New derivatives of the antimalarial drug Pyrimethamine in the control of melanoma tumor growth: an in vitro and in vivo study

Pyrimethamine reduced tumour growth in pre-clinical cancer models: a systematic review to identify potential pre-clinical studies for subsequent human clinical trials

Pyrimethamine (PYR), a STAT3 inhibitor, has been shown to reduce tumour burden in mouse cancer models. It is unclear how much of a reduction occurred or whether the PYR dosages and route of administration used in mice were consistent with the FDA's recommendations for drug repurposing. Search engines such as ScienceDirect, PubMed/MEDLINE, and other databases, including Google Scholar, were thoroughly searched, as was the reference list. The systematic review includes fourteen (14) articles. The risk of bias (RoB) was assessed using SYRCLE's guidelines. Due to the heterogeneity of the data, no meta-analysis was performed. According to the RoB assessment, 13/14 studies fall into the moderate RoB category, with one study classified as high RoB. None adhered to the ARRIVE guideline for transparent research reporting. Oral (FDA-recommended) and non-oral routes of PYR administration were used in mice, with several studies reporting very high PYR dosages that could lead to myelosuppression, while oral PYR dosages of 30 mg/kg or less are considered safe. Direct human equivalent dose translation is probably not the best strategy for comparing whether the used PYR dosages in mice are in line with FDA-approved strength because pharmacokinetic profiles, particularly PYR's half-life (t1/2), between humans (t1/2 = 96 h) and mice (t1/2 = 6 h), must also be considered. Based on the presence of appropriate control and treatment groups, as well as the presence of appropriate clinically proven chemotherapy drug(s) for comparison purposes, only one study (1/14) involving liver cancer can be directed into a clinical trial. Furthermore, oesophageal cancer too can be directed into clinical trials, where the indirect effect of PYR on the NRF2 gene may suppress oesophageal cancer in patients, but this must be done with caution because PYR is an investigational drug for oesophageal cancer, and combining it with proven chemotherapy drug(s) is recommended.

Pyrimethamine upregulates BNIP3 to interfere SNARE-mediated autophagosome-lysosomal fusion in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the most common tumor types and remains a major clinical challenge. Increasing evidence has revealed that mitophagy inhibitors can enhance the effect of chemotherapy on HCC. However, few mitophagy inhibitors have been approved for clinical use in humans. Pyrimethamine (Pyr) is used to treat infections caused by protozoan parasites. Recent studies have reported that Pyr may be beneficial in the treatment of various tumors. However, its mechanism of action is still not clearly defined. Here, we found that blocking mitophagy sensitized cells to Pyr-induced apoptosis. Mechanistically, Pyr potently induced the accumulation of autophagosomes by inhibiting autophagosome-lysosome fusion in human HCC cells. In vitro and in vivo studies revealed that Pyr blocked autophagosome-lysosome fusion by upregulating BNIP3 to inhibit synaptosomal-associated protein 29 (SNAP29)-vesicle-associated membrane protein 8 (VAMP8) interaction. Moreover, Pyr acted synergistically with sorafenib (Sora) to induce apoptosis and inhibit HCC proliferation in vitro and in vivo. Pyr enhances the sensitivity of HCC cells to Sora, a common chemotherapeutic, by inhibiting mitophagy. Thus, these results provide new insights into the mechanism of action of Pyr and imply that Pyr could potentially be further developed as a novel mitophagy inhibitor. Notably, Pyr and Sora combination therapy could be a promising treatment for malignant HCC.

Investigating the anti-cancer potential of pyrimethamine analogues through a modern chemical biology lens

Pharmacological inhibition of dihydrofolate reductase (DHFR) is an established approach for treating a variety of human diseases, including foreign infections and cancer. However, treatment with classic DHFR inhibitors, such as methotrexate (MTX), are associated with negative side-effects and resistance mechanisms that have prompted the search for alternatives. The DHFR inhibitor pyrimethamine (Pyr) has compelling anti-cancer activity in in vivo models, but lacks potency compared to MTX, thereby requiring higher concentrations to induce therapeutic responses. The purpose of this work was to investigate structural analogues of Pyr to improve its in vitro and cellular activity. A series of 36 Pyr analogues were synthesized and tested in a sequence of in vitro and cell-based assays to monitor their DHFR inhibitory activity, cellular target engagement, and impact on breast cancer cell viability. Ten top compounds were identified, two of which stood out as potential lead candidates, 32 and 34. These functionalized Pyr analogues potently engaged DHFR in cells, at concentrations as low as 1 nM and represent promising DHFR inhibitors that could be further explored as potential anti-cancer agents.

