SPN Malignancy Risk Calculator

SPN Malignancy Risk Calculator for Pulmonology. The output is a continuous percentage, conventionally banded into three actions. Low probability (under about 5 to 10 percent) supports serial CT surveillance and generally spares the patient invasive testing. Intermediate probability (roughly 10 to 65 percent) is the zone where FDG-PET or nonsurgical biopsy adds the most value, and where the Herder PET extension is most useful for reclassifying up or down. High probability (above about 65 percent) supports proceeding to biopsy or surgical resection. The exact cut points are guideline- and center-dependent, and small nodules near the size floor cluster in the low band regardless of other features.

How this calculator works

This tool estimates the pretest probability that a radiologically indeterminate solitary pulmonary nodule (SPN) is malignant using the Mayo Clinic logistic regression model of Swensen et al. It combines three clinical variables (age, current or former smoking status, prior extrathoracic cancer diagnosed 5 or more years earlier) with three CT/radiographic features (nodule diameter in mm, presence of spiculation, and upper-lobe location). Each predictor carries a fixed regression coefficient; the weighted sum is passed through the logistic function odds/(1+odds) to yield a percentage probability. When FDG-PET data are available, the Herder extension re-weights the Mayo output by PET uptake intensity (no, faint, moderate, or intense metabolic activity) to refine the estimate.

When to use this calculator

Apply this to solid, indeterminate SPNs roughly 8 to 30 mm on CT in adults being triaged between surveillance, PET/biopsy, and resection. It is the model embedded in British Thoracic Society and referenced in ACCP/CHEST nodule guidance for intermediate-risk lesions. Do not use it for subsolid (ground-glass or part-solid) nodules, nodules under 8 mm, multiple nodules, immunocompromised patients, or those with known active extrapulmonary malignancy, since the derivation cohort excluded these and the model will misclassify them. It is a triage aid, not a substitute for tissue diagnosis.

Inputs used

  • Age
  • Smoking history
  • Cancer history
  • Nodule diameter
  • Nodule location
  • Spiculation
  • Imaging characteristics

Clinical interpretation

The output is a continuous percentage, conventionally banded into three actions. Low probability (under about 5 to 10 percent) supports serial CT surveillance and generally spares the patient invasive testing. Intermediate probability (roughly 10 to 65 percent) is the zone where FDG-PET or nonsurgical biopsy adds the most value, and where the Herder PET extension is most useful for reclassifying up or down. High probability (above about 65 percent) supports proceeding to biopsy or surgical resection. The exact cut points are guideline- and center-dependent, and small nodules near the size floor cluster in the low band regardless of other features.

Worked example

A 68-year-old current smoker with a prior colon cancer resected 7 years ago has a 16 mm spiculated nodule in the right upper lobe. All six risk features are present (older age, smoking, remote cancer, sizable diameter, spiculation, upper-lobe site), so the Mayo model returns a high probability near 70 to 75 percent. This crosses the greater-than-65-percent threshold, so guidelines favor definitive workup (PET then biopsy or surgical resection) rather than CT surveillance. If subsequent FDG-PET shows intense uptake, the Herder-adjusted probability rises above 90 percent, further supporting resection.

Limitations and safety notes

The Mayo model was derived from chest-radiograph-detected nodules in a low-endemic region for granulomatous disease; in populations with high histoplasmosis or tuberculosis prevalence it overcalls malignancy because benign granulomas mimic spiculated solid nodules. External validation found it tends to underestimate risk at the low end of its range (Herder 2005). It ignores emphysema, family history, asbestos, and nodule growth rate, and validation AUCs (roughly 0.75 to 0.85) mean substantial misclassification persists. Head-to-head studies (Balekian 2013, Tanner 2017) found experienced clinicians match or beat the calculator, so a discordant gestalt should not be overridden by the number.

Frequently asked questions

How is the Mayo model different from the Herder model?

The Mayo model uses only clinical and CT features to give a pretest probability. The Herder model takes that pretest estimate and adjusts it by FDG-PET uptake intensity, so it is meant for use after a PET scan has been done and generally sharpens accuracy in the intermediate range.

Can I use this for a ground-glass or part-solid nodule?

No. The model was validated only on solid indeterminate nodules. Subsolid nodules follow different natural histories (often indolent adenocarcinoma spectrum) and require dedicated subsolid-nodule algorithms such as those in the Fleischner or BTS subsolid pathways.

Why does prior cancer only count if it was 5 or more years ago?

The variable was designed to capture a remote cancer history as a marker of malignancy predisposition rather than recent or active disease. A nodule in a patient with a recently treated or active extrathoracic cancer is more likely a metastasis and falls outside the model's derivation population, so the calculator should not be applied there.

What smoking definition does the age/smoking term use?

The model treats smoking as current or former (ever) smoking versus never. It does not weight pack-years continuously, which is one reason heavy-smoking and light-smoking ever-smokers receive the same contribution and why clinical judgment about intensity still matters.

Should the percentage decide surgery on its own?

No. It sets the pretest probability that frames the next step. A high value points toward PET, biopsy, or resection and a low value toward surveillance, but multidisciplinary review, patient comorbidity and preference, and biopsy feasibility drive the final decision.

References

  • Swensen SJ, Silverstein MD, Ilstrup DM, Schleck CD, Edell ES. The probability of malignancy in solitary pulmonary nodules. Application to small radiologically indeterminate nodules. Arch Intern Med. 1997. PMID: 9129544.
  • Swensen SJ, Silverstein MD, Edell ES, Trastek VF, Aughenbaugh GL, Ilstrup DM, Schleck CD. Solitary pulmonary nodules: clinical prediction model versus physicians. Mayo Clin Proc. 1999. PMID: 10221459.
  • Herder GJ, van Tinteren H, Golding RP, Kostense PJ, Comans EF, Smit EF, Hoekstra OS. Clinical prediction model to characterize pulmonary nodules: validation and added value of 18F-fluorodeoxyglucose positron emission tomography. Chest. 2005. PMID: 16236914.
  • Perandini S, Soardi GA, Larici AR, et al. Multicenter external validation of two malignancy risk prediction models in patients undergoing 18F-FDG-PET for solitary pulmonary nodule evaluation. Eur Radiol. 2016. PMID: 27631108.
  • Balekian AA, Silvestri GA, Simkovich SM, Mestaz PJ, Sanders GD, Daniel J, Porcel J, Gould MK. Accuracy of clinicians and models for estimating the probability that a pulmonary nodule is malignant. Ann Am Thorac Soc. 2013. PMID: 24063427.

Editorial review and citation methodology

Reviewed by the Quick Medical Calculator Editorial Team. Last reviewed: May 16, 2026. The review checks calculator inputs, intended population, interpretation, limitations, and source alignment.

  • Prefer original validation studies for scoring systems and prediction tools.
  • Use current specialty society guidance, transplant allocation policy, public health guidance, or regulator resources when they govern clinical use.
  • Include limitations and safety notes when a calculator is population-specific, context-dependent, or unsuitable as a standalone decision tool.

Related reviewed calculators