The CARE Oncology METRICS Protocol: Repurposing Generic Medicines for Metabolic Cancer Therapy

The CARE Oncology / METRICS Protocol represents a pioneering approach to cancer treatment based on the repurposing of safe, off-patent generic medications to target the metabolic vulnerabilities of cancer cells. Developed by Dr. Gilles Frija and colleagues at the Care Oncology Clinic in London, the protocol combines four well-established drugs—metformin, atorvastatin, doxycycline, and mebendazole—as an adjunct to conventional cancer therapy. The protocol and its associated real-world evidence study, METRICS, emerged from a five-year research effort to identify existing medicines whose combined mechanisms could address cancer as a metabolic disease. This essay explores the protocol's foundational principles, the rigorous drug selection process, the scientific rationale for each component, the clinical evidence gathered to date, and the broader implications of this approach for the future of oncology.

---

1. Introduction: A New Model for Evidence Generation

The CARE Oncology Clinic, located on Harley Street in London, was established with a specific mission: to evaluate whether a combination of safe, tolerable, existing generic therapies could slow cancer growth and improve survival for patients in whom standard treatments were no longer available or effective . The clinic's approach, developed by Dr. Gilles Frija and his colleagues, including Clinical Director Dr. Samir Agrawal and SEEK Group strategist Robin Bannister, represents a fundamental departure from traditional drug development .

The challenge they faced was significant. Preclinical and early clinical data increasingly supported the use of a number of non-cancer drugs against multiple tumor types, particularly metabolically targeted agents that showed promise as adjuvant chemo and radiosensitizers . However, the standard pathway of expensive randomized controlled trials was not economically viable for studying patent-expired off-label medications. Moreover, in patient populations where standard of care had already failed, randomized trials raised ethical questions about control arms .

To address this dilemma, the team developed a novel approach to generating clinical evidence. The METRICS study (Metabolic Treatment Initiative for Cancers) was designed as a participant-funded, open-label, non-randomized, real-world study to gather high-quality data on the safety, tolerability, and effectiveness of four off-label metabolically targeted medicines as an adjunctive cancer treatment . This innovative methodology, which emphasizes triangulation of evidence from multiple sources, offers a template for ethical and efficient evaluation of repurposed medications across medicine .

2. The Foundational Philosophy: Cancer as a Metabolic Disease

The CARE Oncology Protocol is built upon a conceptual framework that positions cancer fundamentally as a metabolic disorder rather than primarily a genetic one. This perspective, while increasingly supported by research, remains distinct from the dominant paradigm that has guided oncology for decades.

Cancer cells exhibit profound metabolic alterations that distinguish them from normal cells. Most notably, they display the Warburg effect: a preference for aerobic glycolysis, meaning they metabolize glucose to lactate even in the presence of adequate oxygen. This metabolic reprogramming supports the rapid proliferation of cancer cells by providing biosynthetic precursors for nucleotides, amino acids, and lipids, but it also creates unique vulnerabilities.

The protocol's architects recognized that these metabolic features are common to many cancer types, offering an opportunity for broadly applicable therapeutic strategies . Unlike targeted therapies directed at specific genetic mutations found only in subsets of patients, metabolic interventions could potentially benefit a wide range of individuals regardless of tumor origin or mutational profile. This explains why the METRICS study was designed as an "all comers" study, enrolling patients with various cancer types .

The metabolic approach also addresses a fundamental limitation of conventional oncology: the inevitability of resistance. Cancer cells, with their genomic instability and heterogeneity, eventually evolve resistance to agents that target specific molecular pathways. By attacking core metabolic processes essential for cancer cell survival, metabolic therapy may raise the barrier for resistance development.

3. The Five-Year Drug Selection Process

Identifying the optimal combination of drugs for the protocol was neither simple nor rapid. The team spent approximately five years developing and refining their selection methodology before the first patient was enrolled in late 2013 .

The process began with an enormous pool of candidate drugs. As Robin Bannister explained, there is an "enormous amount of information on an enormous amount of drugs that you could interpret as having some sort of desirable influence in cancer" . The challenge was developing a filtering system to identify the most promising candidates.

The selection framework integrated multiple data streams:

Mechanistic data examined whether drugs affected biological pathways relevant to cancer metabolism. Enzyme data assessed specific biochemical interactions. In vitro data evaluated effects on cancer cell lines. In vivo data from animal models provided evidence of anti-tumor activity. Patient data, where available, offered clues from human experience. Large-scale epidemiological studies revealed associations between drug use and cancer outcomes.

