacles such as the blood-brain barrier. Conventional therapies such as surgery and chemotherapy have shown limited efficacy, whereas immunotherapies, effective in other solid cancers, face obstacles in glioblastoma owing to its unique immunological dysfunction. Despite the development of peptide, neoantigen, cell-based and mRNA-based vaccines, progress to advanced clinical trials has been sluggish. Factors contributing to this slow progress include the immunosuppressive microenvironment of the tumour, the presence of the blood-brain barrier and the inherent instability of glioblastoma vaccines, collectively hindering treatment efficacy. In this context, nanomaterials have emerged as promising owing to their capacity to cross the blood-brain barrier, shield therapeutics from degradation and efficiently target the brain. In this Perspective, we highlight the development of glioblastoma nanovaccination, discussing strategies for nanoparticle engineering to breach the blood-brain barrier and target both immune and glioblastoma cells, paving the way for potential breakthroughs in glioblastoma treatment. Developing vaccines for glioblastoma remains challenging owing to the immunosuppressive microenvironment of the tumour and the presence of the blood-brain barrier. In this Perspective, we explore how nanomaterials can be tailored to address the limitations of glioblastoma vaccination, potentially paving the way for important advancements.

The deployment of structures that enable localized release of bioactive molecules can result in more efficacious treatment of disease and better integration of implantable bionic devices. The strategic design of a biopolymeric coating can be used to engineer the optimal release profile depending on the task at hand. As illustrative examples, here advances in delivery of drugs from bone, brain, ocular, and cardiovascular implants are reviewed. These areas are focused to highlight that both hard and soft tissue implants can benefit from controlled localized delivery. The composition of biopolymers used to achieve appropriate delivery to the selected tissue types, and their corresponding outcomes are brought to the fore. To conclude, key factors in designing drug-loaded biopolymeric coatings for biomedical implants are highlighted.

There is growing need for a safe, efficient, specific and non-pathogenic means for delivery of gene therapy materials. Nanomaterials for nucleic acid delivery offer an unprecedented opportunity to overcome these drawbacks; owing to their tunability with diverse physico-chemical properties, they can readily be functionalized with any type of biomolecules/moieties for selective targeting. Nucleic acid therapeutics such as antisense DNA, mRNA, small interfering RNA (siRNA) or microRNA (miRNA) have been widely explored to modulate DNA or RNA expression Strikingly, gene therapies combined with nanoscale delivery systems have broadened the therapeutic and biomedical applications of these molecules, such as bioanalysis, gene silencing, protein replacement and vaccines. Here, we overview how to design smart nucleic acid delivery methods, which provide functionality and efficacy in the layout of molecular diagnostics and therapeutic systems. It is crucial to outline some of the general design considerations of nucleic acid delivery nanoparticles, their extraordinary properties and the structure-function relationships of these nanomaterials with biological systems and diseased cells and tissues.

Nucleic acid-based therapy emerges as a powerful weapon for the treatment of tumors thanks to its direct, effective, and lasting therapeutic effect. Encouragingly, continuous nucleic acid-based drugs have been approved by the Food and Drug Administration (FDA) and the European Medicines Agency (EMA). Despite the tremendous progress, there are few nucleic acid-based drugs for brain tumors in clinic. The most challenging problems lie on the instability of nucleic acids, difficulty in traversing the biological barriers, and the off-target effect. Herein, nucleic acid-based therapy for brain tumor is summarized considering three aspects: (i) the therapeutic nucleic acids and their applications in clinical trials; (ii) the various administration routes for nucleic acid delivery and the respective advantages and drawbacks. (iii) the strategies and carriers for improving stability and targeting ability of nucleic acid drugs. This review provides thorough knowledge for the rational design of nucleic acid-based drugs against brain tumor.

The interest in nanomedicine for cancer theranostics has grown significantly over the past few decades. However, these nanomedicines need to overcome several physiological barriers intrinsic to the tumor microenvironment (TME) before reaching their target. Intrinsic tumor genetic/phenotypic variations, along with intratumor heterogeneity, provide different cues to each cancer type, making each patient with cancer unique. This brings additional challenges in translating nanotechnology-based systems into clinically reliable therapies. To develop efficient therapeutic strategies, it is important to understand the dynamic interactions between TME players and the complex mechanisms involved, because they constitute invaluable targets to dismantle tumor progression. In this review, we discuss the latest nanotechnology-based strategies for cancer diagnosis and therapy as well as the potential targets for the design of future anticancer nanomedicines.