Small molecule screen identifies pyrimethamine as an inhibitor of NRF2-driven esophageal hyperplasia

OBJECTIVE

NRF2 is a master transcription factor that regulates the stress response. NRF2 is frequently mutated and activated in human esophageal squamous cell carcinoma (ESCC), which drives resistance to chemotherapy and radiation therapy. Therefore, a great need exists for NRF2 inhibitors for targeted therapy of NRF2high ESCC.

DESIGN

We performed high-throughput screening of two compound libraries from which hit compounds were further validated in human ESCC cells and a genetically modified mouse model. The mechanism of action of one compound was explored by biochemical assays.

RESULTS

Using high-throughput screening of two small molecule compound libraries, we identified 11 hit compounds as potential NRF2 inhibitors with minimal cytotoxicity at specified concentrations. We then validated two of these compounds, pyrimethamine and mitoxantrone, by demonstrating their dose- and time-dependent inhibitory effects on the expression of NRF2 and its target genes in two NRF2Mut human ESCC cells (KYSE70 and KYSE180). RNAseq and qPCR confirmed the suppression of global NRF2 signaling by these two compounds. Mechanistically, pyrimethamine reduced NRF2 half-life by promoting NRF2 ubiquitination and degradation in KYSE70 and KYSE180 cells. Expression of an Nrf2E79Q allele in mouse esophageal epithelium (Sox2CreER;LSL-Nrf2E79Q/+) resulted in an NRF2high phenotype, which included squamous hyperplasia, hyperkeratinization, and hyperactive glycolysis. Treatment with pyrimethamine (30 mg/kg/day, p.o.) suppressed the NRF2high esophageal phenotype with no observed toxicity.

CONCLUSION

We have identified and validated pyrimethamine as an NRF2 inhibitor that may be rapidly tested in the clinic for NRF2high ESCC.

Pyrimethamine inhibits cell growth by inducing cell senescence and boosting CD8+ T-cell mediated cytotoxicity in colorectal cancer

BACKGROUND

The emergence of nonresponse or resistance to traditional chemotherapeutic agents is one of the main challenges of colorectal cancer (CRC) therapies. Thus, novel therapeutic drugs that can improve the clinical outcomes of CRC patients are urgently needed. The purpose of this study was to investigate the effects and mechanisms of pyrimethamine in CRC.

METHODS AND RESULTS

In this study, we assessed the role of pyrimethamine on CRC cell growth by cell counting kit-8 and colony formation assays. Cell cycle distribution and cellular senescence were determined by flow cytometry and senescence-associated β-galactosidase staining respectively. RNA-seq analysis and western blotting were used to investigate the potential pathways of pyrimethamine in CRC development. Moreover, animal experiments were performed to evaluate the effect of pyrimethamine in vivo. Our results demonstrated that pyrimethamine could inhibit cell growth by inducing S phase arrest followed by cellular senescence in CRC cells, and the p38MAPK-p53 axis was probably involved in that effect. In addition, pyrimethamine could also boost CD8⁺ T-cell mediated cytotoxicity and exert antitumor activity in vivo.

CONCLUSION

These results indicated that pyrimethamine may be a promising candidate agent for CRC treatment.

The multifaceted antineoplastic role of pyrimethamine against different human malignancies

Cancer accounted for nearly 10 million deaths in 2020 and is the second leading cause of death worldwide. The chemotherapeutic agents that are in clinical practice possess a broad range of severe adverse effects towards vital organs which emphasizes the importance of the discovery of new therapeutic agents or repurposing of existing drugs for the treatment of human cancers. Pyrimethamine is an antiparasitic drug used for the treatment of malaria and toxoplasmosis with a well-documented excellent safety profile. In the last five years, numerous efforts have been made to explore the anticancer potential of pyrimethamine in in vitro and in vivo preclinical models and to repurpose it as an anticancer agent. The studies have demonstrated that pyrimethamine inhibits oncogenic proteins such as STAT3, NF-κB, DX2, MAPK, DHFR, thymidine phosphorylase, telomerase, and many more in a different types of cancer models. Moreover, pyrimethamine has been reported to work in synergy with other anticancer agents, such as temozolomide, to induce apoptosis of tumor cells. Recently, the results of phase-1/2 clinical trials demonstrated that pyrimethamine administration reduces the expression of STAT3 signature genes in tumor tissues of chronic lymphocytic leukemia patients with a good therapeutic response. In the present article, we have reviewed most of the published papers related to the antitumor effects of pyrimethamine in malignancies of breast, liver, lung, skin, ovary, prostate, pituitary, and leukemia in in vitro and in vivo settings. We have also discussed the pharmacokinetic profile and results of clinical trials obtained after pyrimethamine treatment. From these studies, we believe that pyrimethamine has the potential to be repurposed as an anticancer drug. This article is protected by copyright. All rights reserved.

