how to calculate the effective lung volume to calculate the V20 for the lungs. do we have to deduct PTV or the GTV from the total lung volume to know the effective lung volume which will be used to calculate the V20/ V15 etc. kindly provide related evidence.
In clinical Practice, we use the the terminology COMBINED LUNG VOLUME (V20)
This refers to the percentage of both lungs with subtraction of overlapping CTV receiving no more than 20 Gy.
As opposed to the other terminology, TOTAL LUNG VOLUME which is used less frequently which simple refers
percentage of both lungs receiving no more than 20 Gy.
Reference: http://www.nccn.org/professionals/physician_gls/PDF/nscl.pdf
Can you explain to the rationale behind the "Combined Lung Volume"?
Who thought it up?
Apart from artificially lowering the V20, what other purpose does it have?
Are their any physiological studies to show that CTV lung doesn't function for O2 exchange and doesn't contribute to pneumonitis risk?
And what happens if my GTV>CTV margin is 0.3cm, or 0.5cm, or 0.7cm?
How can oncologist's idiosyncracies alter a lung volume?
I volume the whole lung by autocontour. I use 0,800 Hounsfield threshold in Pinnacle. All the lung contributes to pneumonitis risk whether at risk or not.
When marking a GTV, I use the same technique to draw the air-tissue boundary. This method is ABSOLUTELY reproducible - every time the same, guaranteed!
NCCN document requires login, can you download it under files please.
I agree with AAM - i do exactly the same way as him. The logic is that only the GTV will not have the functional capacity as already there is disease. But the CTV and the PTV lung are not only variable but also have a functional lung which will receive the dose and are at risk of pneumonitis.
The concept of: Effective lung volume = total lung volume - GTV ; is the most correct one (as i think) - this is mentioned in NSCLC chapter of perez & brady Vth edition - reproduced from MD Anderson. Some of the RTOG studies have started deducting PTV from the total lung volumes - but i cant understand the logic - this is just a way of artificially making the V20 low.
Anyone having any latest update on this please post.
We all know that the volume of lung irradiation has a direct linear corelation with the incidence of rdaition pneumonitis.
There are various predictive factors for a clinician like like V20, V30, mean lung dose (MLD), and NTCP (normal tissue complication probability models)
V20 is one of the simplest to follow and easy to use which is the favourite for Thoracic radiation Oncologists.
I m afraid just like Emamis tolerance doses which many of you criticise, the V20 comes from the same group of Emami, James Purdy, Perez, with the main author being Graham MV
http://www.ncbi.nlm.nih.gov/pubmed/10487552
The incidence of Gr 2 or worse penumonitis is as follows
V20 < 22% 0%
V20 22-31% 8%
V20 32-40% 13%
V20 >41% 19%
As a rule of thumb most oncologists would accept upto 35% as their higher threshold of V20 as in that study all fatal pneumonitis occured when the V20 ws >35%
I agree its indeed very confusing to find so many studies where V20 has changed over the years.
Some edit the GTV, CTV or PTV while some dont subtract the volume at all and there seems to be no universal standard here.
I think your logic of subtracting GTV as its the non-functioning component makes the most sense of all.
And i agree with you, over the years the margins we give have become more tight, hence its difficult to balance so many variables and correlating this with published data.
At teh end, its kinda simple logic again
If you underestimate the lung volume, your threhold for v20 can be laxed upto perhaps 40%
viceversa, if you overestimate these volumes, the threshold for V20 will have to be more tight, perhaps 30%
So, i m afraid i cant dictate your oncology practice here, and you are free to use whatever you think is appropriate
if you understand these principles and modify your v20 thresholds accordingly.
At teh end, its kinda simple logic again
If you underestimate the lung volume, your threhold for v20 can be laxed upto perhaps 40%; viceversa, if you overestimate these volumes, the threshold for V20 will have to be more tight, perhaps 30%
So, i m afraid i cant dictate your oncology practice here, and you are free to use whatever you think is appropriate if you understand these principles and modify your v20 thresholds accordingly.
That's the point of my questions - it is NOT simple logic applied here.
The ONLY way that I can get an accurate view of potential lung toxicity according to Graham or Emami, or Jackson for that matter, is to volume LUNG, and LUNG only. That organ has a particular density as seen on Hounsefield units, which is why it is so easy. I don't volume [Lung - GTV], I just volume lung in a reproducible manner. Reproducible for me from day to day, AND more importantly, reproducible for you and every other oncologist on the planet. SIMPLE LOGIC dictates that we use the simple reproducible method.
Under and overestimation of V20 is unnecessary, unscientific and useless for pooling data in the long rum - not much logic there.
While oncologists can practice and markup what we like, it should not be dressed up in acceptability by shrugging the shoulders and saying "you can do what you like". We are engaged in the business of treating a patient's anatomy, so we should start using the tools at our disposal properly.
My wife says that "I'm giving some one a bashing". I apologise but we can do this voluming thing well and get data for 10,000x more patients if we are consistent, so we should.
