MétaCan
Menu
Back to cohort
Record W3150901916 · doi:10.1093/aje/kwu311

Re: "Reduced Risk of Lung Cancer With Metformin Therapy in Diabetic Patients: A Systematic Review and Meta-Analysis"

2014· review· en· W3150901916 on OpenAlexaff
Niklas Schmedt, Laurent Azoulay, Sabrina Hense

Bibliographic record

VenueAmerican Journal of Epidemiology · 2014
Typereview
Languageen
FieldBiochemistry, Genetics and Molecular Biology
TopicMetabolism, Diabetes, and Cancer
Canadian institutionsMcGill UniversityJewish General Hospital
Fundersnot available
KeywordsMedicineMetforminMeta-analysisLung cancerOncologyInternal medicineDiabetes mellitusEndocrinology

Abstract

fetched live from OpenAlex

We read with interest the recently published meta-analysis by Zhang et al. (1), in which the authors reported a 29% reduced risk of lung cancer and a 15% reduced risk of respiratory cancer in diabetic patients treated with metformin. The authors concluded that metformin therapy appears to be associated with a lower risk of lung cancer and that this association is more prominent than associations for other cancers of the respiratory system. As the authors briefly mentioned, time-related biases were present in some of the observational studies included in their meta-analysis (1). Such conclusion-altering biases have been shown to greatly exaggerate the benefit of metformin in cancer incidence (2). As such, we are concerned that Zhang et al. failed to differentiate such studies in their meta-analysis. We believe that time-related biases affected 3 of the 6 studies included (3–8). In their case-control study, Mazzone et al. (3) reported that metformin was associated with a 52% reduction in risk of lung cancer. However, their methods suggest that there was differential assessment of metformin exposure between cases and controls. Specifically, for the lung cancer cases, exposure to metformin was correctly assessed prior to the date of diagnosis, whereas for controls, exposure was assessed at any time, which included periods before and after the time of diagnosis of the matched case. As a result, controls had a substantially higher probability of being exposed in comparison with cases, which led to a time-window bias (2). In their cohort study, Hsieh et al. (4) found a 30% increased risk of lung cancer for type 2 diabetes patients being treated with sulfonylurea as compared with metformin monotherapy. However, aside from a very opaque description of their methods, Hsieh et al. did not account for several important time-related factors. To be eligible for analysis, patients had to receive continuous drug coverage for at least 1 year at any time during follow-up, and patients with a diagnosis of cancer before initiation of antidiabetic therapy were excluded. First, since patients in any stage of type 2 diabetes at baseline were included, it is likely that users of sulfonylureas as second-line therapy were at a later stage of disease than those on metformin. This might have resulted in confounding by disease duration (2), since patients in advanced stages of type 2 diabetes may be at higher risk for cancer. In addition, patients who were on sulfonylurea monotherapy might have been on metformin before cohort entry. However, these patients were selectively excluded if they had a cancer diagnosis, and the impact of prior metformin therapy on cancer incidence in those on sulfonylurea monotherapy was disregarded, because no latency time window between start of therapy and the cancer event was considered. We assume that the benefit of metformin observed in this study was triggered by these biases due to time lag and latency. The same applies to the study by Libby et al. (5), which has been discussed in detail elsewhere (2).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.008
metaresearch head score (Gemma)0.030
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Not applicable · Consensus signal: none
GenreCandidate signal: Commentary · Consensus signal: none
Teacher disagreement score0.011
Threshold uncertainty score0.043

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0080.030
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0070.009
Bibliometrics0.0050.005
Science and technology studies0.0000.001
Scholarly communication0.0020.002
Open science0.0020.002
Research integrity0.0020.002
Insufficient payload (model declined to judge)0.0110.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.040
GPT teacher head0.369
Teacher spread0.329 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designNot applicable
Domainnot available
GenreCommentary

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations6
Published2014
Admission routes1
Has abstractno

Explore more

Same venueAmerican Journal of EpidemiologySame topicMetabolism, Diabetes, and CancerFrench-language works237,207