Quick answer: the standard preclinical formula is V = (L × W²) ÷ 2, where L is the longest diameter and W the shortest perpendicular one. A tumour 10 × 8 mm has a volume of 320 mm³. With three measured diameters, use the full ellipsoid V = 0.5236 × L × W × H instead. Note that RECIST does not use volume at all — it uses summed diameters.
Research and educational tool — not for clinical decision-making. These formulas are geometric approximations used mainly in preclinical and laboratory work. They are not a diagnostic instrument, they do not assess treatment response, and they must not be used to interpret your own or a patient's imaging. Clinical response assessment follows RECIST or another validated system and belongs with an oncology team.
Tumor volume calculation formulas, and why the /2
Full ellipsoid: V = 0.5236 × L × W × H (π/6)
Sphere: V = 0.5236 × d³
All three are the same idea with different amounts of information. Start from the ellipsoid: V = (π/6) × L × W × H = 0.5236 × L × W × H.
Now the practical problem. Measuring a subcutaneous tumour with calipers gives you two dimensions, not three — you cannot measure depth through skin. So the modified formula makes one assumption: the unmeasured height equals the measured width. Substituting H = W:
- V = 0.5236 × L × W × W = 0.5236 × L × W²
- Round 0.5236 to 0.5 for hand calculation
- V = (L × W²) ÷ 2
That is the entire derivation. The division by two is not arbitrary — it is π/6 rounded down, with the assumption that the tumour is as deep as it is wide.
Worked example. A tumour 10 mm long and 8 mm wide: V = (10 × 64) ÷ 2 = 320 mm³. If imaging later shows the depth really is 8 mm, the full ellipsoid gives 0.5236 × 10 × 8 × 8 = 335 mm³ — 4.7% higher, purely from the rounded coefficient.
Which formula to use
| Formula | Needs | Typical use | 10×8×8 mm gives |
|---|---|---|---|
| Modified ellipsoid (L × W²)/2 | 2 diameters | Caliper measurement of rodent xenografts — the field standard | 320 mm³ |
| Full ellipsoid 0.5236 × L×W×H | 3 diameters | Imaging with three perpendicular planes | 335 mm³ |
| Sphere 0.5236 × d³ | 1 diameter | Rough estimate; roughly spherical lesions | 524 mm³ (d = 10) |
| Segmentation | Full image stack | Reference standard; irregular tumours | Measured directly |
The sphere row uses the longest diameter and is therefore a substantial overestimate for anything elongated — 524 mm³ against 320. It is included because it appears in quick estimates, not because it is a good default.
The rule that matters more than the choice: pick one formula and keep it for the whole study. A change of method mid-series produces an apparent step change in volume that has nothing to do with the tumour. Report which formula you used.
RECIST uses diameters, not volume — and the gap is large
This is the single most useful thing on the page, because it is where volume figures and clinical reporting get conflated.
RECIST 1.1, the standard for solid-tumour response in clinical trials, works on the sum of longest diameters. Partial response is a ≥30% decrease; progressive disease is a ≥20% increase. Volume never enters it.
But volume scales with the cube of a linear dimension. So those thresholds translate very differently:
| RECIST category | Diameter change | Volume factor | Volume change |
|---|---|---|---|
| Complete response | −100% | 0 | −100% |
| Partial response | −30% | 0.70³ = 0.343 | −65.7% |
| Stable disease (lower edge) | −29% | 0.71³ = 0.358 | −64.2% |
| Stable disease (upper edge) | +19% | 1.19³ = 1.685 | +68.5% |
| Progressive disease | +20% | 1.20³ = 1.728 | +72.8% |
| Doubling in diameter | +100% | 2³ = 8 | +700% |
Read the "stable disease" rows. A tumour can grow by nearly 70% in volume and still be classified as stable disease under RECIST, because its diameter moved less than 20%. That is not a flaw in RECIST — diameter is far more reproducible to measure than volume, and the thresholds were chosen with that in mind. But it does mean a volume change and a RECIST category are answering different questions, and quoting one as though it were the other is misleading.
Volume doubling time
Assuming exponential growth between two measurements separated by time t. A tumour going from 150 to 300 mm³ over 14 days:
- Ratio: 300 ÷ 150 = 2
- ln 2 ÷ ln 2 = 1
- Td = 14 × 1 = 14 days
Which is obvious when the volume exactly doubles — the value of the formula is for the cases where it does not. From 150 to 260 mm³ over the same 14 days gives Td = 14 × 0.693 ÷ 0.550 = 17.6 days.
Real tumours do not grow exponentially indefinitely. Growth typically slows as the tumour outgrows its blood supply, better described by Gompertzian kinetics. Doubling time from two points is a useful comparative figure over short intervals, not a prediction. Two measurements very close together also produce wildly unstable estimates, because measurement noise dominates the ratio.
How much to trust a caliper number
Caliper measurement is standard because it is quick, cheap and non-terminal. It is not accurate in absolute terms, and it helps to know why:
- It includes the skin. Overlying tissue is measured along with the tumour, inflating both dimensions.
- It cannot see inside. Necrotic or cystic cores are counted as tumour.
- Errors compound. W is squared, so a 5% error in width becomes about 10% in volume; a 5% error in both L and W gives roughly 16%.