The critical question for each candidate was whether all these signals pointed in the same direction. Did mechanistic, preclinical, and epidemiological evidence consistently suggest a positive effect on cancer? Only drugs with convergent evidence across multiple domains advanced in the selection process.

Beyond efficacy signals, the team considered practical factors essential for clinical implementation. Side effect profiles had to be acceptable for long-term use in cancer patients. Dosing regimens needed to be practical and well-established. Drug interactions had to be manageable. Finally, the selected drugs had to impact different but related mechanisms in cancer, creating a synergistic combination that addressed multiple vulnerabilities simultaneously .

The goal was to inhibit the unique metabolic processes of cancer cells, particularly glycolysis and the pathways cancer cells use to absorb and utilize nutrients for growth . After five years of systematic evaluation, five drugs emerged as the optimal combination: metformin, statins, doxycycline, mebendazole, and ibuprofen . The final protocol, as implemented and registered in clinical trials, ultimately settled on four agents: metformin, atorvastatin, doxycycline, and mebendazole .

4. The Drug Combination: Mechanisms and Rationale

Each component of the CARE Oncology Protocol was selected for its ability to target distinct but complementary aspects of cancer metabolism and survival. Understanding these individual mechanisms illuminates the rationale for their combination.

Metformin

Metformin, a first-line medication for type 2 diabetes, has accumulated substantial epidemiological evidence suggesting reduced cancer incidence and improved outcomes in diabetic patients taking the drug. Its primary mechanism involves activation of AMP-activated protein kinase, a cellular energy sensor that, when activated, inhibits the mTOR pathway, a key driver of cell growth and proliferation. Metformin also reduces circulating insulin and insulin-like growth factor levels, both of which can promote cancer growth. Additionally, it inhibits mitochondrial complex I, reducing energy production in cancer cells and further activating AMPK. The protocol specifies oral metformin at 500 to 1000 mg once daily, increasing to twice daily if tolerated after two weeks .

Atorvastatin

Statins, including atorvastatin, are widely prescribed for cholesterol reduction, but their potential anticancer effects extend beyond lipid lowering. By inhibiting HMG-CoA reductase, the rate-limiting enzyme in cholesterol biosynthesis, statins deplete cancer cells of intermediates required for post-translational modification of proteins critical for growth and signaling. This includes inhibition of the mevalonate pathway, which produces isoprenoids used for prenylation of small GTPases such as Ras and Rho, proteins frequently mutated or overactive in cancer. The protocol uses oral atorvastatin up to 80 mg once daily .

Doxycycline

Doxycycline is a broad-spectrum antibiotic of the tetracycline class, but its relevance to cancer therapy stems from its effects on mitochondria. Doxycycline inhibits mitochondrial protein synthesis, selectively targeting cancer cells that rely more heavily on mitochondrial function than previously appreciated. By inhibiting translation of mitochondrial-encoded proteins, doxycycline impairs oxidative phosphorylation and increases oxidative stress, pushing cancer cells toward cell death. It also inhibits matrix metalloproteinases, enzymes involved in tissue remodeling and metastasis. The protocol uses oral doxycycline 100 mg once daily .

Mebendazole

Mebendazole is an antihelminthic medication used worldwide for parasitic worm infections, but a growing body of preclinical research supports its repurposing for cancer. Mebendazole exerts multiple anticancer effects: it inhibits tubulin polymerization, disrupting microtubule function and mitotic spindle formation, similar to established chemotherapy agents like vinca alkaloids. It also inhibits angiogenesis by reducing vascular endothelial growth factor (VEGF) signaling, induces apoptosis through multiple pathways, and may inhibit the hedgehog signaling pathway implicated in several cancers. The protocol uses oral mebendazole 100 mg once daily .

The combination of these four agents attacks cancer through multiple independent but complementary mechanisms: metabolic disruption, signaling inhibition, mitochondrial impairment, cytoskeletal interference, and angiogenesis blockade. This multi-targeted approach theoretically reduces the likelihood of resistance and increases the probability of clinical benefit.

5. The METRICS Study: Design and Methodology

The METRICS study (NCT02201381) was designed to generate real-world evidence on the safety, tolerability, and effectiveness of the four-drug regimen . The study's methodology reflects both pragmatic considerations and innovative approaches to evidence generation.

The study was conceived as a non-randomized, open-label, real-world trial comparing outcomes against matched controls . Patients were referred to the Care Oncology Clinic by their healthcare professionals and, after providing informed consent, were enrolled in the study . Eligibility required confirmed cancer diagnosis, age 18 to 85 years, and current or planned standard of care therapy or completion of standard treatment .

