Objetivo: O objetivo do presente estudo foi descrever a relação entre as concentrações plasmáticas totais e livres tumorais do etoposídeo (ETO) e a inibição do crescimento do tumor observada em ratos Wistar portadores de tumor Walker- 256 (W256) utilizando a modelagem farmacocinética/farmacodinâmica (PK/PD). Métodos: Os procedimentos com animais foram aprovados no CEUA/UFRGS sob o número 22302. Os experimentos de farmacocinética foram realizados para determinar concentrações plasmáticas e livres em duas regiões do tumor sólido W256 através de microdiálise. Após a administração do ETO nas doses de 10 ou 20 mg/kg i.v. bolus em ratos Wistar portadores de tumor W256, amostras de sangue e microdialisado de tecido do centro e periferia do tumor foram coletadas simultaneamente, até 7 h pós-dose, para determinar o fator de penetração no tumor. Um método analítico por CLAE-UV foi desenvolvido e validado para quantificação do etoposídeo nas amostras de plasma e dialisado. Os experimentos de farmacodinâmica foram conduzidos em ratos portadores de tumor W256 que receberam ETO 5 e 10 mg/kg i.v. bolus uma vez ao dia por 8 e 4 dias, respectivamente. O volume dos tumores foram monitorados diariamente durante 30 dias. Análise não-compartimental dos dados de PK foi realizada no WinNonlin®. A modelagem dos dados PK e PK/PD foi realizada no Monolix®, utilizando abordagem populacional. Os dados PK/PD foram analisados usando o modelo Simeoni TGI modificado através da introdução de uma função Emax para descrever a relação nãolinear entre a concentração plasmática e tumoral e o efeito. Resultados e Discussão: O método por CLAE-UV foi desenvolvido e validado para quantificar as amostras de ETO em plasma e tecido. A penetração do ETO no tumor foi maior na periferia (61 ± 15 % e 61 ± 29 %) do que no centro do tumor (34 ± 6 % e 28 ± 11 %) após administração das doses 10 e 20 mg/kg, respectivamente (ANOVA, α = 0.05). Um modelo de 4 compartimentos compreendendo uma distribuição saturável (cinética de Michaelis-Menten) nos compartimentos tumorais a partir do compartimento central modelou simultaneamente os perfis de concentração-tempo do ETO em plasma e em ambas regiões do tumor. O modelo populacional PK/PD Simeoni TGI–Emax foi capaz de descrever o efeito antitumoral dependente do regime de administração do ETO utilizando concentrações totais plasmáticas ou livres no tumor, resultando em um maior k2max (potência máxima) para as concentrações livres (25,8 mL.μg-1.dia-1 - intratumoral vs. 12,6 mL.μg-1.dia-1 - plasma total). Conclusões: Os resultados mostram que a utilização das concentrações livres do fármaco no tumor para a modelagem PK/PD pode fornecer um melhor entendimento da relação farmacocinética e farmacodinâmica e melhoram a capacidade de previsão do modelo, considerando que a eficácia dos fármacos antineoplásicos no tratamento de tumores sólidos é dependente da capacidade do fármaco em se distribuir no tecido tumoral. / Objective: The aim of this study was to describe the relationship between total plasma and free interstitial tumor etoposide (ETO) concentrations and the drug tumor growth inhibition observed in a Walker-256 (W256) tumor-bearing Wistar rat model using the pharmacokinetic/pharmacodynamic (PK/PD) modeling. Methods: The experiments with animals were approved by CEUA/UFRGS (protocol number 22302). Pharmacokinetic experiments were conducted to determine total plasma and free intratumoral concentrations in two regions of W256 solid tumor by microdialysis. After administration of ETO 10 or 20 mg/kg i.v. bolus to W256 tumorbearing Wistar rats, blood and tissue microdialysate samples from tumor center and periphery were simultaneously collected up to 7h to determine the tumor penetration factor. An analytical HPLC-UV method was developed and validated for quantification of ETO in plasma and microdialysate samples. The pharmacodynamic experiments were conducted in W256 tumor-bearing rats that received ETO 5 or 10 mg/kg i.v. bolus every day for 8 and 4 days, respectively. Tumor volumes were monitored daily for 30 days. Non-compartmental analysis of PK data was performed in WinNonlin®. The PK and PK/PD modeling by population approach were performed using Monolix®. PK/PD data were analyzed using a modification of Simeoni TGI model by