Nanotechnology for healthcare is coming of age, but automating the design of composite materials poses unique challenges. Although machine learning is supporting groundbreaking discoveries in materials science, new initiatives leveraging learned patterns are required to fully realize the promise of nanodelivery systems and accelerate development pipelines. Nanotechnology for healthcare is coming of age, but automating the design of composite materials poses unique challenges. Although machine learning is supporting groundbreaking discoveries in materials science, new initiatives leveraging learned patterns are required to fully realize the promise of nanodelivery systems and accelerate development pipelines. Main textNanotechnology has seen numerous translational applications over diverse economy sectors, but only recently has it taken healthcare by storm.1Talebian S. Rodrigues T. das Neves J. Sarmento B. Langer R. Conde J. Facts and Figures on Materials Science and Nanotechnology Progress and Investment.ACS Nano. 2021; https://doi.org/10.1021/acsnano.1021c03992Crossref Google Scholar For example, the massively deployed Moderna and BioNTech/Pfizer COVID-19 vaccines use organic nanoparticles to deliver SARS-CoV-2 mRNA.2Talebian S. Conde J. Why Go NANO on COVID-19 Pandemic?.Matter. 2020; 3: 598-601Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar The pursuit of this novel immunization modality has positioned nanotechnology at the center of attention, providing ample clinical validation and further motivating its exploitation in disparate disease areas, e.g., cancer. In this regard, billions of dollars spent in basic/translational nanotechnology have, over the years, allowed a reasonable understanding of the design principles driving efficacy.1Talebian S. Rodrigues T. das Neves J. Sarmento B. Langer R. Conde J. Facts and Figures on Materials Science and Nanotechnology Progress and Investment.ACS Nano. 2021; https://doi.org/10.1021/acsnano.1021c03992Crossref Google Scholar Still, much remains to be explored. The anticipated shift to nanotechnology-centered molecular medicine urges the need for an innovative suite of computational tools that effectively harness the growing amount of information in this space. A data-driven (r)evolution, similar to what we are currently witnessing in chemistry and biology, might not be too distant. We argue that predictive modeling and the de novo design of composite nanodelivery systems will become a reality and ultimately endorse a new era in nanotechnology research. Herein, we critically discuss how machine learning (ML) can reshape next-generation drug delivery and the three challenges that must be addressed to enable continuous innovation through discriminative/generative nanotechnology.Challenge 1: Standardized data reportingQuality data are unavoidably the centerpiece of any ML tool, and the current lack of standardized reporting practices in nanobiotechnology and nanomedicine is a known issue (Figure 1).3Schrurs F. Lison D. Focusing the research efforts.Nat. Nanotechnol. 2012; 7: 546-548Crossref Scopus (83) Google Scholar,4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar This hinders reproducibility and meaningful comparative studies, despite a recent community effort to regulate and improve transparency in the disclosed materials.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar For example, the physicochemical properties (e.g., dimension, shape, surface charge, targeting agent density, and composition), administered dose, and loading in drug delivery systems are cornerstones to modulate pharmacokinetics and efficacy. However, their reporting heterogeneity or lack of explicit information in manuscripts jeopardizes the gain of momentum in nanomedical research.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar Further, we argue the exact composition, injected volume, concentration, and route of administration need to be accurately reported but are only sparingly discriminated. Multiple studies also describe the amount of only one component in the delivery system (e.g., iron or encapsulated drug dosage). Thus, normalizing the others by body weight in in vivo studies is virtually impossible. Paralleling deficiencies in the characterization of nanodelivery materials, one equally finds shortcomings in the report of assay endpoints. Delivery efficacy and tumor accumulation are usually provided as a percentage of initial dose/tumor mass in gram (%ID/g). This is only useful if the initial dose and tumor mass are also reported, which rarely is the case. Further, the percentage of tumor reduction constitutes an endpoint normalization, yielding no tractable information on either delivery efficiency or actual volume change. Finally, one must bear in mind that animal models are a proxy of an actual disease setting, and experiments must be carefully designed to ensure meaningfulness. Otherwise, they may inappropriately reproduce the disease and tumor microenvironment, as in the use of xenograft/allograft heterotopic models instead of orthotopic ones (e.g., lung cells injected subcutaneously).Overall, the design of nanomaterials is intricate, and the established practices in experimental characterization are manifestly insufficient to support translation in healthcare on a wider scale.5Conde J. Above and Beyond Cancer Therapy: Translating Biomaterials into the Clinic.Trends Cancer. 2020; 6: 730-732Abstract Full Text Full Text PDF Scopus (7) Google Scholar Perpetuating those practices will ultimately curb innovation and research throughput. We envisage that a shift to robust reporting standards will be instrumental for automating the development of nanodelivery materials.Challenge 2: Complete datasetsReporting completeness, as urged above, will then enable the construction of nanotechnology databases akin to ChEMBL (Figure 1).6Gaulton A. Hersey A. Nowotka M. Bento A.P. Chambers J. Mendez D. Mutowo P. Atkinson F. Bellis L.J. Cibrián-Uhalte E. et al.The ChEMBL database in 2017.Nucleic Acids Res. 2017; 45: D945-D954Crossref PubMed Scopus (1114) Google Scholar To emphasize the critical nature of quality and complete data, we extracted composition, physicochemical, pharmacokinetics, and dosing information for iron oxide nanoparticles reported in 315 research manuscripts published between 2006 and 2019 in reputed nanotechnology journals. From those, 68% did not report the size, shape, or zeta potential of the nanoparticles. Further, only 1% and 31% presented a pharmacokinetics profile (with elimination/distribution half-lives and delivery efficiency) and dosing information (route and dose), respectively. The observed trends are apparently transferred to other nanoparticle types. In 322 gold and 257 silica nanoparticle research manuscripts, we found 51% and 45% of them missing full physicochemical characterization, respectively. Identical percentages of missing pharmacokinetics and dosing data were found. Together, this highlights a deep-rooted limitation in the nanotechnology field that must be addressed. Until then, inputting or discarding potentially valuable information will be required for modeling, which is far from ideal.While a publicly available resource will likely remain inaccessible in the coming years—even with adequate reporting standards in place—we envisage that a continuous and concerted community effort will be key toward that end. Those efforts might be further assisted by natural language processing and deep learning techniques with the goal of accelerating the extraction of information from the scientific literature.7Öztürk H. Özgür A. Schwaller P. Laino T. Ozkirimli E. Exploring chemical space using natural language processing methodologies for drug discovery.Drug Discov. Today. 