Exclusion criteria were designed to ensure patient safety: pregnancy or lactation, major organ failure, significant liver or kidney dysfunction, diabetic ketoacidosis, history of cardiac failure or recent myocardial infarction, gastrointestinal surgery compromising drug absorption, and hypersensitivity to any study medication .

Enrolled patients received the fixed drug regimen at specified doses and were followed quarterly with blood tests, imaging as clinically indicated, and collection of medical records . The primary outcome measure was overall survival for all cancer types and for each cancer type individually . Secondary outcomes included changes in tumor size, tumor spread, tumor number, and cancer biomarkers . Safety monitoring assessed treatment-related adverse events through quarterly laboratory tests and patient reports .

The study's statistical approach recognized the challenges of non-randomized design. Survival analysis methods estimated overall survival associated with the regimen, interpreted in light of literature data and expert opinion . Bayesian methods were considered where appropriate. The relationship between biochemical changes and efficacy outcomes was investigated graphically. Propensity score methods were planned to adjust for potential bias .

Importantly, the study was designed to include all eligible patients over a specified time frame to reduce selection bias, and the large target enrollment of at least 5000 patients was intended to provide sufficient power for meaningful subgroup analyses . The intention-to-treat philosophy meant that all enrolled patients would be included in primary analysis regardless of adherence .

6. Early Clinical Experience and Outcomes

The METRICS study began enrolling patients in December 2013, and early results, while preliminary, provided encouraging signals . The first patient enrolled, a woman for whom standard treatments were no longer appropriate, demonstrated a "massive reduction in her tumour burden" after beginning the protocol .

The team remained appropriately cautious about interpreting early responses. As Robin Bannister emphasized, "If we'd seen no response in any patient, would it mean we should stop? Absolutely not. And if you see the first patient start to respond, does it mean they all will? Absolutely not" . Nevertheless, observing tumor reduction in the first patient provided validation of the underlying rationale and encouragement to continue.

Recruitment progressed steadily, with patients referred by consultants and individuals contacting the clinic directly. The "all comers" design created a heterogeneous study population with various cancer types, stages, and life expectancies, adding complexity to analysis but also potentially increasing generalizability .

A peer-reviewed article published in 2019 presented preliminary retrospective data from the METRICS study, focusing on glioblastoma patients as a case study for the novel evidence-generation methodology . This publication demonstrated the feasibility of generating meaningful clinical data through real-world, participant-funded studies and offered a template for evaluating other off-label medications .

7. The Challenge of Evidence Generation for Repurposed Drugs

The CARE Oncology Protocol and METRICS study highlight a fundamental challenge in contemporary medicine: how to generate high-quality evidence for promising off-label uses of generic medications when the traditional randomized controlled trial model is economically unviable .

Pharmaceutical companies have little financial incentive to invest millions of dollars in clinical trials for drugs that are already off-patent and available at low cost. The potential return on investment cannot justify the expense, creating a market failure that leaves potentially valuable treatments unevaluated and unavailable to patients.

The METRICS study design offers one potential solution to this dilemma. By operating as a participant-funded, real-world study, it reduces costs while still collecting systematic data under defined protocols . The use of matched controls from literature and expert opinion, combined with advanced statistical methods like propensity score adjustment, attempts to address the limitations of non-randomized design .

The authors of the 2019 perspective article proposed that a pluralistic evidence base and triangulation of evidence from multiple sources could support clinical trial data for off-label drug use in oncology . This approach recognizes that evidence comes in many forms: mechanistic studies, preclinical experiments, epidemiological associations, and real-world clinical data, all of which can contribute to a comprehensive understanding of a treatment's benefits and risks.

For the CARE Oncology Protocol specifically, this means that evidence accumulates incrementally. Each additional patient treated, each retrospective analysis published, and each biological mechanism elucidated adds to the totality of evidence supporting or questioning the regimen.

8. Current Status and Future Directions

As of the most recent update to the clinical trial registry in July 2022, the prospective interventional phase of the METRICS study was listed as "withdrawn" . The detailed description clarifies that the study was designed to collect data prospectively and retrospectively analyze data from the cancer clinic, with the retrospective study ongoing while the prospective study had not yet started .

This status reflects the challenging landscape for conducting clinical research on repurposed medications. Without substantial funding, initiating and maintaining a large-scale prospective trial is difficult. However, the retrospective analysis of clinical data from patients treated at the CARE Oncology Clinic continues, potentially providing valuable real-world evidence on outcomes associated with the protocol .

The vision for the protocol's future, as articulated in 2014, remains compelling. If the combination demonstrates benefit, the low cost of generic medications could enable rapid uptake within national health systems . The estimated annual treatment cost of approximately £100 per patient contrasts starkly with the £100,000 or more for novel targeted therapies . This cost differential could have profound implications for healthcare systems worldwide, particularly in resource-limited settings.