Conducting polymer nanoparticles for a voltage-controlled release of pharmacological chaperones

Pharmacological chaperones (PCs) are low-molecular weight chemical molecules used in patients for the treatment of some rare diseases caused primarily by protein instability. A controlled and on-demand release of PCs via nanoparticles is an alternative for cases in which long treatments are needed and prolonged oral administration could have adverse effects. In this work, pyrimethamine (PYR), which is a potent PC consisting of pyrimidine-2,4-diamine substituted at position 5 by a p-chlorophenyl group and at position 6 by an ethyl group, was successfully loaded in electroresponsive poly(3,4-ethylenedioxythiophene) nanoparticles (PEDOT NPs). The PYR-loading capacity was 11.4 ± 1.5%, with both loaded and unloaded PEDOT NPs exhibiting similar sizes (215 ± 3 and 203 ± 1 nm, respectively) and net surface charges (-26 ± 7 and -29 ± 6 mV, respectively). In the absence of electrical stimulus, the release of PC from loaded NPs is very low (1.6% in 24 h and 18% in 80 days) in aqueous environments. Instead, electrical stimuli that sustained for 30 min enhanced the release of PYR, which was ∼50% when the voltage was scanned from -0.5 V to 0.5 V (cyclic voltammetry) and ∼35% when a constant voltage of 1.0 V was applied (chronoamperometry).

Reimagining dots and dashes: Visualizing structure and function of organelles for high-content imaging analysis

Organelles are responsible for biochemical and cellular processes that sustain life and their dysfunction causes diseases from cancer to neurodegeneration. While researchers are continuing to appreciate new roles of organelles in disease, the rapid development of specifically targeted fluorescent probes that report on the structure and function of organelles will be critical to accelerate drug discovery. Here, we highlight four organelles that collectively exemplify the progression of phenotypic discovery, starting with mitochondria, where many functional probes have been described, then continuing with lysosomes and Golgi and concluding with nascently described membraneless organelles. We introduce emerging probe designs to explore organelle-specific morphology and dynamics and highlight recent case studies using high-content analysis to stimulate further development of probes and approaches for organellar high-throughput screening.

Disentangling the therapeutic tactics in GBM: From bench to bedside and beyond

Glioblastoma multiforme (GBM) is one of the most common and malignant form of adult brain tumor with a high mortality rate and dismal prognosis. The present standard treatment comprising surgical resection followed by radiation and chemotherapy using temozolomide can broaden patient's survival to some extent. However, the advantages are not palliative due to the development of resistance to the drug and tumor recurrence following the multimodal treatment approaches due to both intra- and intertumoral heterogeneity of GBM. One of the major contributors to temozolomide resistance is O⁶ -methylguanine-DNA methyltransferase. Furthermore, deficiency of mismatch repair, base excision repair, and cytoprotective autophagy adds to temozolomide obstruction. Rising proof additionally showed that a small population of cells displaying certain stem cell markers, known as glioma stem cells, adds on to the resistance and tumor progression. Collectively, these findings necessitate the discovery of novel therapeutic avenues for treating glioblastoma. As of late, after understanding the pathophysiology and biology of GBM, some novel therapeutic discoveries, such as drug repurposing, targeted molecules, immunotherapies, antimitotic therapies, and microRNAs, have been developed as new potential treatments for glioblastoma. To help illustrate, "what are the mechanisms of resistance to temozolomide" and "what kind of alternative therapeutics can be suggested" with this fatal disease, a detailed history of these has been discussed in this review article, all with a hope to develop an effective treatment strategy for GBM.

