GIPSS: the genetically inspired prognostic scoring system (Tefferi 2018)
GIPSS is a points score for primary myelofibrosis built entirely from genetic results: the karyotype and four mutation findings. Each finding adds 0, 1 or 2 points, the total runs from 0 to 6, and the total places the patient in one of four risk groups. Magent's GIPSS calculator takes the results you've already assigned and returns the points, the group and the survival figures the 2018 paper reported for that group's cohort.
Not a medical device. Not a diagnosis or dosing recommendation.
What is GIPSS?
Tefferi and colleagues proposed GIPSS as a prognostic system that uses no clinical variables at all. Age, blood counts and symptoms play no part. Instead, the score rests on a three-tier karyotype and on whether certain mutations are present or absent.
The calculator doesn't karyotype or sequence anything. It adds points for results the caller supplies.
The GIPSS formula
GIPSS = karyotype points (favorable 0, unfavorable 1, very high risk 2)
+ 1 if a type 1/like CALR mutation is absent
+ 1 if ASXL1 is mutated
+ 1 if SRSF2 is mutated
+ 1 if U2AF1 Q157 is mutated
total 0 to 6Exact expression the calculator uses:
GIPSS = karyotype (favorable 0, unfavorable 1, vhr 2) + absence of type 1/like CALR 1 + asxl1_mutated 1 + srsf2_mutated 1 + u2af1_q157_mutated 1; Tefferi 2018; max 6; low 0, intermediate_1 1, intermediate_2 2, high ≥3Inputs
| Input | Meaning | Points |
|---|---|---|
karyotype | favorable, unfavorable or vhr (very high risk), as defined in the paper | 0, 1 or 2 |
calr_type1_like | A type 1/like CALR mutation is present | false scores 1, true scores 0 |
asxl1_mutated | ASXL1 mutation | true scores 1 |
srsf2_mutated | SRSF2 mutation | true scores 1 |
u2af1_q157_mutated | U2AF1 mutation at Q157 | true scores 1 |
The favorable tier is a normal karyotype or one of a short list of sole abnormalities, such as 20q− or loss of Y. The very-high-risk tier covers abnormalities such as −7, i(17q) and inv(3). Everything else is unfavorable. A U2AF1 S34 mutation doesn't score.
How the calculation works
- Score the karyotype on the three tiers: 0, 1 or 2.
- Check CALR. A missing type 1/like CALR mutation adds 1. Type 2/like CALR, JAK2, MPL and triple-negative cases all count as absent.
- Add 1 each for ASXL1, SRSF2 and U2AF1 Q157 mutations.
- Assign the group from the total:
| Total | Group | Cohort median survival | Cohort 5-year survival |
|---|---|---|---|
| 0 | low | 26.4 years | 94% |
| 1 | intermediate_1 | 8.0 years | 73% |
| 2 | intermediate_2 | 4.2 years | 40% |
| 3 to 6 | high | 2.0 years | 14% |
The survival columns are the figures Tefferi 2018 reported for each group in the study cohort. They describe groups of patients, not what will happen to one person.
Worked examples
These use Magent's own expected values for the calculator.
Example 1: favorable karyotype with a type 1 CALR mutation
- Karyotype favorable: 0.
- Type 1/like CALR present: 0.
- No ASXL1, SRSF2 or U2AF1 Q157 mutation: 0.
- GIPSS 0, group
low(cohort median 26.4 years, 94% at five years).
Example 2: unfavorable karyotype and ASXL1
- Karyotype unfavorable: 1.
- Type 1/like CALR present: 0.
- ASXL1 mutated: 1.
- 1 + 0 + 1 = 2.
- GIPSS 2, group
intermediate_2(cohort median 4.2 years, 40% at five years).
Example 3: very-high-risk karyotype, no type 1 CALR, ASXL1 and SRSF2
- Karyotype vhr: 2.
- Type 1/like CALR absent: 1.
- ASXL1 mutated: 1. SRSF2 mutated: 1.
- 2 + 1 + 1 + 1 = 5.
- GIPSS 5, group
high(cohort median 2.0 years, 14% at five years).
History of GIPSS
Tefferi and colleagues published GIPSS in Leukemia in 2018, in "GIPSS: genetically inspired prognostic scoring system for primary myelofibrosis". It was designed as a score based on genetic markers alone, unlike earlier systems that also use clinical and blood count variables.
When it's used
When karyotype and mutation results for primary myelofibrosis are available and a group label from the 2018 paper is wanted alongside other prognostic information.
Where it's used
Haematology clinics, multidisciplinary meetings and research.
Who uses it
Haematologists, haematopathologists and researchers. The calculator adds the points and attaches the published group figures.
Limits of the calculation
- It depends entirely on correct karyotype tiering and mutation calls, which the calculator doesn't check.
- The survival figures come from one cohort and describe groups, not individuals.
- It doesn't identify primary myelofibrosis and isn't meant for other myeloid conditions.
- It isn't DIPSS or MIPSS70, which use different inputs.
- The calculator gives no advice about transplant or any other step.
- This page doesn't say what a score means for a particular person.
Clinical decisions belong to a qualified clinician who knows the patient.
How it differs from related calculators
- Absolute lymphocyte count works out a cell count from a blood count; it isn't a prognostic score.
- DIPSS and MIPSS70 also grade myelofibrosis but add clinical variables. They aren't part of this calculator.
Glossary
- Karyotype: the chromosome pattern seen in dividing cells.
- CALR, ASXL1, SRSF2, U2AF1: genes whose mutations are tested in myeloid conditions.
- Type 1/like CALR: a group of CALR mutations treated as favorable in GIPSS.
- Median survival: the time by which half of a group had died.
Common mistakes
- Scoring CALR the wrong way round. Points go to the absence of a type 1/like CALR mutation.
- Counting U2AF1 S34. Q157 scores; S34 doesn't.
- Treating the survival figures as a forecast. They're cohort figures.
- Mixing it up with DIPSS. The inputs are different.
Source
Tefferi A, et al. GIPSS: genetically inspired prognostic scoring system for primary myelofibrosis. Leukemia. 2018;32(7):1631-1642. DOI: 10.1038/s41375-018-0107-z. PMID: 29654267
Frequently asked questions
How is GIPSS calculated?
Add karyotype points (0, 1 or 2) to 1 point each for a missing type 1/like CALR mutation and for ASXL1, SRSF2 and U2AF1 Q157 mutations. The maximum is 6.
What are the GIPSS risk groups?
0 is low, 1 intermediate_1, 2 intermediate_2 and 3 or more high.
Are the survival figures a prediction for one patient?
No. They're the median and five-year survival the 2018 paper reported for each group in its cohort.
Does GIPSS use age or blood counts?
No. It uses genetic results alone.
Is GIPSS the same as DIPSS?
No. DIPSS uses clinical and blood count variables.