The phrase "a marginal survival" is correct and usable in written English.
It can be used to describe something that exists in a limited or minimal capacity, often referring to species, cultures, or practices that persist despite adverse conditions.
Example: "The ancient language is a marginal survival, spoken only by a handful of elderly individuals in the remote village."
Alternatives: "a tenuous existence" or "a fragile continuation".
Monotherapy of pancreatic cancer with gemcitabine results in only a marginal survival benefit [ 3].
However, due to pre-existing or acquired chemoresistance, gemcitabine treatment has a marginal survival benefit and yields an objective tumor response rate of < 10% [ 5, 6].
In addition, these Singaporean women with local cancer showed a marginal survival advantage over the Swedish cohort in the later period.
In a phase III trial comparing gemcitabine versus 5-flurouracil in advanced pancreatic cancer, patients who received gemcitabine showed a modest improvement in response rate, a marginal survival advantage, and most important, superior clinical response [ 20].
Gemcitabine monotherapy has modest clinical benefit and a marginal survival advantage in patients with advanced PDAC, 2 however, the median survival of patients with metastatic PDAC remains poor, and is often less than 6 months. 2 More recently, encouraging results have been observed in clinical trials with the FOLRIFINOX regimen, 3 although many patients are unable to tolerate this regimen.
However, Eq. (1) is at the individual level and often it is desired to present an average (marginal survival).
Hsu et al. converted interval-censored data problems to right censored data problems using auxiliary variables via multiple imputation in a situation with independent censoring to estimate the marginal survival distribution [ 10].
In order to use the information from the auxiliary variables to improve the marginal survival estimate, Hsu et al. considered a situation of right censored data with possibly multiple time-independent or time-dependent continuous covariates and proposed deriving two risk scores [ 17].
The traditional lifetable approaches have averaged the numerator and denominator separately, which is not the same as taking an average of individual relative survival and is a key reason why traditional approaches have potential bias when estimating marginal survival.
A less formal strategy is the graphical comparison between the Kaplan-Meier marginal survival probability and the marginal probability obtained from the transformation model without covariates.
In 1990 Alan Grafen showed that a handicap-like signalling system was evolutionarily stable if higher quality signallers paid lower marginal survival costs for their signals.