Pyrimethamine exerts significant antitumor effects on human ovarian cancer cells both in vitro and in vivo

Pyrimethamine has been used principally to treat infections from protozoan parasites. Although previous studies have shown that pyrimethamine exhibited anticancer activity by inducing cellular apoptosis, there are none that show that pyrimethamine possesses anticancer activity with respect to ovarian cancer. We examined the roles of pyrimethamine on apoptosis and proliferation, DNA damage, and cell cycle distribution of human ovarian cancer cell lines in vitro. To investigate the antitumor efficacy of pyrimethamine in vivo, we established two intraperitoneal ovarian carcinoma models in nude mice. Pyrimethamine significantly induced apoptosis of ovarian cancer cells via growth inhibition, cell cycle arrest, and nuclear DNA damage in vitro and manifested antitumor activity by inhibiting tumor growth, thereby prolonging the survival time of tumor-bearing mice. We also demonstrated that pyrimethamine increased the expression of caspase-9 and decreased the expression of X-linked inhibitor of apoptosis protein. In conclusion, the antitumor effects of pyrimethamine were associated with enhanced apoptosis of tumor cells and inhibition of the growth of intratumoral microvessels. Our results indicate that pyrimethamine may provide an effective approach toward inhibiting the growth of ovarian cancer with minimal adverse effects.

Pyrimethamine Elicits Antitumor Effects on Prostate Cancer by Inhibiting the p38-NF-κB Pathway

Since incurable castration-resistant prostate cancer (CRPC) inevitably develops following treatment with androgen deprivation therapy, there is an urgent need to devise new therapeutic strategies to treat this cancer. Pyrimethamine, an FDA-approved antimalarial drug, is known to exert an antitumor activity in various types of human cancer cells. However, whether pyrimethamine can inhibit prostate cancer is not well established. Hence, the present study aimed to characterize the mechanism of action of pyrimethamine on prostate cancer. We investigated the potential effect of pyrimethamine on cell proliferation, cell cycle, and apoptosis in metastatic DU145 and PC3 prostate cancer cells. We found that pyrimethamine inhibited cell proliferation, induced cell cycle arrest in the S phase, and promoted cell apoptosis of prostate cells in vitro; it also suppressed tumor growth in xenograft models. In addition, we observed that pyrimethamine suppressed prostate cancer growth by inhibiting the p38-NF-κB axis in vitro and in vivo. Thus, this study demonstrates that pyrimethamine is a novel p38 inhibitor that can exert antiproliferative and proapoptotic effects in prostate cancer by affecting cell cycle and intrinsic apoptotic signaling, thereby providing a novel strategy for using pyrimethamine in CRPC treatment.

Drug rechanneling: A novel paradigm for cancer treatment

Cancer continues to be one of the leading contributors towards global disease burden. According to NIH, cancer incidence rate per year will increase to 23.6 million by 2030. Even though cancer continues to be a major proportion of the disease burden worldwide, it has the lowest clinical trial success rate amongst other diseases. Hence, there is an unmet need for novel, affordable and effective anti-neoplastic medications. As a result, a growing interest has sparkled amongst researchers towards drug repurposing. Drug repurposing follows the principle of polypharmacology, which states, "any drug with multiple targets or off targets can present several modes of action". Drug repurposing also known as drug rechanneling, or drug repositioning is an economic and reliable approach that identifies new disease treatment of already approved drugs. Repurposing guarantees expedited access of drugs to the patients as these drugs are already FDA approved and their safety and toxicity profile is completely established. Epidemiological studies have identified the decreased occurrence of oncological or non-oncological conditions in patients undergoing treatment with FDA approved drugs. Data from multiple experimental studies and clinical observations have depicted that several non-neoplastic drugs have potential anticancer activity. In this review, we have summarized the potential anti-cancer effects of anti-psychotic, anti-malarial, anti-viral and anti-emetic drugs with a brief overview on their mechanism and pathways in different cancer types. This review highlights promising evidences for the repurposing of drugs in oncology.

Autophagy dysfunctions associated with cancer cells and their therapeutic implications

Genomic analysis of human cancers indicates that the loss or mutation of core autophagy related genes, (ATG) is uncommon, whereas oncogenic events that activate autophagy and lysosomal biogenesis have been identified. Several studies have demonstrated that autophagy plays a wide variety of physiological and pathophysiological roles in cells: a cellular process that maintains the homeostasis of the normal cell, while self-defects can lead to a lawsuit to accelerate tumorigenesis and developing diseases, such as cancer. Depending on different contexts, autophagy dysfunctions may play a role: neutral, tumor-suppressive, or tumor-promoting. The process of autophagy may function in tumor suppression by mitigating metabolic stress and, in concert with apoptosis, by preventing tumor cell death by necrosis. In this case, optimal combination of autophagy inhibition (CQ, HCQ) with other conventional therapies - chemo or radiotherapy in a variety of tumor types in different phases can be successful approaches for improve the effect of anticancer therapies. This review examines recent insights of the molecular mechanism of autophagy and the potential roles of autophagy in cell death, cancer development, overview of the most recent therapeutic strategies involving autophagy modulators in cancer prevention and therapeutic opportunities.