2020; 25: 689-705Crossref PubMed Scopus (36) Google Scholar By encompassing multiple unexplored data patterns, those data resources are expected to endorse automated processes and support the implementation of ML tools that allow more efficient experiment prioritizations.Challenge 3: A machine-readable nanotechnology languageWhile predictive modeling8Tao H. Wu T. Aldeghi M. Wu T.C. Aspuru-Guzik A. Kumacheva E. Nanoparticle synthesis assisted by machine learning.Nat. Rev. Mater. 2021; 6: 701-716Crossref Scopus (44) Google Scholar,9Hart G.L.W. Mueller T. Toher C. Curtarolo S. Machine learning for alloys.Nat. Rev. Mater. 2021; 6: 730-755Crossref Scopus (51) Google Scholar can be executed with quality datasets and established heuristics, generative design of composite materials requires the development of new toolkits. In small-molecule discovery, the SMILES or SELFIES10Krenn M. Häse F. Nigam A. Friederich P. Aspuru-Guzik A. Self-referencing embedded strings (SELFIES): A 100% robust molecular string representation.Mach. Learn. Sci. Technol. 2020; 1: 045024Crossref Scopus (102) Google Scholar languages encode atom connectivity, which implicitly hardwires all physicochemical and biological properties for a given entity. By learning this language, a computer is then able to programmatically generate new words/molecules (as strings of characters) according to a probability distribution for each newly added character. In doing so, researchers become armed with a powerful means to virtually access new chemical matter and more efficiently explore a vast search space. We argue that a similar approach can be pursued to generate and tailor composite delivery materials (Figure 1). Considering that information on constituents, including which entities, their percentage, and/or their concentration, is key to determining all the underlying physicochemical and biological properties, it becomes essential to devise a new language and ontology for canonical representation of composite material systems—both already reported and imagined by a computer. Said language ought to holistically represent the nanomaterial and thus be transferable to any use case aside from the drug delivery systems we focus on here. Once this technology is in hand, the research community will gain access to an untapped concept for the de novo design of composite materials. If employed correctly, we expect that such ML models—which might be available in a short/medium term—could impact nanotechnology for healthcare similarly to how they are transforming discovery chemistry.Overall, we anticipate a gain of momentum for nanotechnology research and preview its future developments leveraged by ML concepts. The solutions we propose to the three outstanding challenges are realistic but surprisingly still not tackled by the research community. We expect the tight integration of computational technologies with robotics to result in a digital nanotechnology era that will see prototyping innovative and life-changing therapeutics done at a fraction of the time currently needed. Main textNanotechnology has seen numerous translational applications over diverse economy sectors, but only recently has it taken healthcare by storm.1Talebian S. Rodrigues T. das Neves J. Sarmento B. Langer R. Conde J. Facts and Figures on Materials Science and Nanotechnology Progress and Investment.ACS Nano. 2021; https://doi.org/10.1021/acsnano.1021c03992Crossref Google Scholar For example, the massively deployed Moderna and BioNTech/Pfizer COVID-19 vaccines use organic nanoparticles to deliver SARS-CoV-2 mRNA.2Talebian S. Conde J. Why Go NANO on COVID-19 Pandemic?.Matter. 2020; 3: 598-601Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar The pursuit of this novel immunization modality has positioned nanotechnology at the center of attention, providing ample clinical validation and further motivating its exploitation in disparate disease areas, e.g., cancer. In this regard, billions of dollars spent in basic/translational nanotechnology have, over the years, allowed a reasonable understanding of the design principles driving efficacy.1Talebian S. Rodrigues T. das Neves J. Sarmento B. Langer R. Conde J. Facts and Figures on Materials Science and Nanotechnology Progress and Investment.ACS Nano. 2021; https://doi.org/10.1021/acsnano.1021c03992Crossref Google Scholar Still, much remains to be explored. The anticipated shift to nanotechnology-centered molecular medicine urges the need for an innovative suite of computational tools that effectively harness the growing amount of information in this space. A data-driven (r)evolution, similar to what we are currently witnessing in chemistry and biology, might not be too distant. We argue that predictive modeling and the de novo design of composite nanodelivery systems will become a reality and ultimately endorse a new era in nanotechnology research. Herein, we critically discuss how machine learning (ML) can reshape next-generation drug delivery and the three challenges that must be addressed to enable continuous innovation through discriminative/generative nanotechnology.Challenge 1: Standardized data reportingQuality data are unavoidably the centerpiece of any ML tool, and the current lack of standardized reporting practices in nanobiotechnology and nanomedicine is a known issue (Figure 1).3Schrurs F. Lison D. Focusing the research efforts.Nat. Nanotechnol. 2012; 7: 546-548Crossref Scopus (83) Google Scholar,4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar This hinders reproducibility and meaningful comparative studies, despite a recent community effort to regulate and improve transparency in the disclosed materials.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar For example, the physicochemical properties (e.g., dimension, shape, surface charge, targeting agent density, and composition), administered dose, and loading in drug delivery systems are cornerstones to modulate pharmacokinetics and efficacy. However, their reporting heterogeneity or lack of explicit information in manuscripts jeopardizes the gain of momentum in nanomedical research.