Metformin Associated With Improved Outcomes in Diabetic Patients With Laryngeal and Oropharyngeal Carcinoma

Dear Editor,

The 5-year survival rate of patients with squamous cell carcinoma (SCC) of the head and neck continues to be under 50%.¹ Reducing metastasis and recurrence of these cancers increases prognosis greatly so discovering therapeutic options to accomplish these aims is paramount to future treatment. The antidiabetic drug, metformin, modulates molecular pathways that are known to contribute to cancer progression in the setumors. Specifically, metformin activates 5^’ adenosine monophosphate activated kinase and increases insulin sensitivity in type 2 diabetics.^2,3 While the previously proposed benefit of these mechanisms is to lower blood glucose, these mechanisms also may produce a therapeutic effect on SCC of the larynx or oropharynx. Regarding metformin’s current use in cancer treatment of other organ systems, the literature shows a consensus in the significant benefit of metformin as a neoadjuvant to doxorubicin in the treatment of breast cancer,⁴ and there is evidence to indicate that metformin therapy also improves outcomes in patients diagnosed with prostate, colonandlungcancer.^5,6 Our aim was to investigate the potential therapeutic benefit of metformin therapy in patients with SCC of the larynx and oropharynx.

To investigate this further, patients at the VA in Memphis with SCC of the larynx or oropharynx and pre-existing type 2 diabetes who remained on metformin during their cancer treatment were compared to controls never on metformin in a retrospective chart review. To have been included in the metformin group, type 2 diabetic patients had to have been on metformin for at least 1 year before their pathologic diagnosis of SCC and at least 1 year after, or they must have been on metformin at the time of death if they died within 1 year of being diagnosed with cancer. To be included in the nonmetformin group, type 2 diabetic patients must never have been on metformin. The 2 groups were evaluated for confounding factors including age, race, hemoglobin A1c, primary cancer treatment and tumor stage. End points of the study were metastases, cancer recurrence, 5-year survival rate and the presence of new malignancies.

Of 310 patients in the VA tumor registry with SCC of the larynx or oropharynx, 54 were found to also carry the diagnosis of type 2 diabetes. Of these 54 patients, 34 met our inclusion criteria. The nonmetformin group (n = 23) and the metformin group (n = 11) were found to be similar for baseline parameters such as age, primary cancer treatment, race, glucose control, age at death and tumor stage of SCC.

As shown in Table 1, there were significantly fewer instances of metastasis, as only 18.1% of the metformin group had metastasis compared to 82.6% in the nonmetformin group (P < 0.001). In addition, 72.7% of the metformin group reached 5-year survival postpathology diagnosis while only 34.7% of the nonmetformin group did (P = 0.038). There were fewer recurrences of cancer and fewer new malignancies in the metformin group, but these findings were not statistically significant.

TABLE 1.

End points for assessment of squamous cell carcinoma of larynx and oropharynx in type 2 diabetic patients.

	Metformin group (%) n = 11	Nonmetformin group (%) n = 23	P Value
Metastasis	18.1	82.6	< 0.001
Recurrence	45.5	47.8	0.897
New Malignancies	9.1	30.4	0.170
5-year survival	72.7	34.7	0.038

The strength of our results was limited by a smaller sample size than anticipated. Unfortunately, we found that many potential study patients failed to pass our study inclusion criteria because they were taken off metformin before or during their cancer treatment. From our observation, these cancer patients generally had eating impairment due to tumor mass effect or primary cancer treatment, which then led to metformin no longer being needed for diabetic control. Still, in the patients who did meet the inclusion criteria, we found that the metformin group exhibited a statistically significant reduction in tumor metastasis for SCC of the larynx or oropharynx as well as an increase in 5-year survival rate. Thus, although the strength of our conclusions is limited by small sample size, it appears there is an association with metformin therapy and reduction in metastasis and increased 5-year survival that warrants further investigation in a larger retrospective analysis or clinical trial.

There are several mechanisms that could potentially explain this association. For example, mammalian target of rapamycin activation has been shown to be a common molecular event leading to lymph node metastasis in head and neck SCC in mice.⁷ Metformin’s ability to inhibit mammalian target of rapamycin, either directly through 5′ adenosine monophosphate activated kinase activation or indirectly by lowering levels of circulating insulin, is a potential antineoplastic mechanism that could account for the lower frequency of metastases that we observed in type 2 diabetic patients on metformin with SCC of the larynx or oropharynx.² In addition, metformin’s ability to lower circulating insulin levels in type 2 diabetics may independently reduce tumorigenesis as high insulin has been proposed to contribute to neoplastic signaling.⁸ Metformin has also shown direct effects on oncogenes, such as in the inhibition of expression of human epidermal growth factor receptor 2 and tyrosine kinase in breast cancer.⁹

Clinical trials testing the efficacy and effects of metformin in squamous cell cancer of the larynx and oropharynx are limited. There are a few trials underway, and some were terminated due to low number of participants, but clinical trials would provide more definitive conclusions to the findings of our study.