There is a recently proposed method to speed up grid search:
"Fast Cross validation via sequential analysis"
http://www.scribd.com/doc/76134034/Fast-Cross-Validation-Via-Sequential-Analysis-Talk
Basically, they're doing a normal grid search, but try to eliminate bad parameters early in the process and not waste too much computation on them. It's fairly new and I don't know independent evaluations of their method, but I'm currently implementing it and want to give it a try.
To elaborate on Frank Harrell's answer, what the Epi package did was to fit a logistic regression, and make a ROC curve with outcome predictions of the following form:
$$
outcome = \frac {1}{1+e^{-(\beta_0 + \beta_1 s100b + \beta_2 ndka)}}
$$
In your case, the fitted values are $\beta_0$ (intercept) = -2.379, $\beta_1$ (s100b) = 5.334 and $\beta_2$ (ndka) = 0.031. As you want your predicted outcome to be 0.312 (the "optimal" cutoff), you can then substitute this as (hope I didn't introduce errors here):
$$
0.312 = \frac {1}{1+e^{-(-2.379 + 5.334 s100b + 0.031 ndka)}}
$$
$$
1.588214 = 5.334 s100b + 0.031 ndka
$$
or:
$$
s100b = \frac{1.588214 - 0.031 ndka}{5.334}
$$
Any pair of (s100b, ndka) values that satisfy this equality is "optimal". Bad luck for you, there are an infinity of these pairs. For instance, (0.29, 1), (0, 51.2), etc. Even worse, most of them don't make any sense. What does the pair (-580, 10000) mean? Nothing!
In other words, you can't establish cut-offs on the inputs - you have to do it on the outputs, and that's the whole point of the model.
Best Answer
The optimal tuning parameter ($\texttt{spar}$) in R function $\texttt{smooth.spline()}$ is automatically determined by min GCV by default (with option $\texttt{cv=FALSE}$). The other option (by setting $\texttt{cv=TRUE}$) is the jackknife (i.e., leave-one-out) based PRESS. Recall that GCV is simplified from PRESS by merely replacing the diagonal elements $S_{ii}$ of the smoothing matrix, say, $\mathbf{S}$ with their average $\mbox{tr}(\mathbf{S})/n$. Usually the best tuning parameter choice from PRESS is similar to that from GCV, though GCV is easier to compute.