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar Further, we argue the exact composition, injected volume, concentration, and route of administration need to be accurately reported but are only sparingly discriminated. Multiple studies also describe the amount of only one component in the delivery system (e.g., iron or encapsulated drug dosage). Thus, normalizing the others by body weight in in vivo studies is virtually impossible. Paralleling deficiencies in the characterization of nanodelivery materials, one equally finds shortcomings in the report of assay endpoints. Delivery efficacy and tumor accumulation are usually provided as a percentage of initial dose/tumor mass in gram (%ID/g). This is only useful if the initial dose and tumor mass are also reported, which rarely is the case. Further, the percentage of tumor reduction constitutes an endpoint normalization, yielding no tractable information on either delivery efficiency or actual volume change. Finally, one must bear in mind that animal models are a proxy of an actual disease setting, and experiments must be carefully designed to ensure meaningfulness. Otherwise, they may inappropriately reproduce the disease and tumor microenvironment, as in the use of xenograft/allograft heterotopic models instead of orthotopic ones (e.g., lung cells injected subcutaneously).Overall, the design of nanomaterials is intricate, and the established practices in experimental characterization are manifestly insufficient to support translation in healthcare on a wider scale.5Conde J. Above and Beyond Cancer Therapy: Translating Biomaterials into the Clinic.Trends Cancer. 2020; 6: 730-732Abstract Full Text Full Text PDF Scopus (7) Google Scholar Perpetuating those practices will ultimately curb innovation and research throughput. We envisage that a shift to robust reporting standards will be instrumental for automating the development of nanodelivery materials.Challenge 2: Complete datasetsReporting completeness, as urged above, will then enable the construction of nanotechnology databases akin to ChEMBL (Figure 1).6Gaulton A. Hersey A. Nowotka M. Bento A.P. Chambers J. Mendez D. Mutowo P. Atkinson F. Bellis L.J. Cibrián-Uhalte E. et al.The ChEMBL database in 2017.Nucleic Acids Res. 2017; 45: D945-D954Crossref PubMed Scopus (1114) Google Scholar To emphasize the critical nature of quality and complete data, we extracted composition, physicochemical, pharmacokinetics, and dosing information for iron oxide nanoparticles reported in 315 research manuscripts published between 2006 and 2019 in reputed nanotechnology journals. From those, 68% did not report the size, shape, or zeta potential of the nanoparticles. Further, only 1% and 31% presented a pharmacokinetics profile (with elimination/distribution half-lives and delivery efficiency) and dosing information (route and dose), respectively. The observed trends are apparently transferred to other nanoparticle types. In 322 gold and 257 silica nanoparticle research manuscripts, we found 51% and 45% of them missing full physicochemical characterization, respectively. Identical percentages of missing pharmacokinetics and dosing data were found. Together, this highlights a deep-rooted limitation in the nanotechnology field that must be addressed. Until then, inputting or discarding potentially valuable information will be required for modeling, which is far from ideal.While a publicly available resource will likely remain inaccessible in the coming years—even with adequate reporting standards in place—we envisage that a continuous and concerted community effort will be key toward that end. Those efforts might be further assisted by natural language processing and deep learning techniques with the goal of accelerating the extraction of information from the scientific literature.7Öztürk H. Özgür A. Schwaller P. Laino T. Ozkirimli E. Exploring chemical space using natural language processing methodologies for drug discovery.Drug Discov. Today. 2020; 25: 689-705Crossref PubMed Scopus (36) Google Scholar By encompassing multiple unexplored data patterns, those data resources are expected to endorse automated processes and support the implementation of ML tools that allow more efficient experiment prioritizations.Challenge 3: A machine-readable nanotechnology languageWhile predictive modeling8Tao H. Wu T. Aldeghi M. Wu T.C. Aspuru-Guzik A. Kumacheva E. Nanoparticle synthesis assisted by machine learning.Nat. Rev. Mater. 2021; 6: 701-716Crossref Scopus (44) Google Scholar,9Hart G.L.W. Mueller T. Toher C. Curtarolo S. Machine learning for alloys.Nat. Rev. Mater. 2021; 6: 730-755Crossref Scopus (51) Google Scholar can be executed with quality datasets and established heuristics, generative design of composite materials requires the development of new toolkits. In small-molecule discovery, the SMILES or SELFIES10Krenn M. Häse F. Nigam A. Friederich P. Aspuru-Guzik A. Self-referencing embedded strings (SELFIES): A 100% robust molecular string representation.Mach. Learn. Sci. Technol. 2020; 1: 045024Crossref Scopus (102) Google Scholar languages encode atom connectivity, which implicitly hardwires all physicochemical and biological properties for a given entity. By learning this language, a computer is then able to programmatically generate new words/molecules (as strings of characters) according to a probability distribution for each newly added character. In doing so, researchers become armed with a powerful means to virtually access new chemical matter and more efficiently explore a vast search space. We argue that a similar approach can be pursued to generate and tailor composite delivery materials (Figure 1). Considering that information on constituents, including which entities, their percentage, and/or their concentration, is key to determining all the underlying physicochemical and biological properties, it becomes essential to devise a new language and ontology for canonical representation of composite material systems—both already reported and imagined by a computer. Said language ought to holistically represent the nanomaterial and thus be transferable to any use case aside from the drug delivery systems we focus on here. Once this technology is in hand, the research community will gain access to an untapped concept for the de novo design of composite materials. If employed correctly, we expect that such ML models—which might be available in a short/medium term—could impact nanotechnology for healthcare similarly to how they are transforming discovery chemistry.Overall, we anticipate a gain of momentum for nanotechnology research and preview its future developments leveraged by ML concepts. The solutions we propose to the three outstanding challenges are realistic but surprisingly still not tackled by the research community. We expect the tight integration of computational technologies with robotics to result in a digital nanotechnology era that will see prototyping innovative and life-changing therapeutics done at a fraction of the time currently needed. Nanotechnology has seen numerous translational applications over diverse economy sectors, but only recently has it taken healthcare by storm.1Talebian S. Rodrigues T. das Neves J. Sarmento B. Langer R. Conde J. Facts and Figures on Materials Science and Nanotechnology Progress and Investment.ACS Nano. 2021; https://doi.org/10.1021/acsnano.1021c03992Crossref Google Scholar For example, the massively deployed Moderna and BioNTech/Pfizer COVID-19 vaccines use organic nanoparticles to deliver SARS-CoV-2 mRNA.2Talebian S. Conde J. Why Go NANO on COVID-19 Pandemic?.Matter. 