MIPI Calculator for Mantle Cell Lymphoma

Q: Is MIPI a points score?

No, and that is the most important thing about it. It is a continuous linear predictor: 0.03535 × age + 0.6978 if ECOG is above 1 + 1.367 × log10(LDH / ULN) + 0.9393 × log10(white cells per microlitre), banded at 5.7 and 6.2. The points-based version, with a 0-to-11 total, is the simplified MIPI, which is a separate instrument with its own cut-offs.

Q: What unit does the white cell count go in?

The published formula uses the count per 10⁶/L, which is per microlitre, so 9 ×10⁹/L is 9,000. This page asks for ×10⁹/L and multiplies internally. Doing it by hand with the count in ×10⁹/L subtracts 0.9393 × 3 = 2.818 from the index, which is enough to turn a high-risk patient into a low-risk one.

Q: Why does the LDH go in as a ratio?

Because LDH is an enzyme activity, not a concentration: it depends on the method and on the temperature the assay runs at, and upper limits of normal in routine use span roughly 190 to 280 U/L. Normalising to the reporting laboratory's own limit is the only way the coefficient transfers between hospitals. A ratio of exactly 1 contributes nothing to the index.

Q: What is MIPI-b?

The biological MIPI, which adds 0.02142 × the Ki-67 proliferation index as a percentage to the four-term formula — so a Ki-67 of 30% adds 0.64. It is not computed on this page because the published risk cut-offs for the biological index differ from the 5.7 and 6.2 used for the standard MIPI, and those cut-offs could not be read in any accessible source.

MIPI is NOT a points score. It is a continuous linear predictor in age, performance status, the LDH ratio and the logarithm of the white cell count, with published coefficients, banded afterwards — so a one-year change in age moves it, which no sum can do.

Last updated 8 Oct 2026

MIPI (Hoster 2008)

continuous index, then banded

Age (years)Enters the formula CONTINUOUSLY, multiplied by 0.03535. One extra year adds 0.03535 to the index, so ten years adds 0.3535 — enough on its own to cross the 0.5-wide intermediate band. This is the feature that distinguishes MIPI from every points score on this site: there is no age band and no rounding.

ECOG performance status 2 or worseThe only non-continuous term. The published formula adds 0.6978 when the performance status is greater than 1 and nothing otherwise, so the item is a step of 0.6978 — larger than the whole intermediate band is wide, which means this single answer can move a patient two strata.

Measured serum LDHThe measured activity in whatever unit the report uses. Enter that laboratory’s upper limit of normal below; only the ratio of the two enters the formula, so the unit cancels.

That laboratory’s upper limit of normal for LDHFrom the same report. The published term is 1.367 × log10(LDH / ULN), so a ratio of exactly 1 contributes nothing, a ratio of 2 contributes 0.41, and a ratio of 10 contributes 1.367. Entering an absolute LDH where a ratio belongs adds roughly 3.3 to the index and reads every patient as high risk.

White cell count (×10⁹/L)In ×10⁹/L, which is how every report prints it, and this page multiplies by 1,000 internally because the published formula’s term is log10 of the count per 10⁶/L — that is, per microlitre. The distinction matters: entering 9 where the formula wants 9,000 subtracts 0.9393 × 3 = 2.818 from the index, which takes a high-risk patient below the low-risk boundary. The conversion is done for you here; it is the single commonest error in hand calculations of this index.

6.12index valueExample

Age 65; ECOG 1; LDH 300 U/L against a laboratory upper limit of 250 U/L; white cells 9 ×10⁹/L

The published formula, its units, and the sum that is a different index

MIPI = 0.03535 × age (years) + 0.6978 (if ECOG > 1) + 1.367 × log₁₀(LDH / ULN) + 0.9393 × log₁₀(white cells per 10⁶/L)

Low risk <5.7 · intermediate 5.7 to <6.2 · high ≥ 6.2

MIPI-b, when Ki-67 is available, adds 0.02142 × Ki-67 (%)
Simplified MIPI is a different instrument: 0 to 3 points for each of age, ECOG, the LDH ratio and the white cell count, total 0–11, with low 0–3, intermediate 4–5 and high 6–11

