In SBRT for lung tumors, does Monte Carlo dose calculation algorithm versus AAA dose calculation give a better representation of lung target dose and normal tissue dose?

I am not a physicist, and they are much better suited to answering this question than I am, but I like physics and I'm willing to take a stab at it.

Monte Carlo is the gold standard for dose calculation.  There is nothing better aside from other Monte Carlo alogrithms.  When I first heard the terms I assumed it was a radiation oncology specific dose calc algorithm that was described by a physicist while living in Monte Carlo (good gig, Mr/Ms Physicist).  Not so.  Monte Carlo simulation is a computational method that samples the probability of an event many times and then creates a plot of predicted results.  It was originally described while working on the atomic bomb in the 40's (wiki page: http://en.wikipedia.org/wiki/Monte_Carlo_method).  It's used for a lot of different applicaitons/fields like engineering, video games, wall street stock predictions, etc.  In RT it takes the path of a photon and simulates the interaction with tissue and gives probability of interactions that sum into a dose.  It also considers all the interactions that can come in the path of that photon a well (secondary electrons, etc).  All that computation provides the robustness of Monte Carlo, but also is the limitation of MC...it takes a long time to calc dose.  Other algorithms, like AAA, are based on MC, but simplified for quicker calc.  Many times they have a "MC kernel", but uses other algorithms for the rest.  Most of that is beyond my knowledge base.  I found a great talk on the topic online: http://medicaldosimetry.org/pub/397f61d4-2354-d714-5172-744e9b92e649  Funny, thorough, and discusses your question specifically.  Summary: MC is the best, but the calculation times take so long that it is impractical in a Rad Onc clinic so we use simplier calc algorithms that are within 1-2% of MC and quicker.

In the future cars will drive themselves, robots will clean our houses, and radiation oncologists will have MC dose calcs done every day for daily treatments.  We will image the patients, MC will calc a dose of the day, adjust the plan to fit our prescriptions and constraints, and IMRT will deliver it.  It will be great for patients, and I don't think that is really far from becoming a reality (ie you'll have it before you have Rosie from the Jetsons).