# Lasso cox regression with bootstrap

I'm looking at building a nomogram for cancer prognosis based on 20 variables. This will be derived from a cox ph model. In the past I used poor methodology including dichotomization and stepwise selection. I am looking to improve the methodology substantially on an upcoming study.

I understand that using Lasso for variable selection is now built into the glmnet R package. This is what I will use initially to select variables for cox regression.

However for internal validation, I understand that bootstrapping may be a superior process to k-fold cross validation.

While I am reasonably familiar with R, this methodological sequence to build the nomogram is very foreign to me. In terms of work flow, can anyone point me in the right direction on how to build a Cox PH model with Lasso that is internally validated with bootstrapping?

If you have a very large number of candidate predictors then you will have to use a penalized approach. As your interest is in prediction, consider whether ridge regression (also available in glmnet) might be a better choice than LASSO. The specific predictors selected by LASSO are likely to change drastically among samples, as you can see by repeating your LASSO modeling on multiple bootstrap samples from your data and noting the differences among the sets of predictors. Ridge will include all of your predictors but with the magnitudes of their coefficients reduced to avoid overfitting. If you do need to do predictor selection with LASSO, recognize from the beginning the necessarily arbitrary choices it will make from a set of correlated predictors.
There are some packages in R that might help with this process: BootValidation (which seems to extend the validation procedures of the rms package to glmnet models) and c060 (which provides "functions to perform stability selection, model validation and parameter tuning for glmnet models"). I don't have direct experience with them, however. The fall-back is to take advantage of the boot package in R to automate the bootstrapping, model building, and validation tests.