2020; 3: 598-601Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar The pursuit of this novel immunization modality has positioned nanotechnology at the center of attention, providing ample clinical validation and further motivating its exploitation in disparate disease areas, e.g., cancer. In this regard, billions of dollars spent in basic/translational nanotechnology have, over the years, allowed a reasonable understanding of the design principles driving efficacy.1Talebian S. Rodrigues T. das Neves J. Sarmento B. Langer R. Conde J. Facts and Figures on Materials Science and Nanotechnology Progress and Investment.ACS Nano. 2021; https://doi.org/10.1021/acsnano.1021c03992Crossref Google Scholar Still, much remains to be explored. The anticipated shift to nanotechnology-centered molecular medicine urges the need for an innovative suite of computational tools that effectively harness the growing amount of information in this space. A data-driven (r)evolution, similar to what we are currently witnessing in chemistry and biology, might not be too distant. We argue that predictive modeling and the de novo design of composite nanodelivery systems will become a reality and ultimately endorse a new era in nanotechnology research. Herein, we critically discuss how machine learning (ML) can reshape next-generation drug delivery and the three challenges that must be addressed to enable continuous innovation through discriminative/generative nanotechnology. Challenge 1: Standardized data reportingQuality data are unavoidably the centerpiece of any ML tool, and the current lack of standardized reporting practices in nanobiotechnology and nanomedicine is a known issue (Figure 1).3Schrurs F. Lison D. Focusing the research efforts.Nat. Nanotechnol. 2012; 7: 546-548Crossref Scopus (83) Google Scholar,4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar This hinders reproducibility and meaningful comparative studies, despite a recent community effort to regulate and improve transparency in the disclosed materials.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar For example, the physicochemical properties (e.g., dimension, shape, surface charge, targeting agent density, and composition), administered dose, and loading in drug delivery systems are cornerstones to modulate pharmacokinetics and efficacy. However, their reporting heterogeneity or lack of explicit information in manuscripts jeopardizes the gain of momentum in nanomedical research.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar Further, we argue the exact composition, injected volume, concentration, and route of administration need to be accurately reported but are only sparingly discriminated. Multiple studies also describe the amount of only one component in the delivery system (e.g., iron or encapsulated drug dosage). Thus, normalizing the others by body weight in in vivo studies is virtually impossible. Paralleling deficiencies in the characterization of nanodelivery materials, one equally finds shortcomings in the report of assay endpoints. Delivery efficacy and tumor accumulation are usually provided as a percentage of initial dose/tumor mass in gram (%ID/g). This is only useful if the initial dose and tumor mass are also reported, which rarely is the case. Further, the percentage of tumor reduction constitutes an endpoint normalization, yielding no tractable information on either delivery efficiency or actual volume change. Finally, one must bear in mind that animal models are a proxy of an actual disease setting, and experiments must be carefully designed to ensure meaningfulness. Otherwise, they may inappropriately reproduce the disease and tumor microenvironment, as in the use of xenograft/allograft heterotopic models instead of orthotopic ones (e.g., lung cells injected subcutaneously).Overall, the design of nanomaterials is intricate, and the established practices in experimental characterization are manifestly insufficient to support translation in healthcare on a wider scale.5Conde J. Above and Beyond Cancer Therapy: Translating Biomaterials into the Clinic.Trends Cancer. 2020; 6: 730-732Abstract Full Text Full Text PDF Scopus (7) Google Scholar Perpetuating those practices will ultimately curb innovation and research throughput. We envisage that a shift to robust reporting standards will be instrumental for automating the development of nanodelivery materials. Quality data are unavoidably the centerpiece of any ML tool, and the current lack of standardized reporting practices in nanobiotechnology and nanomedicine is a known issue (Figure 1).3Schrurs F. Lison D. Focusing the research efforts.Nat. Nanotechnol. 2012; 7: 546-548Crossref Scopus (83) Google Scholar,4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar This hinders reproducibility and meaningful comparative studies, despite a recent community effort to regulate and improve transparency in the disclosed materials.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar For example, the physicochemical properties (e.g., dimension, shape, surface charge, targeting agent density, and composition), administered dose, and loading in drug delivery systems are cornerstones to modulate pharmacokinetics and efficacy. However, their reporting heterogeneity or lack of explicit information in manuscripts jeopardizes the gain of momentum in nanomedical research.4Faria M. Björnmalm M. Thurecht K.J. Kent S.J. Parton R.G. Kavallaris M. Johnston A.P.R. Gooding J.J. Corrie S.R. Boyd B.J. et al.Minimum information reporting in bio-nano experimental literature.Nat. Nanotechnol. 2018; 13: 777-785Crossref PubMed Scopus (349) Google Scholar Further, we argue the exact composition, injected volume, concentration, and route of administration need to be accurately reported but are only sparingly discriminated. Multiple studies also describe the amount of only one component in the delivery system (e.g., iron or encapsulated drug dosage). Thus, normalizing the others by body weight in in vivo studies is virtually impossible. Paralleling deficiencies in the characterization of nanodelivery materials, one equally finds shortcomings in the report of assay endpoints. Delivery efficacy and tumor accumulation are usually provided as a percentage of initial dose/tumor mass in gram (%ID/g). This is only useful if the initial dose and tumor mass are also reported, which rarely is the case. Further, the percentage of tumor reduction constitutes an endpoint normalization, yielding no tractable information on either delivery efficiency or actual volume change. Finally, one must bear in mind that animal models are a proxy of an actual disease setting, and experiments must be carefully designed to ensure meaningfulness. Otherwise, they may inappropriately reproduce the disease and tumor microenvironment, as in the use of xenograft/allograft heterotopic models instead of orthotopic ones (e.g., lung cells injected subcutaneously). Overall, the design of nanomaterials is intricate, and the established practices in experimental characterization are manifestly insufficient to support translation in healthcare on a wider scale.5Conde J. Above and Beyond Cancer Therapy: Translating Biomaterials into the Clinic.Trends Cancer. 