introducing an Emax function to describe the nonlinear relationship between tumor and plasma concentrations and effect. Results and Discussion: The HLPCUV method was developed and validated to determine plasma and tissue samples of ETO. ETO tumor penetration was higher in the tumor periphery (61 ± 15 % and 61 ± 29 %) than center (34 ± 6 % and 28 ± 11 %) following 10 and 20 mg/kg doses, respectively (ANOVA, α = 0.05). A 4-compartment structural model comprising a saturable distribution (Michaelis-Menten kinetics) into the tumor compartments from the central compartment simultaneously described the ETO concentration–time profiles in plasma and both tumor regions. The PK/PD population Simeoni TGI–Emax model was capable of describing the schedule-dependent antitumor effects of ETO using total plasma or free tumor concentrations obtained in a W256-tumor bearing Wistar rat model, resulting in higher k2max (maximal potency) for free concentrations (25.8 mL.μg-1.day-1 - intratumoral vs. 12.6 mL.μg-1.day-1 total plasma). Conclusions: The results showed that the use of free intratumoral drug concentrations in the PK/PD modeling can provide a better understanding of the pharmacokinetics and pharmacodynamics relationship and improve the forecasting ability of the models considering that the efficacy of antineoplastic drugs in the treatment of solid tumors is dependent on the drug ability to distribute into the tumor.

Objetivo: O objetivo do presente estudo foi descrever a relação entre as concentrações plasmáticas totais e livres tumorais do etoposídeo (ETO) e a inibição do crescimento do tumor observada em ratos Wistar portadores de tumor Walker- 256 (W256) utilizando a modelagem farmacocinética/farmacodinâmica (PK/PD). Métodos: Os procedimentos com animais foram aprovados no CEUA/UFRGS sob o número 22302. Os experimentos de farmacocinética foram realizados para determinar concentrações plasmáticas e livres em duas regiões do tumor sólido W256 através de microdiálise. Após a administração do ETO nas doses de 10 ou 20 mg/kg i.v. bolus em ratos Wistar portadores de tumor W256, amostras de sangue e microdialisado de tecido do centro e periferia do tumor foram coletadas simultaneamente, até 7 h pós-dose, para determinar o fator de penetração no tumor. Um método analítico por CLAE-UV foi desenvolvido e validado para quantificação do etoposídeo nas amostras de plasma e dialisado. Os experimentos de farmacodinâmica foram conduzidos em ratos portadores de tumor W256 que receberam ETO 5 e 10 mg/kg i.v. bolus uma vez ao dia por 8 e 4 dias, respectivamente. O volume dos tumores foram monitorados diariamente durante 30 dias. Análise não-compartimental dos dados de PK foi realizada no WinNonlin®. A modelagem dos dados PK e PK/PD foi realizada no Monolix®, utilizando abordagem populacional. Os dados PK/PD foram analisados usando o modelo Simeoni TGI modificado através da introdução de uma função Emax para descrever a relação nãolinear entre a concentração plasmática e tumoral e o efeito. Resultados e Discussão: O método por CLAE-UV foi desenvolvido e validado para quantificar as amostras de ETO em plasma e tecido. A penetração do ETO no tumor foi maior na periferia (61 ± 15 % e 61 ± 29 %) do que no centro do tumor (34 ± 6 % e 28 ± 11 %) após administração das doses 10 e 20 mg/kg, respectivamente (ANOVA, α = 0.05). Um modelo de 4 compartimentos compreendendo uma distribuição saturável (cinética de Michaelis-Menten) nos compartimentos tumorais a partir do compartimento central modelou simultaneamente os perfis de concentração-tempo do ETO em plasma e em ambas regiões do tumor. O modelo populacional PK/PD Simeoni TGI–Emax foi capaz de descrever o efeito antitumoral dependente do regime de administração do ETO utilizando concentrações totais plasmáticas ou livres no tumor, resultando em um maior k2max (potência máxima) para as concentrações livres (25,8 mL.μg-1.dia-1 - intratumoral vs. 12,6 mL.