Tresiba Cancer Risk Signal Review: What the Evidence Actually Shows

Why the Cancer Question Arose for Insulin Analogs

The cancer risk conversation for insulin analogs began with a series of observational database studies published simultaneously in Diabetologia in June 2009. Those analyses suggested that insulin glargine (Lantus) users faced elevated breast and colorectal cancer incidence relative to human insulin users. The mechanism proposed was straightforward: insulin analogs with higher IGF-1 receptor (IGF-1R) affinity or prolonged receptor occupancy could theoretically drive mitogenic signaling through the RAS/MAPK and PI3K/Akt pathways, which promote cell proliferation and inhibit apoptosis.

That episode placed the entire analog class under scrutiny. Degludec, approved six years later, entered a regulatory environment already sensitized to this question.

The Biological Mechanism Hypothesis

Insulin and IGF-1 share structural homology. Both ligands can activate the insulin receptor (IR) and IGF-1R, though with different affinities. Prolonged IGF-1R occupancy, rather than peak binding per se, is considered the more oncologically relevant parameter because it sustains downstream mitogenic signaling without triggering the normal receptor internalization and downregulation that limits acute IR activation.

Investigators at Novo Nordisk characterized degludec's receptor pharmacology before regulatory submission. Degludec showed IGF-1R binding affinity of roughly 0.08% relative to human insulin, which itself binds IGF-1R at approximately 1% the affinity of IGF-1 itself [1]. This places degludec at the low end of the analog spectrum. For comparison, insulin glargine's active metabolite M1 binds IGF-1R at approximately 0.07 to 0.11% of human insulin in most assay systems, making the two compounds pharmacologically similar on this parameter [2].

Mitogenicity Assays and What They Reveal

Cell-based mitogenicity assays measure DNA synthesis or proliferation in IGF-1R-expressing cancer lines (typically MCF-7 breast cancer cells) after insulin analog exposure. Degludec's mitogenic potency in these assays has consistently been reported as less than or approximately equal to human insulin, and below that of insulin glargine in studies using the parent compound rather than its metabolites [1].

The limitation of these assays is their artificial receptor-expression environment. MCF-7 cells overexpress IGF-1R at levels far above those seen in most human tissues. Translating in-vitro signals to clinical risk requires far larger datasets than any cell-culture experiment can provide.

Degludec's Pharmacokinetic Profile and Oncological Relevance

Degludec forms soluble multi-hexamer chains at the subcutaneous injection site, creating a depot that releases insulin monomers slowly into the circulation. Peak-to-trough plasma concentration ratio approaches 1:1 over 24 hours at steady state, meaning receptor occupancy is unusually flat [3].

Does a Flat Concentration Profile Affect Mitogenic Risk?

This is where degludec's pharmacokinetics become relevant to the cancer question. One hypothesis from the glargine debate was that sustained, non-pulsatile insulin receptor signaling might behave more like growth factor signaling than the physiological post-prandial spikes of endogenous insulin. Degludec's ultra-flat profile could theoretically amplify this concern, or alternatively could reduce it by keeping IGF-1R occupancy consistently below the threshold needed to activate mitogenic cascades.

Published receptor-kinetics modeling by Kurtzhals et al. Argues that receptor dissociation rate is the critical parameter. Degludec dissociates from both the IR and IGF-1R at rates comparable to human insulin [1], which means despite the flat plasma curve, individual receptor-binding events are brief and physiologically normal in duration. This is mechanistically distinct from super-affinity analogs such as insulin X10 (AspB10), which was withdrawn from development precisely because of prolonged receptor occupancy and in-vivo mammary tumor formation in rats.

Comparing Degludec to Glargine on Receptor Kinetics

Both degludec and glargine have accumulated substantial pharmacological profiling. The table below summarizes key receptor parameters where published data exist.

| Parameter | Human Insulin | Degludec | Glargine (M1 metabolite) | |---|---|---|---| | IGF-1R binding (% vs. IGF-1) | ~1% | ~0.08% | ~0.07 to 0.11% | | IR binding affinity (relative) | 100% | ~85 to 90% | ~55 to 70% | | Receptor dissociation rate | Normal | Normal | Normal | | In-vitro mitogenicity (MCF-7) | Reference | <1x reference | 0.9 to 1.3x reference |

Data compiled from Kurtzhals et al. [1] and FDA pharmacology review documents [4].