You are right here andrew but at the same time Rohit is telling what is being done.. so he is the messenger here so as to speak. That said the concept of excluding the lung from PTV and reporting the V20 for that seems rather lamebrain to me too. But variability in countouring/voluming is a necessary consequence of our reliance on inadequately reported radiotherapy results / techniques and personal ego … two things that are essentially "reporducible"
yes, I am well aware that the individual does what they want, but that is no substitute for proper logic and proper use.
In other areas like H&N the voluming is much much more difficult, but in lung the ability to volume lung is simple and built into the software. The dosimetrist can do it easily. So there is no excuse for variability in lung dose-volume relationships as the lung can be defined extremely accurately and reproducibly.
The actuality is not
"variability in countouring/voluming is a necessary consequence of our reliance on inadequately reported radiotherapy results / techniques and personal ego"
but rather that
inadequately reported radiotherapy results are a necessary consequence of our reliance on variability in countouring/voluming/techniques and personal ego.
i communicated with Dr Joe Y Chang - head of thoracic radiotherapy unit - MD Anderson and i reproduce his words :
"Most papers use Total lung minus GTV. Ideally, however, it should be Total lung minus PTV since lung within PTV will be damaged by RT anyway."
so now, what to say guys ? im confused with all this stuff LOL.
In head and neck IMRT PTV overlaps some part of parotid gland but we calculate dose for the whole parotid gland & try to keep mean dose < 26Gy ideally or V30 of one gland to be <50%. then why we exclude PTV in lung.
Nice one Abhinav, can you ask him the rationale behind this citing this doubt of yours.. so that he can give a more detailed answer. It is appreciated that the lung within the PTV will be damaged but is that volume not of consequence when the morbidity from pneumonitis is concerned?
Yes Santam, i sent him another mail asking these doubts lets see what he has to say.
I never expected V20 business would ever get so serious.
I am now beginning to think whether does it make a huge difference at the end of the day, if i subtract
1. GTV from the total lung volume and have my standard thresholds for v20
2. CTV from the total lung volume and have another set of standard thresholds for V20
3. PTV from the total lung volume and have a third set of standard threshold for V20
4. OR NOTHING AND use the most simplest of techniques as described by Andrew.
If any of you can tell me what are the incidence of Gr 1, 2, 3 or 4 pneumonitis correlating with your V20
All of you will then understand the working of THE MATRIX and will have the vision of Keanu Reeves.
To me all seems to correlate with one another as long as the total lung volume remains in the equation.
It simply is the way you look at it.
Dont get me wrong, but there does not seem to be a clinical need for pooling 10,000 pts for this. Waste of time and waste of resources.
At the end of your conclusion a few decades from now, even if it were published, i would doubt how many would follow those guidelines as you would always have oncologists criticising in one way or the other.
An international consensus would simply be achieved only if RTOG, DAHANCA, EORTC, etc met at a simple meeting and reached a consensus on this, thats the only way u and i would do the same thing ever.
Over and out
Rohit !
Dear Friends i received the answer from Prof Joe Y chang from M D Anderson hospital im copy and psting the final line of the mail:
"If you ask me what you should use for clinical practice using published dose volume constraints, I would say total lung minus GTV"
he is the one who wrote the chapter of NSCLC in perez & brady Vth edition ; so i think my puzzle is solved.
I think you misunderstand something about the lung definition I use.
I use lung Hounsefield values to recognise lung tissue, thus the GTV IS NEVER INCLUDED. I volume the lung, the whole lung and nothing but the lung (so #1 above is what I do …. with the proviso below).
My point relates to GTV definition rather than lung definition, but both are based on the same thing.
Where the GTV interfaces with the lung, THE METHOD THAT DEFINES THE INTERFACE ACCURATELY ACTUALLY DEFINES BOTH THE LUNG AND GTV INTERFACE CORRECTLY.
When I see the term [Lung - GTV], I think it usually relates to a hand drawn GTV, this is inaccurate. The published data shows that these differences are substantial. If the Lung and GTV are constructed by autocontour, then it is accurate.
I use an autocontour method to define the GTV wherever it contacts the lung also. I will describe the details below.
When voluming the GTV, I make a new ROI called "GTV" and set the autothreshold to 0,800 as I did for the lung autothreshold. On a slice where the tumour is visible, I use the autocontour to mark the tumour.
- If the tumour is an island within the lung, then the lung contour will already look like an annulus, and my GTV volume will correspond to the internal island (or 'donut hole') that matches the lung border. But the point sthat I make is that it matches exactly, and whether done in Wollongong or MDACC or PMH or Chandighar, it will be exactly the same because it is based on Hounsefield units.
- If the tumour is part of the lung outline, i.e., pushing in to/out from the mediastinum, then the lung outline will look like a 'dish' with no hole. The GTV when autocontoured then looks just like the lung contour, and then I use the drawing tools to draw the soft tissue boundary in the chest wall/mediastinum, leaving the air-tissue interface untouched. This process creates an annulus, and I then delete the outer ring.