μg-1.dia-1 - plasma total). Conclusões: Os resultados mostram que a utilização das concentrações livres do fármaco no tumor para a modelagem PK/PD pode fornecer um melhor entendimento da relação farmacocinética e farmacodinâmica e melhoram a capacidade de previsão do modelo, considerando que a eficácia dos fármacos antineoplásicos no tratamento de tumores sólidos é dependente da capacidade do fármaco em se distribuir no tecido tumoral. / Objective: The aim of this study was to describe the relationship between total plasma and free interstitial tumor etoposide (ETO) concentrations and the drug tumor growth inhibition observed in a Walker-256 (W256) tumor-bearing Wistar rat model using the pharmacokinetic/pharmacodynamic (PK/PD) modeling. Methods: The experiments with animals were approved by CEUA/UFRGS (protocol number 22302). Pharmacokinetic experiments were conducted to determine total plasma and free intratumoral concentrations in two regions of W256 solid tumor by microdialysis. After administration of ETO 10 or 20 mg/kg i.v. bolus to W256 tumorbearing Wistar rats, blood and tissue microdialysate samples from tumor center and periphery were simultaneously collected up to 7h to determine the tumor penetration factor. An analytical HPLC-UV method was developed and validated for quantification of ETO in plasma and microdialysate samples. The pharmacodynamic experiments were conducted in W256 tumor-bearing rats that received ETO 5 or 10 mg/kg i.v. bolus every day for 8 and 4 days, respectively. Tumor volumes were monitored daily for 30 days. Non-compartmental analysis of PK data was performed in WinNonlin®. The PK and PK/PD modeling by population approach were performed using Monolix®. PK/PD data were analyzed using a modification of Simeoni TGI model by introducing an Emax function to describe the nonlinear relationship between tumor and plasma concentrations and effect. Results and Discussion: The HLPCUV method was developed and validated to determine plasma and tissue samples of ETO. ETO tumor penetration was higher in the tumor periphery (61 ± 15 % and 61 ± 29 %) than center (34 ± 6 % and 28 ± 11 %) following 10 and 20 mg/kg doses, respectively (ANOVA, α = 0.05). A 4-compartment structural model comprising a saturable distribution (Michaelis-Menten kinetics) into the tumor compartments from the central compartment simultaneously described the ETO concentration–time profiles in plasma and both tumor regions. The PK/PD population Simeoni TGI–Emax model was capable of describing the schedule-dependent antitumor effects of ETO using total plasma or free tumor concentrations obtained in a W256-tumor bearing Wistar rat model, resulting in higher k2max (maximal potency) for free concentrations (25.8 mL.μg-1.day-1 - intratumoral vs. 12.6 mL.μg-1.day-1 total plasma). Conclusions: The results showed that the use of free intratumoral drug concentrations in the PK/PD modeling can provide a better understanding of the pharmacokinetics and pharmacodynamics relationship and improve the forecasting ability of the models considering that the efficacy of antineoplastic drugs in the treatment of solid tumors is dependent on the drug ability to distribute into the tumor.

Objetivo: O objetivo do presente estudo foi descrever a relação entre as concentrações plasmáticas totais e livres tumorais do etoposídeo (ETO) e a inibição do crescimento do tumor observada em ratos Wistar portadores de tumor Walker- 256 (W256) utilizando a modelagem farmacocinética/farmacodinâmica (PK/PD). Métodos: Os procedimentos com animais foram aprovados no CEUA/UFRGS sob o número 22302. Os experimentos de farmacocinética foram realizados para determinar concentrações plasmáticas e livres em duas regiões do tumor sólido W256 através de microdiálise. Após a administração do ETO nas doses de 10 ou 20 mg/kg i.v. bolus em ratos Wistar portadores de tumor W256, amostras de sangue e microdialisado de tecido do centro e periferia do tumor foram coletadas simultaneamente, até 7 h pós-dose, para determinar o fator de penetração no tumor. Um método analítico por CLAE-UV foi desenvolvido e validado para quantificação do etoposídeo nas amostras de plasma e dialisado. Os experimentos de farmacodinâmica foram conduzidos em ratos portadores de tumor W256 que receberam ETO 5 e 10 mg/kg i.v. bolus uma vez ao dia por 8 e 4 dias, respectivamente. O volume dos tumores foram monitorados diariamente durante 30 dias. Análise não-compartimental dos dados de PK foi realizada no WinNonlin®. A modelagem dos dados PK e PK/PD foi realizada no Monolix®, utilizando abordagem populacional. Os dados PK/PD foram analisados usando o