How do you best communicate SBRT planning directives, specifically in regard to how "hot" the plan should be?

This is where common language becomes uncommonly important. There are two languages one can speak re: doses in external beam radiation therapy. In language one, the 100% line always equals the MD's prescribed gray* dose. In language two, the 100% equals the maximum gray dose in the plan. Most of my day is spent speaking the former language; in SRS and (some) SBRT cases though, it's easier to speak language two. Why this is, is that in general for SRS and SBRT cases, the treatment volumes are smaller and/or nearer more critical structures (although clearly in lung SBRT sensitive propinquities are not going to always be an issue) so conformality and dose fall-off becomes a priority. A mathematical example of how one can be speaking these two different languages but saying the same thing is in a cranial SRS case when treating a 1 cc brain met. Let's say that the 20 Gy isovolume just perfectly covers the lesion and that ~1 cm away from the surface of the lesion in all directions lies the 10 Gy isovolume. At the very core of the lesion lies a 30 Gy voxel of dose. One can say, I'm giving 100% of the dose to 100% of the volume with a 150% "hot spot." Or, one can say I'm giving 20 Gy to the lesion at the 66.7% isodose line (66.7% = 20 Gy/30 Gy). The hot spot, H, is always related to the isodose line, D, by H=1/D. So one knows that if you're giving 20 Gy at the 66.7%, the hot spot equals 1/0.667=1.5, or 150%, or 1.5*20Gy = 30Gy.

If one says "For lung tumors I typical prescribe 100% of the dose to 95% of the PTV volume"... and... "I will either accept 90% coverage by the prescription dose or prescribe to an isodose line (i.e. 90%)," I do not have a good idea what the absolute Gy dose will be to the lesion because the languages, and even the prescription, are conflicting. If the prescription is: 20 Gy to 95% of the PTV, I expect the 20 Gy isovolume to cover 95% of the PTV. If the 20 Gy isovolume covers 90% of the PTV, then the prescription is de facto 20 Gy to 90% of the PTV. If you're speaking the second language I mentioned above and PRESCRIBE TO AN ISODOSE PERCENT, then the 90% *becomes* the 20 Gy (i.e. the 100% is no longer the 20 Gy, it's 1/0.9*20 = 22.2 Gy), and now maybe 100% of the 20 Gy dose would cover 100% of the treatment volume... maybe. That's what I would assume if you prescribe to an isodose line *in percent* and the isodose line *in percent* covers the target. (This becomes a HUGE distinction in Eclipse treatment planning software for example. In Eclipse, if you normalize to a percent, the 100% *always* equals the absolute Gy prescription dose. If you prescribe to a percent, the 100% *never* equals the absolute Gy prescription dose. If you do both, normalize and prescribe to a percent, it's doubly confusing. Normalization and prescribing to a percent are mathematically equivalent, but graphically what they do to isodose % lines in Eclipse are different; yet they do the same thing to absolute Gy doses and beam MUs etc.) Again, looking at languages for doses, if you say "...I will usually prescribe to an isodose line of around 80%... don’t be surprised to see a 20-30% hot spot." I can only assume that the 80% equals the absolute prescription dose in grays, thus the 100% would be 25% hotter than that, and if you had 120-130% isodoses in the plan, you *really* are talking about doses 130/80 = 1.625 = ~62.5% hotter than the prescribed dose. Is that what you're saying? It's got to be, becuase if you prescribe to the 80% and max is 100% in the plan, you can't have a 30% hot spot... only a 1/0.8 = 1.25 = 25% hot spot.

Percent isodoses always correlate to absolute isodoses, and this correlation is always mathematically direct but differs depending on which language you're speaking. Again... it's important to speak the same language the same way throughout the case. To the original question, my prescription in essentially every case that involves a PTV is an absolute dose prescription. Whether I'm planning or in concert with my dosimetrist, we know the Rx is for most of the PTV to get the absolute Rx dose (or more). By definition in cranial SRS, the Rx is the minimal lesional dose in grays. For other IMRT and SBRT cases, I too like to see 95% of the PTV getting the absolute Rx dose or more. I prescribe to isodose lines in SRS cases (i.e. I speak the second language above where 100% always equals the plan absolute max dose and there are no percent isodose lines greater than 100); in other IMRT/Rapidarc/SBRT cases, I prescribe an absolute dose to a percent (usually 95%) of the planning target volume, and there will almost always be >100% isodose lines in those cases, graphically speaking. AAPM TG-101, "SBRT," listed these metrics for planning and reporting SBRT:

• Prescription dose
• Prescription ICRU reference point or dose/volume e.g., isodose covering PTV to a particular percentage
• Target coverage
• Plan conformity (example: Ratio of prescription isodose volume to PTV or a conformity index such as proposed by Hazard et al. IJROBP 73:562–570, 2009.)
• Dose falloff outside the target (example: Ratio of the volume of the 50% of prescription isodose curve to PTV)
• Heterogeneity index e.g., the ratio of highest dose received by 5% of PTV to lowest dose received by 95% of PTV
• Notable areas of high or low dose outside of the PTV

And finally regarding things "hot" and "hot spots," I'm not even speaking the proper language here myself. The ICRU doesn't describe "hot spots" if they're inside the treatment volume; they're extra-treatment-volume entities. Here is an excerpt from ICRU-50, "Prescribing, Recording, and Reporting Photon Beam Therapy":

• Treated Volume: Volume enclosed by an isodose surface (e.g. 95% isodose), selected and specified by radiation oncologist as being appropriate to achieve the purpose of treatment. Ideally, Treated Volume would be identical to PTV, but may also be considerably larger than PTV
• Maximum Dose: Highest dose in PTV. A volume is considered clinically meaningful if its minimum diameter exceeds 15 mm; however, if it occurs in a small organ (e.g. the eye, optic nerve, larynx), a dimension smaller than 15 mm has to be considered
• Minimum Dose: Lowest dose in PTV. In contrast to Maximum Dose, no volume limit is recommended.
• Hot Spots: Volume outside the PTV which receives dose larger than 100% of the specified PTV dose. In general considered significant only if minimum diameter exceeds 15 mm; however, if it occurs in a small organ (e.g. the eye, optic nerve, larynx), a dimension smaller than 15 mm has to be considered.

*(never capitalize "gray" i.e. J/kg)

One of the most important concepts that people need to understand when planning SBRT is the tug-of-war going on between achieving optimal conformality and dose homogeneity. Most planning algorithms struggle to achieve both effectively. My top priority is conformality (close to 1 as possible) so I really do not pay as much attention to hot spots within the target volumes. Conformality is a complex issue so I’ll refer people a Red Journal manscript “Conformality Index: A Review” Feuvret Loic, et al, 2006. The other factor I pay attention to, especially in lung treatment, is the R50. This lets you assess low dose spillage outside of your target volumes. Most of the RTOG SBRT protocols provide good references based on PTV volume for both peripheral and central tumors.

For lung tumors I typical prescribe 100% of the dose to 95% of the PTV volume. I will alter this based on location of the tumor to critical structures (ie bronchus). Typically then I will either accept 90% coverage by the prescription dose or prescribe to an isodose line (ie 90%), which means the plan gets hotter. For spine plans I will usually prescribe to an isodose line of around 80%. This allows for you to stay off of the spinal cord, knowing the plan will get hotter within the PTV. When using these techniques, don’t be surprised to see a 20-30% hot spot. As long as this occurs within the PTV, there will usually be no issues. This speaks to the necessary of adequate immobilization to minimize setup error so your plan heterogeneity does not lead to dose outside the target volume.