- Irregular and flat tumours break the model. The ellipsoid assumption fails, usually overestimating.
- Operators differ. Inter-observer variability is frequently larger than the effect sizes studies are trying to detect.
Published comparisons against imaging or excised tumour weight routinely show discrepancies of 20% or more. The practical implication: calipers are good for relative change within a consistently measured series, and poor for absolute volume. If absolute volume matters, use imaging with segmentation.
| Measurement error | Effect on volume (L×W²/2) |
|---|---|
| +5% on length only | +5% |
| +5% on width only | +10.3% |
| +5% on both | +15.8% |
| +10% on both | +33.1% |
| +1 mm on a 5 mm tumour (both) | +72.8% |
That last row is why small tumours are so hard to measure reliably. At 5 mm, a single millimetre of caliper imprecision changes the calculated volume by nearly three-quarters.
Common mistakes & pro tips
- Swapping L and W. W must be the shorter perpendicular dimension. Reversing them on a 10 × 8 tumour gives 400 mm³ instead of 320 — a 25% error.
- Changing formula mid-study. Produces an artefactual step in the growth curve.
- Comparing a volume change to a RECIST category. They are different measures; see the table above.
- Reporting volume without the method. "320 mm³" is not reproducible unless the formula is stated.
- Mixing units. Millimetres give mm³; centimetres give cm³ (= mL). 1,000 mm³ = 1 cm³.
- Pro tip: one operator, one instrument. Inter-observer variability usually exceeds the biological effect being measured.
- Pro tip: record L and W, not just V. Raw dimensions let anyone recompute with a different formula later; a volume alone cannot be reverse-engineered.
How to use this calculator
- Modified ellipsoid — the caliper standard. Enter the long and short diameters.
- Full ellipsoid — enter three perpendicular diameters from imaging.
- Sphere — a single diameter, for a rough estimate only.
- Doubling time — two volumes and the interval between them.
- Read the comparison rows — each mode shows what the other formulas would have given for the same measurements.
Frequently asked questions
How do you calculate tumor volume?
The most widely used method in preclinical research is the modified ellipsoid formula: volume = length × width² ÷ 2. Length is the longest diameter and width is the shortest diameter perpendicular to it. A tumour measuring 10 × 8 mm has a volume of 10 × 8² ÷ 2, which is 320 mm³. If three perpendicular diameters are available from imaging, the full ellipsoid formula multiplies all three by 0.5236.
Why is the tumor volume formula divided by two?
Because it approximates an ellipsoid in which the two unmeasured dimensions are both taken to equal the width. The full ellipsoid formula is π/6 × length × width × height, which is 0.5236 × the product. Setting height equal to width and rounding 0.5236 to 0.5 gives the familiar length × width² ÷ 2. The approximation is deliberate, and it makes the formula workable with a single pair of caliper readings.
What is the difference between the ellipsoid and modified ellipsoid formulas?
The full ellipsoid formula uses three measured diameters and the coefficient 0.5236, which is π/6. The modified ellipsoid formula uses only two, assumes the third equals the width, and rounds the coefficient to 0.5. For a tumour 10 × 8 × 8 mm the two give 335 and 320 mm³ respectively, a difference of about 4.5%. The modified version is standard in caliper-based rodent studies because only two dimensions can be measured through skin.
Does RECIST use tumor volume?
No. RECIST version 1.1, the standard for clinical trial response assessment in solid tumours, uses the sum of longest diameters, not volume. Partial response is a decrease of at least 30% in that sum, and progressive disease is an increase of at least 20%. Because volume scales with the cube of a diameter, those diameter thresholds correspond to roughly a 66% fall and a 73% rise in volume, which is why volume figures and RECIST categories cannot be compared directly.
How do you calculate tumor doubling time?
Doubling time = (time between measurements × ln 2) ÷ ln(V₂ ÷ V₁). A tumour growing from 150 to 300 mm³ over 14 days has a doubling time of 14 × 0.693 ÷ 0.693, which is exactly 14 days. The calculation assumes exponential growth, which real tumours only approximate.
How accurate is caliper measurement of tumor volume?
It is reproducible enough to track a trend but not accurate in absolute terms. Caliper readings include overlying skin, cannot detect necrotic cores, and become unreliable for irregular or flat tumours. Published comparisons against imaging and against excised tumour weight commonly show discrepancies of 20% or more. Because both dimensions are squared or multiplied together, a small measurement error compounds, so the same operator should take all readings in a study.
References & further reading
- Tomayko, M. M. & Reynolds, C. P. (1989), "Determination of subcutaneous tumor size in athymic (nude) mice", Cancer Chemotherapy and Pharmacology — the caliper method.
- Euhus, D. M. et al. (1986), Journal of Surgical Oncology — the modified ellipsoid formula.
- Eisenhauer, E. A. et al. (2009), "New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1)", European Journal of Cancer.
- Faustino-Rocha A, Oliveira PA, Pinho-Oliveira J, et al. Estimation of rat mammary tumor volume using caliper and ultrasonography measurements. Lab Anim (NY) 2013;42(6):217–224. doi:10.1038/laban.254
Research and educational use only. This calculator applies published geometric formulas. It is not a diagnostic tool, does not perform response assessment, and must not inform clinical decisions. See our accuracy policy.