modelo Simeoni TGI modificado através da introdução de uma função Emax para descrever a relação nãolinear entre a concentração plasmática e tumoral e o efeito. Resultados e Discussão: O método por CLAE-UV foi desenvolvido e validado para quantificar as amostras de ETO em plasma e tecido. A penetração do ETO no tumor foi maior na periferia (61 ± 15 % e 61 ± 29 %) do que no centro do tumor (34 ± 6 % e 28 ± 11 %) após administração das doses 10 e 20 mg/kg, respectivamente (ANOVA, α = 0.05). Um modelo de 4 compartimentos compreendendo uma distribuição saturável (cinética de Michaelis-Menten) nos compartimentos tumorais a partir do compartimento central modelou simultaneamente os perfis de concentração-tempo do ETO em plasma e em ambas regiões do tumor. O modelo populacional PK/PD Simeoni TGI–Emax foi capaz de descrever o efeito antitumoral dependente do regime de administração do ETO utilizando concentrações totais plasmáticas ou livres no tumor, resultando em um maior k2max (potência máxima) para as concentrações livres (25,8 mL.μg-1.dia-1 - intratumoral vs. 12,6 mL.μg-1.dia-1 - plasma total). Conclusões: Os resultados mostram que a utilização das concentrações livres do fármaco no tumor para a modelagem PK/PD pode fornecer um melhor entendimento da relação farmacocinética e farmacodinâmica e melhoram a capacidade de previsão do modelo, considerando que a eficácia dos fármacos antineoplásicos no tratamento de tumores sólidos é dependente da capacidade do fármaco em se distribuir no tecido tumoral. / Objective: The aim of this study was to describe the relationship between total plasma and free interstitial tumor etoposide (ETO) concentrations and the drug tumor growth inhibition observed in a Walker-256 (W256) tumor-bearing Wistar rat model using the pharmacokinetic/pharmacodynamic (PK/PD) modeling. Methods: The experiments with animals were approved by CEUA/UFRGS (protocol number 22302). Pharmacokinetic experiments were conducted to determine total plasma and free intratumoral concentrations in two regions of W256 solid tumor by microdialysis. After administration of ETO 10 or 20 mg/kg i.v. bolus to W256 tumorbearing Wistar rats, blood and tissue microdialysate samples from tumor center and periphery were simultaneously collected up to 7h to determine the tumor penetration factor. An analytical HPLC-UV method was developed and validated for quantification of ETO in plasma and microdialysate samples. The pharmacodynamic experiments were conducted in W256 tumor-bearing rats that received ETO 5 or 10 mg/kg i.v. bolus every day for 8 and 4 days, respectively. Tumor volumes were monitored daily for 30 days. Non-compartmental analysis of PK data was performed in WinNonlin®. The PK and PK/PD modeling by population approach were performed using Monolix®. PK/PD data were analyzed using a modification of Simeoni TGI model by introducing an Emax function to describe the nonlinear relationship between tumor and plasma concentrations and effect. Results and Discussion: The HLPCUV method was developed and validated to determine plasma and tissue samples of ETO. ETO tumor penetration was higher in the tumor periphery (61 ± 15 % and 61 ± 29 %) than center (34 ± 6 % and 28 ± 11 %) following 10 and 20 mg/kg doses, respectively (ANOVA, α = 0.05). A 4-compartment structural model comprising a saturable distribution (Michaelis-Menten kinetics) into the tumor compartments from the central compartment simultaneously described the ETO concentration–time profiles in plasma and both tumor regions. The PK/PD population Simeoni TGI–Emax model was capable of describing the schedule-dependent antitumor effects of ETO using total plasma or free tumor concentrations obtained in a W256-tumor bearing Wistar rat model, resulting in higher k2max (maximal potency) for free concentrations (25.8 mL.μg-1.day-1 - intratumoral vs. 12.6 mL.μg-1.day-1 total plasma). Conclusions: The results showed that the use of free intratumoral drug concentrations in the PK/PD modeling can provide a better understanding of the pharmacokinetics and pharmacodynamics relationship and improve the forecasting ability of the models considering that the efficacy of antineoplastic drugs in the treatment of solid tumors is dependent on the drug ability to distribute into the tumor.