2020; 6: 730-732Abstract Full Text Full Text PDF Scopus (7) Google Scholar Perpetuating those practices will ultimately curb innovation and research throughput. We envisage that a shift to robust reporting standards will be instrumental for automating the development of nanodelivery materials. Challenge 2: Complete datasetsReporting completeness, as urged above, will then enable the construction of nanotechnology databases akin to ChEMBL (Figure 1).6Gaulton A. Hersey A. Nowotka M. Bento A.P. Chambers J. Mendez D. Mutowo P. Atkinson F. Bellis L.J. Cibrián-Uhalte E. et al.The ChEMBL database in 2017.Nucleic Acids Res. 2017; 45: D945-D954Crossref PubMed Scopus (1114) Google Scholar To emphasize the critical nature of quality and complete data, we extracted composition, physicochemical, pharmacokinetics, and dosing information for iron oxide nanoparticles reported in 315 research manuscripts published between 2006 and 2019 in reputed nanotechnology journals. From those, 68% did not report the size, shape, or zeta potential of the nanoparticles. Further, only 1% and 31% presented a pharmacokinetics profile (with elimination/distribution half-lives and delivery efficiency) and dosing information (route and dose), respectively. The observed trends are apparently transferred to other nanoparticle types. In 322 gold and 257 silica nanoparticle research manuscripts, we found 51% and 45% of them missing full physicochemical characterization, respectively. Identical percentages of missing pharmacokinetics and dosing data were found. Together, this highlights a deep-rooted limitation in the nanotechnology field that must be addressed. Until then, inputting or discarding potentially valuable information will be required for modeling, which is far from ideal.While a publicly available resource will likely remain inaccessible in the coming years—even with adequate reporting standards in place—we envisage that a continuous and concerted community effort will be key toward that end. Those efforts might be further assisted by natural language processing and deep learning techniques with the goal of accelerating the extraction of information from the scientific literature.7Öztürk H. Özgür A. Schwaller P. Laino T. Ozkirimli E. Exploring chemical space using natural language processing methodologies for drug discovery.Drug Discov. Today. 2020; 25: 689-705Crossref PubMed Scopus (36) Google Scholar By encompassing multiple unexplored data patterns, those data resources are expected to endorse automated processes and support the implementation of ML tools that allow more efficient experiment prioritizations. Reporting completeness, as urged above, will then enable the construction of nanotechnology databases akin to ChEMBL (Figure 1).6Gaulton A. Hersey A. Nowotka M. Bento A.P. Chambers J. Mendez D. Mutowo P. Atkinson F. Bellis L.J. Cibrián-Uhalte E. et al.The ChEMBL database in 2017.Nucleic Acids Res. 2017; 45: D945-D954Crossref PubMed Scopus (1114) Google Scholar To emphasize the critical nature of quality and complete data, we extracted composition, physicochemical, pharmacokinetics, and dosing information for iron oxide nanoparticles reported in 315 research manuscripts published between 2006 and 2019 in reputed nanotechnology journals. From those, 68% did not report the size, shape, or zeta potential of the nanoparticles. Further, only 1% and 31% presented a pharmacokinetics profile (with elimination/distribution half-lives and delivery efficiency) and dosing information (route and dose), respectively. The observed trends are apparently transferred to other nanoparticle types. In 322 gold and 257 silica nanoparticle research manuscripts, we found 51% and 45% of them missing full physicochemical characterization, respectively. Identical percentages of missing pharmacokinetics and dosing data were found. Together, this highlights a deep-rooted limitation in the nanotechnology field that must be addressed. Until then, inputting or discarding potentially valuable information will be required for modeling, which is far from ideal. While a publicly available resource will likely remain inaccessible in the coming years—even with adequate reporting standards in place—we envisage that a continuous and concerted community effort will be key toward that end. Those efforts might be further assisted by natural language processing and deep learning techniques with the goal of accelerating the extraction of information from the scientific literature.7Öztürk H. Özgür A. Schwaller P. Laino T. Ozkirimli E. Exploring chemical space using natural language processing methodologies for drug discovery.Drug Discov. Today. 2020; 25: 689-705Crossref PubMed Scopus (36) Google Scholar By encompassing multiple unexplored data patterns, those data resources are expected to endorse automated processes and support the implementation of ML tools that allow more efficient experiment prioritizations. Challenge 3: A machine-readable nanotechnology languageWhile predictive modeling8Tao H. Wu T. Aldeghi M. Wu T.C. Aspuru-Guzik A. Kumacheva E. Nanoparticle synthesis assisted by machine learning.Nat. Rev. Mater. 2021; 6: 701-716Crossref Scopus (44) Google Scholar,9Hart G.L.W. Mueller T. Toher C. Curtarolo S. Machine learning for alloys.Nat. Rev. Mater. 2021; 6: 730-755Crossref Scopus (51) Google Scholar can be executed with quality datasets and established heuristics, generative design of composite materials requires the development of new toolkits. In small-molecule discovery, the SMILES or SELFIES10Krenn M. Häse F. Nigam A. Friederich P. Aspuru-Guzik A. Self-referencing embedded strings (SELFIES): A 100% robust molecular string representation.Mach. Learn. Sci. Technol. 