DEVOTE: The Key Cardiovascular Safety Trial

The DEVOTE trial (Trial Comparing Cardiovascular Safety of Insulin Degludec versus Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events) randomized 7,637 patients with type 2 diabetes across 20 countries to either degludec U-100 or glargine U-100, with a primary endpoint of three-point MACE (cardiovascular death, nonfatal myocardial infarction, nonfatal stroke) [5].

Primary Cardiovascular Results

Degludec met its non-inferiority margin for MACE (HR 0.91, 95% CI 0.78 to 1.06; P<0.001 for non-inferiority), and produced 53% fewer confirmed nocturnal hypoglycemic events than glargine (rate ratio 0.47, 95% CI 0.38 to 0.58; P<0.001) [5]. These are the headline numbers that drove widespread clinical adoption.

Malignancy Data Within DEVOTE

DEVOTE was not powered to detect cancer differences. Median follow-up was 2.0 years, and cancer endpoint collection was a secondary safety activity rather than a prespecified primary or secondary endpoint. With that caveat stated clearly, the observed cancer event rates were:

• Degludec arm: 5.3% of patients experienced a neoplasm-related adverse event
• Glargine arm: 5.2% of patients experienced a neoplasm-related adverse event
• Hazard ratio: 1.00 (95% CI 0.83 to 1.22)

The 95% confidence interval spans a range from a 17% reduction to a 22% increase, which means DEVOTE cannot rule out small excesses in either direction. It also cannot confirm any clinically meaningful difference. Two years of follow-up is insufficient to detect most solid-tumor signals, given typical latency periods of 10 to 20 years for carcinogen-driven malignancies.

Post-Marketing Pharmacoepidemiological Evidence

Post-approval observational studies have extended the DEVOTE data into longer observation windows and broader patient populations. These studies carry their own limitations, including confounding by indication, immortal time bias, and incomplete outcome ascertainment, but they represent the best available long-term evidence.

Nordic Register Data

Scandinavian countries maintain population-level prescription and cancer registries that enable linkage studies with low ascertainment bias. A Swedish-Danish register analysis examining degludec initiators versus glargine initiators found no statistically significant difference in incident colorectal, breast, or prostate cancer over a median follow-up of 3.1 years (HR range 0.89 to 1.11 across cancer types, none reaching P<0.05) [6]. The study covered approximately 28,000 degludec initiations and is the largest real-world comparator dataset published to date.

Mechanistic Consideration: Hyperinsulinemia vs. Analog-Specific Risk

A recurring methodological problem in insulin-cancer research is separating the effects of exogenous insulin dose from any analog-specific molecular signal. Patients with poorly controlled diabetes receive higher total daily insulin doses, experience more hyperglycemia, and have greater baseline cancer risk through obesity-related mechanisms, regardless of which analog they use. Failure to adequately adjust for these confounders can generate spurious analog-specific signals.

The FDA addressed this explicitly in its 2012 Drug Safety Communication reviewing the glargine database, noting that "the observational studies have important methodological limitations and do not establish a causal relationship" between insulin glargine and cancer [7]. The same reasoning applies to degludec: any pharmacoepidemiological signal must survive rigorous confounder adjustment before clinical implications can be drawn.

Regulatory and Guideline Positions

FDA Label Status

The current FDA-approved prescribing information for Tresiba (degludec) does not contain any cancer warning, precaution, or contraindication [4]. Pharmacovigilance databases maintained by the FDA's Adverse Event Reporting System (FAERS) have not generated a disproportionality signal for malignancy associated with degludec that exceeded the reporting threshold for formal safety evaluation as of the most recent public summary report.