Programmed Cell Death, from a Cancer Perspective: An Overview

Programmed cell death (PCD) is probably the most widely discussed subject among the topics of cancer therapy. Over the last 2 decades an astonishing boost in our perception of cell death has been seen, and its role in cancer and cancer therapy has been thoroughly investigated. A number of discoveries have clarified the molecular mechanism of PCD, thus expounding the link between PCD and therapeutic tools. Even though PCD is assumed to play a major role in anticancer therapy, the clinical relevance of its induction remains uncertain. Since PCD involves multiple death programs including programmed necrosis and autophagic cell death, it has contributed to our better understanding of cancer pathogenesis and therapeutics. In this review, we discuss a brief outline of PCD types as well as their role in cancer therapeutics. Since irregularities in the cell death process are frequently found in various cancers, key proteins governing cell death type could be used as therapeutic targets for a wide range of cancer.

Drug repurposing for the treatment of glioblastoma multiforme

Background

Glioblastoma Multiforme is the deadliest type of brain tumor and is characterized by very poor prognosis with a limited overall survival. Current optimal therapeutic approach has essentially remained unchanged for more than a decade, consisting in maximal surgical resection followed by radiotherapy plus temozolomide.

Main body

Such a dismal patient outcome represents a compelling need for innovative and effective therapeutic approaches. Given the development of new drugs is a process presently characterized by an immense increase in costs and development time, drug repositioning, finding new uses for existing approved drugs or drug repurposing, re-use of old drugs when novel molecular findings make them attractive again, are gaining significance in clinical pharmacology, since it allows faster and less expensive delivery of potentially useful drugs from the bench to the bedside. This is quite evident in glioblastoma, where a number of old drugs is now considered for clinical use, often in association with the first-line therapeutic intervention. Interestingly, most of these medications are, or have been, widely employed for decades in non-neoplastic pathologies without relevant side effects. Now, the refinement of their molecular mechanism(s) of action through up-to-date technologies is paving the way for their use in the therapeutic approach of glioblastoma as well as other cancer types.

Short conclusion

The spiraling costs of new antineoplastic drugs and the long time required for them to reach the market demands a profoundly different approach to keep lifesaving therapies affordable for cancer patients. In this context, repurposing can represent a relatively inexpensive, safe and fast approach to glioblastoma treatment. To this end, pros and cons must be accurately considered.

Therapeutic strategies of drug repositioning targeting autophagy to induce cancer cell death: from pathophysiology to treatment

The 2016 Nobel Prize in Physiology or Medicine was awarded to the researcher that discovered autophagy, which is an evolutionally conserved catabolic process which degrades cytoplasmic constituents and organelles in the lysosome. Autophagy plays a crucial role in both normal tissue homeostasis and tumor development and is necessary for cancer cells to adapt efficiently to an unfavorable tumor microenvironment characterized by hypo-nutrient conditions. This protein degradation process leads to amino acid recycling, which provides sufficient amino acid substrates for cellular survival and proliferation. Autophagy is constitutively activated in cancer cells due to the deregulation of PI3K/Akt/mTOR signaling pathway, which enables them to adapt to hypo-nutrient microenvironment and exhibit the robust proliferation at the pre-metastatic niche. That is why just the activation of autophagy with mTOR inhibitor often fails in vain. In contrast, disturbance of autophagy–lysosome flux leads to endoplasmic reticulum (ER) stress and an unfolded protein response (UPR), which finally leads to increased apoptotic cell death in the tumor tissue. Accumulating evidence suggests that autophagy has a close relationship with programmed cell death, while uncontrolled autophagy itself often induces autophagic cell death in tumor cells. Autophagic cell death was originally defined as cell death accompanied by large-scale autophagic vacuolization of the cytoplasm. However, autophagy is a “double-edged sword” for cancer cells as it can either promote or suppress the survival and proliferation in the tumor microenvironment. Furthermore, several studies of drug re-positioning suggest that “conventional” agents used to treat diseases other than cancer can have antitumor therapeutic effects by activating/suppressing autophagy. Because of ever increasing failure rates and high cost associated with anticancer drug development, this therapeutic development strategy has attracted increasing attention because the safety profiles of these medicines are well known. Antimalarial agents such as artemisinin and disease-modifying antirheumatic drug (DMARD) are the typical examples of drug re-positioning which affect the autophagy regulation for the therapeutic use. This review article focuses on recent advances in some of the novel therapeutic strategies that target autophagy with a view to treating/preventing malignant neoplasms.