2020; 1: 045024Crossref Scopus (102) Google Scholar languages encode atom connectivity, which implicitly hardwires all physicochemical and biological properties for a given entity. By learning this language, a computer is then able to programmatically generate new words/molecules (as strings of characters) according to a probability distribution for each newly added character. In doing so, researchers become armed with a powerful means to virtually access new chemical matter and more efficiently explore a vast search space. We argue that a similar approach can be pursued to generate and tailor composite delivery materials (Figure 1). Considering that information on constituents, including which entities, their percentage, and/or their concentration, is key to determining all the underlying physicochemical and biological properties, it becomes essential to devise a new language and ontology for canonical representation of composite material systems—both already reported and imagined by a computer. Said language ought to holistically represent the nanomaterial and thus be transferable to any use case aside from the drug delivery systems we focus on here. Once this technology is in hand, the research community will gain access to an untapped concept for the de novo design of composite materials. If employed correctly, we expect that such ML models—which might be available in a short/medium term—could impact nanotechnology for healthcare similarly to how they are transforming discovery chemistry.Overall, we anticipate a gain of momentum for nanotechnology research and preview its future developments leveraged by ML concepts. The solutions we propose to the three outstanding challenges are realistic but surprisingly still not tackled by the research community. We expect the tight integration of computational technologies with robotics to result in a digital nanotechnology era that will see prototyping innovative and life-changing therapeutics done at a fraction of the time currently needed. While predictive modeling8Tao H. Wu T. Aldeghi M. Wu T.C. Aspuru-Guzik A. Kumacheva E. Nanoparticle synthesis assisted by machine learning.Nat. Rev. Mater. 2021; 6: 701-716Crossref Scopus (44) Google Scholar,9Hart G.L.W. Mueller T. Toher C. Curtarolo S. Machine learning for alloys.Nat. Rev. Mater. 2021; 6: 730-755Crossref Scopus (51) Google Scholar can be executed with quality datasets and established heuristics, generative design of composite materials requires the development of new toolkits. In small-molecule discovery, the SMILES or SELFIES10Krenn M. Häse F. Nigam A. Friederich P. Aspuru-Guzik A. Self-referencing embedded strings (SELFIES): A 100% robust molecular string representation.Mach. Learn. Sci. Technol. 2020; 1: 045024Crossref Scopus (102) Google Scholar languages encode atom connectivity, which implicitly hardwires all physicochemical and biological properties for a given entity. By learning this language, a computer is then able to programmatically generate new words/molecules (as strings of characters) according to a probability distribution for each newly added character. In doing so, researchers become armed with a powerful means to virtually access new chemical matter and more efficiently explore a vast search space. We argue that a similar approach can be pursued to generate and tailor composite delivery materials (Figure 1). Considering that information on constituents, including which entities, their percentage, and/or their concentration, is key to determining all the underlying physicochemical and biological properties, it becomes essential to devise a new language and ontology for canonical representation of composite material systems—both already reported and imagined by a computer. Said language ought to holistically represent the nanomaterial and thus be transferable to any use case aside from the drug delivery systems we focus on here. Once this technology is in hand, the research community will gain access to an untapped concept for the de novo design of composite materials. If employed correctly, we expect that such ML models—which might be available in a short/medium term—could impact nanotechnology for healthcare similarly to how they are transforming discovery chemistry. Overall, we anticipate a gain of momentum for nanotechnology research and preview its future developments leveraged by ML concepts. The solutions we propose to the three outstanding challenges are realistic but surprisingly still not tackled by the research community. We expect the tight integration of computational technologies with robotics to result in a digital nanotechnology era that will see prototyping innovative and life-changing therapeutics done at a fraction of the time currently needed. The authors acknowledge financial support from FCT Portugal in the framework of PhD grant 2020.06638.BD (to D.P.S.), and the European Research Council grant agreement 848325 (J. Conde for the ERC Starting Grant). T.R. is an Investigador Auxiliar supported by FCT Portugal ( CEECIND/00684/2018 ). A.L., B.B.M., J. Conniot, and D.P.S. collected and analyzed data. J. Conde and T.R. conceived and supervised the research. All authors contributed to writing the manuscript and agreed on its final version. J. Conde and T.R. are co-founders and shareholders of TargTex S.A.

Correction for ‘Revisiting gene delivery to the brain: silencing and editing’ by João Conniot et al., Biomater. Sci., 2021, DOI: 10.1039/D0BM01278C.

Colorectal cancer (CRC) is among the five most commonly diagnosed cancers worldwide, constituting 6% of all cancers and the third leading cause of cancer death. CRC is the third and second most frequent cancer in men and women worldwide, accounting for 14% and 13% of all cancer incidence rates, respectively. CRC incidence is decreasing in older populations, but it has been significantly rising worldwide in adolescents and adults younger than 50 years old. Significant advances in the screening methods and surgical procedures have been underlying the reduction of the CRC incidence rate in older populations. However, there is an urgent demand for the development of alternative effective therapeutic options to overcome advanced metastatic CRC, while preventing disease recurrence. This review addresses the immune and CRC biology, summarizing the recent advances on the immune and/or therapeutic regimens currently in clinical use. We will focus on the emerging role of nanotechnology in the development of combinational therapies targeting and thereby regulating the function of the major players in CRC progression and immune evasion.

A low response rate, acquired resistance and severe side effects have limited the clinical outcomes of immune checkpoint therapy. Here, we show that combining cancer nanovaccines with an anti-PD-1 antibody (αPD-1) for immunosuppression blockade and an anti-OX40 antibody (αOX40) for effector T-cell stimulation, expansion and survival can potentiate the efficacy of melanoma therapy. Prophylactic and therapeutic combination regimens of dendritic cell-targeted mannosylated nanovaccines with αPD-1/αOX40 demonstrate a synergism that stimulates T-cell infiltration into tumours at early treatment stages. However, this treatment at the therapeutic regimen does not result in an enhanced inhibition of tumour growth compared to αPD-1/αOX40 alone and is accompanied by an increased infiltration of myeloid-derived suppressor cells in tumours. Combining the double therapy with ibrutinib, a myeloid-derived suppressor cell inhibitor, leads to a remarkable tumour remission and prolonged survival in melanoma-bearing mice. The synergy between the mannosylated nanovaccines, ibrutinib and αPD-1/αOX40 provides essential insights to devise alternative regimens to improve the efficacy of immune checkpoint modulators in solid tumours by regulating the endogenous immune response.