this is a regression, not a score: four terms with four published coefficients, summed continuously and banded afterwards. There is no age band, no white-cell band and no LDH band in the index itself; treating it as a points sum gives a DIFFERENT instrument with different cut-offs
0.03535 per year of age: so a decade of age moves the index 0.3535 and twenty years moves it 0.707. The intermediate band is 0.5 wide, which means age alone spans more than one stratum across the range of a typical clinic
0.6978 for ECOG greater than 1: a single step, applied when the performance status is 2 or worse and not otherwise. It is larger than the whole intermediate band, so this one answer can carry a patient from low risk to high risk
1.367 × log10(LDH / ULN): the base-10 logarithm of the RATIO to the reporting laboratory’s upper limit of normal. A ratio of 1 contributes exactly nothing, 2 contributes 0.411, 0.5 contributes −0.411. An absolute LDH of 300 entered where a ratio belongs contributes 3.39 instead of 0.11 and reads high risk on every patient
0.9393 × log10(white cells per 10⁶/L): per 10⁶/L is per microlitre, so 9 ×10⁹/L is 9,000 in the formula. Two of the three documents read lost the superscript, so the unit was checked by magnitude too: in ×10⁹/L a typical presentation scores about 3.1, outside the index’s published range, while per microlitre it scores about 5.96
MIPI-b and Ki-67: where a Ki-67 proliferation index is available, the biological MIPI adds 0.02142 × Ki-67 as a percentage — so a Ki-67 of 30% adds 0.64. It is not computed here, because the published risk cut-offs for the biological index differ from the ones above and could not be read in any source available for this page
the cohort and its era: 455 patients with advanced-stage mantle cell lymphoma in three German and European trials, about 18% transplanted in first remission. Median survival not reached in the low-risk group after a median 32 months (five-year overall survival 60%), 51 months intermediate, 29 months high. The enrolment years could not be read in any accessible source and are not stated; the cohort predates BTK inhibitors and CAR T-cell therapy entirely
what a median is: a median is the middle of the derivation cohort and not a forecast: half of that cohort reached the figure and half did not, and a figure measured in a cohort treated decades ago is a historical measurement rather than an expectation

Worked example

Age 65; ECOG 1; LDH 300 U/L against a laboratory upper limit of 250 U/L; white cells 9 ×10⁹/L
Age term: 0.03535 × 65 = 2.2978
Performance status: ECOG 1 is not greater than 1, so 0
LDH ratio = 300 / 250 = 1.20, so the term is 1.367 × log₁₀(1.20) = 1.367 × 0.07918 = 0.1082
White cells: 9 ×10⁹/L is 9,000 per microlitre, so the term is 0.9393 × log₁₀(9000) = 0.9393 × 3.95424 = 3.7140
Total = 2.2978 + 0 + 0.1082 + 3.7140 = 6.12, which is the 5.7-to-6.2 intermediate band — median survival 51 months in the derivation cohort
Now add one year of age. 66 gives 6.16: the index moved by exactly 0.03535, which is what a continuous formula does and what no points sum can do
The two unit traps, both of which this page closes. Entering the white cell count as 9 rather than 9,000 gives a total of 3.30 — comfortably low risk, wrong by two strata. Entering the measured LDH of 300 as though it were the ratio gives 9.40 — high risk on a barely raised LDH
The simplified MIPI on the same patient scores 2 (age 60–69) + 0 (ECOG 0–1) + 2 (ratio 1.00–1.49) + 1 (white cells 6.7–9.999) = 5 of 11, which is its intermediate band. The two instruments agree here and do not always

The four terms, their coefficients and their units

Notice which term dominates at ordinary values: the white cell logarithm, because log10(9000) is nearly 4. That is also why the unit matters — a thousandfold error subtracts 2.818 from a figure whose whole low-to-high range is about 1.
Term	Coefficient	Enters as	Contribution at the default inputs
Age	0.03535	Years, continuous	2.2978 at 65 years
ECOG performance status	0.6978	A step, applied only when ECOG is greater than 1	0 at ECOG 1
LDH	1.367	log₁₀ of the ratio to the laboratory ULN	0.1082 at a ratio of 1.20
White cell count	0.9393	log₁₀ of the count per 10⁶/L (per microlitre)	3.7140 at 9 ×10⁹/L
Ki-67, in MIPI-b only	0.02142	Percentage, continuous	not computed on this page

The continuous MIPI and the simplified MIPI, which are different instruments

Both are computed above. The simplified version exists so the index can be scored without a calculator, and the price is that it bands three continuous variables and so disagrees with the continuous form near the boundaries, in both directions. The point table was read as text out of a statistical analysis plan on clinicaltrials.gov — which also says that for an upper limit of 240 U/L the LDH cut-points fall at 180, 240 and 360 U/L. 240 and 360 are exactly the 1.00 and 1.50 ratio boundaries; 180 is a ratio of 0.75, not the 0.67 the same table prints, which would put the first cut-point at 161 U/L. The printed band table is what is implemented, and the discrepancy can only affect a sub-normal LDH.
	MIPI	Simplified MIPI
Shape	Continuous linear predictor	Points sum
Scale	About 4 to 8 in practice	0 to 11
Age	0.03535 per year	0 if under 50, 1 if 50–59, 2 if 60–69, 3 if 70 or over
ECOG	0.6978 if greater than 1	0 if 0–1, 2 if 2–4
LDH ratio	1.367 × log₁₀(ratio)	0 if under 0.67, 1 if 0.67–0.99, 2 if 1.00–1.49, 3 if 1.5 or over
White cells	0.9393 × log₁₀(per microlitre)	0 if under 6.7, 1 if 6.7–9.999, 2 if 10–14.999, 3 if 15 or over (×10⁹/L)
Low risk	Below 5.7	0 – 3
Intermediate risk	5.7 to below 6.2	4 – 5
High risk	6.2 or above	6 – 11

The one index here that computes rather than sums

The Mantle Cell Lymphoma International Prognostic Index is the odd one out in this category, and the difference is not cosmetic. Every other instrument here assigns points to answers and adds them up; MIPI is a Cox regression — four terms, four published coefficients, summed on a continuous scale and only then cut into three strata at 5.7 and 6.2. Three of the four terms are continuous, so one extra year of age moves the result by exactly 0.03535 and a 10% change in the white cell count moves it by about 0.039. A points sum cannot do either, and that is the test of whether an implementation has understood what it is computing.