I can pull a picture and upload a series if this is not clear (because these types of description often are clear to the author and opaque to the reader!)
My point about 10,000+ patient data is that we all do lung DVHs for radical lung treatment. If you see patients in follow up for 1 year, we then have enough data to do much more accurate DVH measures than the miniscule stuff in the literature. And what extra work is involved? Lots if your data doesn't sit in your OIS, minimal if it does. Both ARIA and MULTIACCESS/LANTIS have clinical assessments to cope with this. See here for more explanation!
Good explaination Andrew. While the method is reasonable for peripheral tumors what do you do for hilar based masses?
Dear AAM,
Thanks for the images - I contour almost the same way, by autocontouring lung using Lung HUs and the GTV accordingly.
QUANTEC is here and probably we can get the paper and throw new light about importance , merits , demerits of tolerance, volume etc.
QUANTEC says to use the GTV too as the use of PTV can lead to variability due to inter-institutional variations in PTV margin selection. Other major issues highlighted by the paper are:
P.S. Andrew thanks for the nice explanation.. I was fuzzy about the step 3 onwards as that cant be done by autosegementation and as you have rightly said will be variable
K , now i understand what you mean.
Those pics are great.easy and self explanatory.
This method surely looks great. It probably inherently does not concentrate on the gtv, and gets the functioning lung volume as per hounslow units.
In the UK, strangely believe it or not an oncologist only draws the tumour volumes. All critical structures are marked by physicists or dosimetrists. You are of course quite privileged to draw these critical volumes.
Thanks
Dear Collegues,
Agreed of logic of using Lung-GTV.
I always wondered what really happens in vivo? when the patient is having inspiration does the lung dose averages out.
What happen’s in the later part of treatment when consolidation resolves and aeriation improves?
Is upper zone radiation less worse than lower zone RT(more vascularized) which is functionally more important?
How things should be differant in COPD or previous TB patient and role of baseline PFT?
These patients are often smokers and their baseline function is poor but nobody does PFT prior and than blame goes to RT for complication.
And now the most important doubt. Many people use 5-7 beams for the thorasic IMRT which increases the low dose volumes more and lung being a parallel organ which is not desirable. I have an argument with some of physicists who for the sake of conformity put extra beams and increases the dose otherwise untouched lung. What’s the take of the house on these issues.
You are right sandeep, in vivo its gonna be difficult to know what really happens. Your guess would be as good as one who would do a lenghty thesis and phd on it.
The part of the GTV or CTV which is gonna receive the brunt of radiotherapy —- are you concerned about that part of lung function improving on treatment. Its just a small cost you pay for treatment. My simple advice would be mark it, treat it and forget about it.
Regards your view on nobody does PFT prior, i think this is " GROSS " and " Poor Medical Practice "
You, the radiation oncologist is responsible for RT Pneumonitis especially if this has been inappropriately offered to a patient who looks clinically well but on lung function test has an FEV1 = 0.6 litres ( oh i remember this figure from a previous very lengthy discussion on Isocentre… my God no more discussions on this please)
It is mandatory to know your Lung function Tests esp FEV1 and transfer factors before you consider pts for Radical RT.
There have been medicolegal cases from patients relatives suing the Hospital for death of patients from RT pneumonitis where pts were treated with Radical RT and on case reviews it was found , that no lung function tests were done.
Another compulsory thing here in UK is to do PET scans for any pts being considered for RADICAL Lung treatment.
15-20% of the times the damn PET scan would change the decision from Radical RT to high dose Pall RT because of distant mets.
Your concern about upper zone and lower zone can be objectively assessed only by Ventilation perfusion scan, but we dont use that in normal clinical practice.
We simply subjectively assess that on clinical grounds.
Thus in general pts with FEV1 = 1.5 litres being the usual cut-off at the lower end of limit for lower zone tumours.
However, if there was a pt with a small peripheral upper zone T1a or T1b Lung cancer (medically inoperable), we may clinically accept treating such pts even if the FEV1 was 1.4 or 1.3 litres. These are of course clinical decisions and have to be taken on a case to case basis depending on its merit, explaining to the patients about the risks involved.
The UK view of treating lung cancer is a simple 3 or 4 beam conformal technique where we want realistic radical doses
1. Conventional: 60-64 Gy / 30-32 # / 6 - 6.5 weeks
2. CHART: 54 Gy / 30 # / 12 days
3. Hypofractionated: 55 Gy / 20 # / 4 weeks
If in the context of clinical trial there are plans for dose escalation, then we use IMRT
At the end of the day its clinical judgement and experience which determines with regards to when to treat, what to mark, what plans to accept , what V20 levels to accept and what risks to accept.
A good and wise oncologist will know his or her boundaries and know when to say " NO " as a recommendation or suggestion to a patient as opposed to tell the patient to take your own risk.
Risk taking comes in picture when patients challenge you, question your decision, are desperate for treatment at which point you carefully consent them informing them about the benefits and risks involved.
Over & Out
Rohit !