The modulation of immune checkpoint receptors has been one of the most successful, exciting, and explored approaches for cancer immunotherapy. Currently, several immune checkpoint modulators, mainly monoclonal antibodies, are showing remarkable results. However, the failure to show a response in most patients and the induction of severe immune-related adverse effects are the major drawbacks. Novel approaches concerning the development of immune modulatory small molecules have emerged as an alternative. Nevertheless, the lack of structural information about immune checkpoint receptors has hindered the rational design of those small-molecule modulators by preventing the use of methodologies such as computer-aided drug design. Herein, we provide an overview and critical analysis of the structural and dynamic details of immune checkpoint receptors (cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4), programmed cell death protein 1 (PD-1), and glucocorticoid-induced TNFR-related protein (GITR)) and their interaction with known modulators. This knowledge is essential to advance the understanding of their binding mode and guide the design of novel effective targeted anticancer medicines.

Particulate delivery systems can protect entrapped material from chemical and enzymatic degradation, resulting in increased blood circulation time, by avoiding the uptake by the mononuclear phagocyte system and rapid clearance via the kidneys. In addition, those carriers allow the concomitant delivery of multiple components for a sustained release of the entrapped active molecules, prolonging their therapeutic effects. To achieve a specific therapeutic outcome, the scientific community has done considerable efforts on developing different strategies to target tissues and specific cells through the development of site-directed nanocarriers. By modulating nanoparticle (NP) size, surface charge, or hydrophobicity, it is possible to regulate the endocytic pathways and facilitate endosomal escape, leading to the cytosolic delivery of therapeutic molecules. The modification of NPs by organelle-specific targeting macromolecules (drugs, proteins, DNA, short interference RNA, among others) has an extreme potential for the delivery of molecules to intracellular target receptors, constituting a particularly important strategy to develop nanomedicines with extended efficacy and specificity. This chapter addresses the current strategies explored to achieve the delivery and accumulation of bioactive molecules to targeted organelles by nanotechnology-based systems.

Poly(lactic acid) (PLA) is one of the most successful and versatile polymers explored for controlled delivery of bioactive molecules. Its attractive properties of biodegradability and biocompatibility in vivo have contributed in a meaningful way to the approval of different products by the FDA and EMA for a wide range of biomedical and pharmaceutical applications, in the past two decades. This polymer has been widely used for the preparation of particles as delivery systems of several therapeutic molecules, including vaccines. These PLA vaccine carriers have shown to induce a sustained and targeted release of different bacterial, viral and tumor-associated antigens and adjuvants in vivo, triggering distinct immune responses. The present review intends to highlight and discuss the major advantages of PLA as a promising polymer for the development of potent vaccine delivery systems against pathogens and cancer. It aims to provide a critical discussion based on preclinical data to better understand the major effect of PLA-based carrier properties on their interaction with immune cells and thus their role in the modulation of host immunity.Statement of SignificanceDuring the last decades, vaccination has had a great impact on global health with the control of many severe diseases. Polymeric nanosystems have emerged as promising strategies to stabilize vaccine antigens, promoting their controlled release to phagocytic cells, thus avoiding the need for multiple administrations. One of the most promising polymers are the aliphatic polyesters, which include the poly(lactic acid). This is a highly versatile biodegradable and biocompatible polymer. Products containing this polymer have already been approved for all food and some biomedical applications. Despite all favorable characteristics presented above, PLA has been less intensively discussed than other polymers, such as its copolymer PLGA, including regarding its application in vaccination and particularly in tumor immunotherapy. The present review discusses the major advantages of poly(lactic acid) for the development of potent vaccine delivery systems, providing a critical view on the main properties that determine their effect on the modulation of immune cells. (C) 2016 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Nanotechnology-based strategies can dramatically impact the treatment, prevention and diagnosis of a wide range of diseases. Despite the unprecedented success achieved with the use of nanomaterials to address unmet biomedical needs and their particular suitability for the effective application of a personalized medicine, the clinical translation of those nanoparticulate systems has still been impaired by the limited understanding on their interaction with complex biological systems. As a result, unexpected effects due to unpredicted interactions at biomaterial and biological interfaces have been underlying the biosafety concerns raised by the use of nanomaterials. This review explores the current knowledge on how nanoparticle (NP) physicochemical and surface properties determine their interactions with innate immune cells, with particular attention on the activation of pattern-recognition receptors and inflammasome. A critical perspective will additionally address the impact of biological systems on the effect of NP on immune cell activity at the molecular level. We will discuss how the understanding of the NP-innate immune cell interactions can significantly add into the clinical translation by guiding the design of nanomedicines with particular effect on targeted cells, thus improving their clinical efficacy while minimizing undesired but predictable toxicological effects.

Nanomedicines have been in the forefront of pharmaceutical research in the last decades, creating new challenges for research community, industry, and regulators. There is a strong demand for the fast development of scientific and technological tools to address unmet medical needs, thus improving human health care and life quality. Tremendous advances in the biomaterials and nanotechnology fields have prompted their use as promising tools to overcome important drawbacks, mostly associated to the non-specific effects of conventional therapeutic approaches. However, the wide range of application of nanomedicines demands a profound knowledge and characterization of these complex products. Their properties need to be extensively understood to avoid unpredicted effects on patients, such as potential immune reactivity. Research policy and alliances have been bringing together scientists, regulators, industry, and, more frequently in recent years, patient representatives and patient advocacy institutions. In order to successfully enhance the development of new technologies, improved strategies for research-based corporate organizations, more integrated research tools dealing with appropriate translational requirements aiming at clinical development, and proactive regulatory policies are essential in the near future. This review focuses on the most important aspects currently recognized as key factors for the regulation of nanomedicines, discussing the efforts under development by industry and regulatory agencies to promote their translation into the market. Regulatory Science aspects driving a faster and safer development of nanomedicines will be a central issue for the next years.