European Medicines Agency Assessment

The EMA's European Public Assessment Report (EPAR) for degludec reviewed pre-clinical carcinogenicity data including 2-year rat studies at supratherapeutic exposures. No treatment-related tumor findings emerged that were considered relevant to human risk [8]. The EMA's pharmacovigilance committee (PRAC) has not issued a signal assessment specific to degludec and malignancy.

ADA Standards of Care

The American Diabetes Association 2024 Standards of Care in Diabetes state that "insulin analogs currently approved for clinical use have not demonstrated an increased risk of cancer in large randomized trials" and do not recommend cancer-specific monitoring beyond standard age-appropriate screening for patients using any basal insulin analog [9].

The ADA guideline language is consistent with the available evidence: "There is no basis for selecting or avoiding a specific insulin analog based on cancer risk considerations in patients who otherwise qualify for basal insulin therapy" [9].

Specific Cancer Types: What the Sub-Group Data Show

Breast Cancer

Breast cancer has received disproportionate attention in insulin-analog safety discussions because IGF-1R is overexpressed in a significant proportion of ER-positive breast tumors and because MCF-7 assays use breast cancer cells. DEVOTE did not report breast cancer separately in its primary publication. The Nordic register study found an HR of 0.94 (95% CI 0.71 to 1.24) for incident breast cancer in degludec versus glargine initiators, providing no evidence of excess risk [6].

Colorectal Cancer

Colorectal cancer risk in insulin users has been a concern given the trophic effects of insulin on colonocytes. Observational data for degludec specifically show HRs clustering near 1.0, with no study reaching statistical significance for an increased risk [6].

Prostate Cancer

Prostate cancer data for degludec are sparse because the key trials enrolled predominantly older male patients with established cardiovascular disease, a population already at elevated prostate cancer baseline risk. No analog-specific signal has emerged in available datasets [6].

Practical Prescribing Considerations

When to Choose Degludec Over Glargine

The DEVOTE trial's secondary finding of 53% fewer nocturnal hypoglycemic events compared with glargine [5] is the primary clinical differentiator. Patients with nocturnal hypoglycemia unawareness, shift workers with variable sleep schedules, or patients who benefit from flexible once-daily dosing windows (up to 8 hours variation permitted per labeling) are reasonable degludec candidates on efficacy grounds. Cancer risk considerations do not factor into this decision based on available evidence.

Patients With Active or Prior Malignancy

No guideline from the ADA, AACE, or Endocrine Society specifically restricts degludec use in patients with active cancer or cancer survivors. Insulin therapy decisions in oncology patients are typically driven by glucose target considerations, hypoglycemia risk during chemotherapy-related anorexia, and steroid-induced hyperglycemia management protocols rather than by analog-specific molecular concerns. The endocrine-surgery literature generally recommends continuing pre-existing basal insulin regimens perioperatively unless a specific clinical reason exists to switch [9].

Monitoring Protocol

Standard cancer screening per age, sex, and risk-factor guidelines applies to all patients with diabetes, independent of insulin analog choice. There is no published recommendation for additional oncology surveillance in degludec users. Patients with strong family histories of breast, colorectal, or prostate cancer should follow standard genetics and screening referral pathways regardless of their insulin regimen.

Interpreting the Evidence: A Framework for Clinicians

Four questions structure a rational assessment of any insulin-cancer signal:

1. Is the in-vitro receptor-binding data biologically plausible as a human risk driver? For degludec, IGF-1R affinity is low and dissociation kinetics are normal, making a strong biological case for excess risk unlikely.

2. Do randomized trial data show a signal? DEVOTE (N=7,637) showed HR 1.00 for neoplasm events at 2 years, with confidence intervals too wide to exclude small differences but no point-estimate concern [5].

3. Do long-term observational data with confounder adjustment show a signal? Available Nordic register data covering more than 28,000 degludec initiations show HRs near 1.0 across major cancer types [6].

4. Have regulators identified a pharmacovigilance signal? Neither the FDA nor the EMA has issued a cancer-related safety communication specific to degludec [4][8].

All four answers currently point in the same direction: no actionable cancer signal exists for degludec at this time. Continued post-marketing surveillance, particularly studies with 10-year or longer follow-up, would strengthen this conclusion.