Beyond immediate clinical application, the metabolic approach to cancer therapy could enable more fundamental shifts in oncology. Robin Bannister envisioned a future where individuals at high risk for certain cancers might receive preventive metabolic therapy, potentially avoiding radical interventions like prophylactic mastectomy . While speculative, such possibilities illustrate the transformative potential of the underlying paradigm.

9. Critical Considerations and Limitations

Despite its promise, the CARE Oncology Protocol must be evaluated with appropriate attention to its limitations. The evidence base, while growing, remains preliminary. The published retrospective data are limited, and no randomized controlled trials have been completed. The withdrawal of the prospective METRICS study leaves questions about the feasibility of generating definitive evidence through this model.

The non-randomized design of the available studies introduces potential biases. Patients who seek out and can afford care at a private London clinic may differ systematically from the general cancer population. Without randomized comparison, distinguishing treatment effects from prognostic factors and natural history is challenging.

The heterogeneity of the patient population, while intended to increase generalizability, complicates interpretation. Patients with different cancer types, stages, prior treatments, and concurrent therapies may respond differently to the regimen. Identifying which patients are most likely to benefit requires larger sample sizes and sophisticated subgroup analyses.

Safety considerations also merit attention. While each individual drug has an established safety profile in its approved indications, their combined use in cancer patients over extended periods raises questions about additive or synergistic toxicities. The study's safety monitoring addresses this concern, but comprehensive safety data require larger patient numbers and longer follow-up .

The metabolic paradigm underlying the protocol, while supported by substantial evidence, represents only one dimension of cancer biology. Cancers are heterogeneous, and metabolic vulnerabilities may vary between and within tumors. Some cancers may be more dependent on the targeted pathways than others, potentially explaining variable responses.

10. Conclusion

The CARE Oncology / METRICS Protocol represents a bold and scientifically grounded attempt to translate insights from cancer metabolism research into accessible, affordable treatment for patients. Developed through a five-year process of systematic drug selection, the combination of metformin, atorvastatin, doxycycline, and mebendazole targets multiple complementary vulnerabilities of cancer cells, potentially slowing growth and improving outcomes when added to standard therapy.

The protocol's innovative approach to evidence generation, using real-world data and retrospective analysis to evaluate off-label medications, offers a template for addressing the market failure that leaves many promising repurposed drugs unevaluated. By operating outside the traditional pharmaceutical development pathway, the CARE Oncology Clinic has generated preliminary data supporting the regimen's potential while treating patients who might otherwise have exhausted options.

The early clinical experiences, including dramatic responses in some patients, provide encouraging signals that validate the underlying rationale. The first patient's tumor reduction, the accumulating retrospective data, and the consistency of the mechanistic evidence all point toward genuine biological activity.

Yet important questions remain. The prospective METRICS study was withdrawn, leaving the evidence base reliant on retrospective analysis. The heterogeneity of responses across cancer types and individual patients requires further characterization. Safety in long-term use requires continued monitoring. Most fundamentally, the absence of randomized controlled data means that the magnitude of benefit attributable to the regimen, independent of patient selection and natural history, remains uncertain.

For patients considering the protocol, these limitations must be weighed against the alternatives. For those who have exhausted standard options or face poor prognoses with conventional therapy, the addition of safe, well-tolerated generic medications with plausible anticancer mechanisms may represent a reasonable choice, particularly when pursued under medical supervision with appropriate monitoring.

The broader significance of the CARE Oncology Protocol extends beyond its specific drug combination. It demonstrates the feasibility of patient-funded, real-world research. It validates the metabolic approach to cancer therapy. It highlights the untapped potential of repurposed generic medications. And it challenges the oncology community to develop new models for evaluating and accessing treatments that cannot attract pharmaceutical investment.

As Robin Bannister observed, the ultimate vision is transformative: a future where cancer care is not only more effective but also more accessible, where prevention replaces radical intervention, and where the cost of treatment does not bankrupt patients or healthcare systems . The CARE Oncology Protocol represents one step toward that vision, grounded in rigorous science, executed with pragmatic innovation, and sustained by the hope of patients and physicians alike.

11. Key Published Works and Resources

Publication: A New Method for Ethical and Efficient Evidence Generation for Off-Label Medication Use in Oncology (A Case Study in Glioblastoma), Agrawal et al., 2019

Clinical Trial Registration: METRICS Study, NCT02201381, ClinicalTrials.gov

Clinic: Care Oncology Clinic, London, United Kingdom

Media Coverage: Pharmaceutical Technology feature on the METRICS study, December 2014