No / Limited tumor penetrability of anti-cancer drugs is recognized as one of the major factors that lead to poor anti-tumor activity. SRJ09 (3,19-(2-bromobenzylidene) andrographolide) has been identified as a lead anti-cancer agent for colon cancer. Recently, this compound was shown by us to be a mutant K-Ras binder. In this present study, the penetrability of SRJ09 through the DLD-1 colon cancer multicell layer (MCL) was evaluated. The amount of SRJ09 that penetrated through the MCL was quantitated by utilizing high performance liquid chromatography (HPLC). Histopathological staining was used to visualize the morphology of MCL. A chemosensitivity assay was performed to assess the anti-cancer activity of SRJ09 in DLD-1 cells. SRJ09 was able to penetrate through DLD-1 MCL and is inversely proportional with the MCL thickness. The flow rates for SRJ09 through MCL were 0.90 ± 0.20 μM/min/cm2 and 0.56 ± 0.06 μM/min/cm2 for days 1 and 5, respectively, which are better than doxorubicin. Histopathological examination revealed that the integrity of the DLD-1 MCL was retained and no visible damage was inflicted on the cell membrane, confirming the penetration of SRJ09 was by diffusion. Short term exposure (1 h) in DLD-1 cells demonstrated SRJ09 had IC50 of 41 μM which was approximately 4-folds lower than andrographolide, the parent compound of SRJ09. In conclusion, SRJ09 successfully penetrated through DLD-1 MCL by diffusion and emerged as a potential candidate to be developed as a clinically viable anti-colon cancer drug.

Recent advancements in pharmaceutical technology based on principles of nanotechnology, polymer chemistry, and biomedical engineering have resulted in the creation of novel drug delivery systems with the potential to revolutionize current strategies in cancer chemotherapy. In oncology, realization of significant improvements in therapeutic efficacy requires minimization of drug exposure to healthy tissues and concentration of the drug within the tumor. As such, encapsulation of chemotherapeutic agents inside nanoparticles capable of enhancing tumor-targeted drug delivery is a particularly promising innovation. Yet, initial investigations into the intratumoral fate of nanomedicines have suggested that they may be heterogeneously distributed and achieve limited access to cancer cells located distant from the tumor vasculature. As such, uncovering the determinants of nanoparticle transport at the intratumoral level is critical to the development of optimized delivery vehicles capable of fully exploiting the therapeutic potential of nanomedicines. In this work, the chemotherapeutic agent, docetaxel (DTX), was incorporated into nano-sized, biocompatible PEG-b-PCL block copolymer micelles (BCMs). Encapsulation of DTX in micelles via chemical conjugation or physical entrapment resulted in a dramatic increase in drug solubility and customizable drug release rate. The use of multicellular tumor spheroids (MCTS) was established as a viable platform for assessing the efficacy and tumor tissue penetration of nanomedicines in vitro. A series of complementary assays was validated for analysis of DTX-loaded micelle (BCM+DTX) toxicity in monolayer and spheroid cultures relative to Taxotere®. Cells cultured as spheroids were less responsive to treatment relative to monolayer cultures due to mechanisms of drug resistance associated with structural and microenvironmental properties of the 3-D tissue. Computational, image-based methodologies were used to assess the spatial and temporal penetration of BCMs in spheroids and corresponding human tumor xenografts. Using this approach, the tumor penetration of micelles was found to be nanoparticle-size-, tumor tissue type- and time- dependent. Furthermore, spheroids were found to be a valuable platform for the prediction of trends in nanoparticle transport in vivo. Overall, the results reported herein serve to demonstrate important determinants of nanoparticle intratumoral transport and to establish computational in vitro and in vivo methodologies for the rational design and optimization of nanomedicines.