Systematic assessment of multi-gene predictors of pan-cancer cell line sensitivity to drugs exploiting gene expression data

Background: Selected gene mutations are routinely used to guide the selection of cancer drugs for a given patient tumour. Large pharmacogenomic data sets, such as those by Genomics of Drug Sensitivity in Cancer (GDSC) consortium, were introduced to discover more of these single-gene markers of drug sensitivity. Very recently, machine learning regression has been used to investigate how well cancer cell line sensitivity to drugs is predicted depending on the type of molecular profile. The latter has revealed that gene expression data is the most predictive profile in the pan-cancer setting. However, no study to date has exploited GDSC data to systematically compare the performance of machine learning models based on multi-gene expression data against that of widely-used single-gene markers based on genomics data. Methods: Here we present this systematic comparison using Random Forest (RF) classifiers exploiting the expression levels of 13,321 genes and an average of 501 tested cell lines per drug. To account for time-dependent batch effects in IC ₅₀ measurements, we employ independent test sets generated with more recent GDSC data than that used to train the predictors and show that this is a more realistic validation than standard k-fold cross-validation. Results and Discussion: Across 127 GDSC drugs, our results show that the single-gene markers unveiled by the MANOVA analysis tend to achieve higher precision than these RF-based multi-gene models, at the cost of generally having a poor recall (i.e. correctly detecting only a small part of the cell lines sensitive to the drug). Regarding overall classification performance, about two thirds of the drugs are better predicted by the multi-gene RF classifiers. Among the drugs with the most predictive of these models, we found pyrimethamine, sunitinib and 17-AAG. Conclusions: Thanks to this unbiased validation, we now know that this type of models can predict in vitro tumour response to some of these drugs. These models can thus be further investigated on in vivo tumour models. R code to facilitate the construction of alternative machine learning models and their validation in the presented benchmark is available at http://ballester.marseille.inserm.fr/gdsc.transcriptomicDatav2.tar.gz.

Systematic assessment of multi-gene predictors of pan-cancer cell line sensitivity to drugs exploiting gene expression data

Background: Selected gene mutations are routinely used to guide the selection of cancer drugs for a given patient tumour. Large pharmacogenomic data sets, such as those by Genomics of Drug Sensitivity in Cancer (GDSC) consortium, were introduced to discover more of these single-gene markers of drug sensitivity. Very recently, machine learning regression has been used to investigate how well cancer cell line sensitivity to drugs is predicted depending on the type of molecular profile. The latter has revealed that gene expression data is the most predictive profile in the pan-cancer setting. However, no study to date has exploited GDSC data to systematically compare the performance of machine learning models based on multi-gene expression data against that of widely-used single-gene markers based on genomics data. Methods: Here we present this systematic comparison using Random Forest (RF) classifiers exploiting the expression levels of 13,321 genes and an average of 501 tested cell lines per drug. To account for time-dependent batch effects in IC ₅₀ measurements, we employ independent test sets generated with more recent GDSC data than that used to train the predictors and show that this is a more realistic validation than standard k-fold cross-validation. Results and Discussion: Across 127 GDSC drugs, our results show that the single-gene markers unveiled by the MANOVA analysis tend to achieve higher precision than these RF-based multi-gene models, at the cost of generally having a poor recall (i.e. correctly detecting only a small part of the cell lines sensitive to the drug). Regarding overall classification performance, about two thirds of the drugs are better predicted by the multi-gene RF classifiers. Among the drugs with the most predictive of these models, we found pyrimethamine, sunitinib and 17-AAG. Conclusions: Thanks to this unbiased validation, we now know that this type of models can predict in vitro tumour response to some of these drugs. These models can thus be further investigated on in vivo tumour models. R code to facilitate the construction of alternative machine learning models and their validation in the presented benchmark is available at http://ballester.marseille.inserm.fr/gdsc.transcriptomicDatav2.tar.gz.