Two terms carry a unit trap, and between them they account for most wrong MIPI values. The LDH term is 1.367 times the base-10 logarithm of the RATIO to the reporting laboratory’s own upper limit of normal: a ratio of 1 contributes exactly nothing, while an absolute LDH of 300 entered where a ratio belongs contributes 3.39 instead of 0.11 and makes every patient high risk. The white cell term is 0.9393 times the logarithm of the count PER 10⁶/L — per microlitre, not per litre — so 9 ×10⁹/L is 9,000 in the formula, and entering 9 subtracts 2.818 from a figure whose whole low-to-high span is about one unit. This page takes the count in ×10⁹/L, as every report prints it, and converts it itself.

The simplified MIPI is shown alongside because it is frequently quoted as though it were the same number. It bands age, the LDH ratio and the white cell count into four each, scores ECOG as 0 or 2, and totals 0 to 11 with strata at 0–3, 4–5 and 6–11; banding three continuous variables discards the information the continuous form uses, so the two disagree near the boundaries in both directions. Neither is wrong, and a record should say which was computed. The index is also narrow in its population: derived in ADVANCED-STAGE disease, in 455 patients in three European trials, about 18% of them transplanted in first remission, with median survival not reached in the low-risk group, 51 months intermediate and 29 months high. Every outcome figure here describes the cohort the index was derived in and not the patient in front of you; the spread within one stratum is wider than the gap between strata. Those figures predate cytarabine-containing induction, rituximab maintenance, BTK inhibitors and CAR T-cell therapy. Every laboratory threshold here is method- and laboratory-dependent, so the reader’s own laboratory’s reference interval takes precedence. A risk stratum is not a diagnosis and not a plan. This page computes the index, names the stratum and reports what it predicted in the derivation cohort; what follows is a decision for the treating team with the patient.

Frequently asked questions

Is MIPI a points score?

No, and that is the most important thing about it. It is a continuous linear predictor: 0.03535 × age + 0.6978 if ECOG is above 1 + 1.367 × log10(LDH / ULN) + 0.9393 × log10(white cells per microlitre), banded at 5.7 and 6.2. The points-based version, with a 0-to-11 total, is the simplified MIPI, which is a separate instrument with its own cut-offs.

What unit does the white cell count go in?

The published formula uses the count per 10⁶/L, which is per microlitre, so 9 ×10⁹/L is 9,000. This page asks for ×10⁹/L and multiplies internally. Doing it by hand with the count in ×10⁹/L subtracts 0.9393 × 3 = 2.818 from the index, which is enough to turn a high-risk patient into a low-risk one.

Why does the LDH go in as a ratio?

Because LDH is an enzyme activity, not a concentration: it depends on the method and on the temperature the assay runs at, and upper limits of normal in routine use span roughly 190 to 280 U/L. Normalising to the reporting laboratory’s own limit is the only way the coefficient transfers between hospitals. A ratio of exactly 1 contributes nothing to the index.

What is MIPI-b?

The biological MIPI, which adds 0.02142 × the Ki-67 proliferation index as a percentage to the four-term formula — so a Ki-67 of 30% adds 0.64. It is not computed on this page because the published risk cut-offs for the biological index differ from the 5.7 and 6.2 used for the standard MIPI, and those cut-offs could not be read in any accessible source.

Related calculators

References

Hoster E, et al. A new prognostic index for patients with advanced-stage mantle cell lymphoma (the MIPI). Blood. 2008. The derivation paper, named as the source of the coefficients by every document above that carries them; 455 advanced-stage patients in three German and European trials.
van de Schans SAM, et al. Validation of the Mantle Cell Lymphoma International Prognostic Index in a population-based cohort of 178 patients. Haematologica. 2010. Source of the MIPI cut-offs of 5.7 and 6.2.
BeiGene. Protocol for NCT03206970: the MIPI and MIPI-b formulas with every coefficient, the LDH term as LDH divided by the upper limit of normal, and the Ki-67 coefficient. clinicaltrials.gov.
Kite Pharma. Statistical analysis plan for ZUMA-2 (NCT02601313), Appendix 4: simplified MIPI point table and risk categories. clinicaltrials.gov.
MIPI derivation cohort and outcomes, from two independent sources: 455 advanced-stage patients in three German and European trials, about 18% transplanted in first remission; median survival not reached in the low-risk group (5-year overall survival 60%), 51 months intermediate, 29 months high.

Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/