Recent advancements in pharmaceutical technology based on principles of nanotechnology, polymer chemistry, and biomedical engineering have resulted in the creation of novel drug delivery systems with the potential to revolutionize current strategies in cancer chemotherapy. In oncology, realization of significant improvements in therapeutic efficacy requires minimization of drug exposure to healthy tissues and concentration of the drug within the tumor. As such, encapsulation of chemotherapeutic agents inside nanoparticles capable of enhancing tumor-targeted drug delivery is a particularly promising innovation. Yet, initial investigations into the intratumoral fate of nanomedicines have suggested that they may be heterogeneously distributed and achieve limited access to cancer cells located distant from the tumor vasculature. As such, uncovering the determinants of nanoparticle transport at the intratumoral level is critical to the development of optimized delivery vehicles capable of fully exploiting the therapeutic potential of nanomedicines. In this work, the chemotherapeutic agent, docetaxel (DTX), was incorporated into nano-sized, biocompatible PEG-b-PCL block copolymer micelles (BCMs). Encapsulation of DTX in micelles via chemical conjugation or physical entrapment resulted in a dramatic increase in drug solubility and customizable drug release rate. The use of multicellular tumor spheroids (MCTS) was established as a viable platform for assessing the efficacy and tumor tissue penetration of nanomedicines in vitro. A series of complementary assays was validated for analysis of DTX-loaded micelle (BCM+DTX) toxicity in monolayer and spheroid cultures relative to Taxotere®. Cells cultured as spheroids were less responsive to treatment relative to monolayer cultures due to mechanisms of drug resistance associated with structural and microenvironmental properties of the 3-D tissue. Computational, image-based methodologies were used to assess the spatial and temporal penetration of BCMs in spheroids and corresponding human tumor xenografts. Using this approach, the tumor penetration of micelles was found to be nanoparticle-size-, tumor tissue type- and time- dependent. Furthermore, spheroids were found to be a valuable platform for the prediction of trends in nanoparticle transport in vivo. Overall, the results reported herein serve to demonstrate important determinants of nanoparticle intratumoral transport and to establish computational in vitro and in vivo methodologies for the rational design and optimization of nanomedicines.

Une faible biodisponibilité et une haute toxicité des médicaments anticancéreux, ajoutées à une résistance aux médicaments, constituent des obstacles majeurs pour le traitement du cancer. L'application des nanotechnologies pour la délivrance de médicaments est largement pressentie pour aborder ces problèmes. Premièrement, nous avons utilisé un peptide CRGDK comme ligand spécifique pour les cellules cancéreuses que nous avons conjugué au DSPE-PEG2000 pour préparer les nanomicelles encapsulant le médicament anticancéreux doxorubicine. Le peptide CRGDK conjugué aux nanomicelles provoque la liaison aux récepteurs à NRP-1, conduisant à l'absorption cellulaire spécifique et à l’amélioration de l'activité anticancéreuse in vitro. Les résultats in vivo ont également confirmé que les nanomicelles décorées de CRGDK pourraient efficacement pénétrer et s’accumuler dans les tumeurs profondes.Deuxièmement, nous avons a été consacrée à la mise au point de nanomicelles originales utilisant un dendrimère amphiphile (AmDM). Ces nanomicelles sont capables d’encapsuler efficacement la doxorubicine. Les taux de remplissage de ces nanomicelles sont extrêmement élevés. Ces nanomicelles montrent une efficacité supérieure à la doxorubicine libre et ceci sur divers types de cellules cancéreuses. De plus, ce mécanisme de pénétration cellulaire permet à ces nanomicelles de contourner le relargage du médicament médié par la pompe à efflux glycoprotéine, et ainsi surmonter la résistance à la doxorubicine. L’étude sur souris montre également un excellent effet anticancéreux associé à une diminution des effets toxiques de la doxorubicine. / Poor tumor penetration and high toxicity of anticancer drugs, together with the developed drug resistance constitute challenging hurdles for cancer therapy. The application of nanotechnology for anticancer drug delivery is expected to address these issues and bring new hope for cancer treatment. In the first part of my PhD thesis, we used a new tumor-penetrating peptide, CRGDK, to conjugate onto the surface of doxorubicin encapsulated DSPE-PEG2000 nanomicelles. The CRGDK peptide conjugated on the nanomicelles triggered specific binding to Nrp-1 receptors, leading to enhanced cellular uptake and anticancer activity in vitro. The in vivo results further confirmed that the CRGDK-decorated nanomicelles could efficiently accumulate and penetrate into deeper tumors. In the second part of my PhD thesis, we established an original nanomicellar drug delivery system based on an amphiphilic dendrimer (AmDM), which could generate supramolecular micelles to effectively encapsulate the anticancer drug doxorubicin (DOX) with high drug loading capacity (> 40%), thanks to the unique dendritic structure creating large void space for drug accommodation. The resulting AmDM/DOX nanomicelles are able to specifically accumulate at tumor sites via EPR effect and penetrate deeper into tumor tissues thanks to their small size. Most importantly, these nanomicelles exhibit significantly improved anticancer activity and reduced systemic toxicity, and are very effective even towards drug resistant cancers by virtue of their macropinocytotic cell uptake mechanism and their ability to bypass cell drug efflux pumps.