Discussion with insurance company

[deleted4401]

I had an interesting discussion with an insurance company physician yesterday. The insurer is a large, multi-state medical system that also provides it's own insurance, providers, and hospitals - an HMO. The physician I spoke to was not some Caribbean school flunky that sold their soul to the insurance company. Rather, it was a radiation oncologist that was a service provider for the HMO that was trained within the last 10-15 years at what people on this forum consider to be a top 5 program. This insurer tends to be very strict on what they allow/deny, yet I don't find myself arguing with them often, because they are evidence-based and very reasonable.

I had recommended IMRT to a patient with breast cancer. It was right sided, so it was initially denied. The reason for denial was that there were no normal structures to be avoided. There were two problems with this denial. One, I don't recommend IMRT for breast cancer often and when I do, their is usually a legitimate reason for it, so I get pissed when denied. Two, this insurance company (and most others) have a completely mistaken concept of what the role/relevance of IMRT for breast cancer is, and I had to call them on it.

This is basically the conversation I had with her, and it was one of the first times we've been able to reverse their decisions.

She tried to go down the path of critical structures and ding me for that, but I made it very clear my reasoning had nothing to do with that. The majority of the time I recommend forward planned IMRT is when I have a large breasted woman that I cannot bring the hot spot down to a manageable level (we don't have a prone board, yet). The two randomized trials (Pignol and Donovan) comparing IMRT and 2D/wedged conventional treatment don't make mention of critical structures. The whole goal of IMRT in breast cancer has nothing to do with reducing dose to heart/lung. It has everything to do with increasing homogeneity of dose - i.e. decreasing the Dmax/hot spots, decreasing the V105%, etc. I told her I had tried traditional methods and could not get the hot spot down with wedging or 3D iterative techniques. Based on level one evidence, forward planned IMRT would allow me to statistically decrease both acute and late toxicity. She ended up reversing it, and said that they would approve this, and in the future that I should go ahead and simulate/plan these patients before approval, but I would need to send: 1) central axis CT slice showing the separation was large (>30cm in this case) 2) a summary of what I just said above 3) DVH showing the Dmax and V105%. So, if you are trying to get approval with a major insurer for breast - try that approach - it worked on these guys and they are tough.

Anyway, it was fun to win. Now, whether we should get reimbursed as high as we do for that, completely different story. But, it's still a completely different technique and considerably more labor intensive than iterative 3D/FiF or conventional 2D with wedging.

-S

---

Members don't see this ad.

 

[academia]

Nicely done Simul and thanks so much for sharing!

I feel like this should be published in some Rad Onc business/practice journal! Perhaps the ASTRO Blog might like to see this as a means for disseminating some very useful info?

 

[Pewl]

Strong work!

Quick question, though, if she is from a "Top 5" rad onc program why was she not aware of the IMRT breast study you mentioned!

 

[deleted4401]

Oh no - that was the point - she was well aware of the study, but they know that most people that call in try to get it for L breast to decrease to heart and lung. And sometimes they actually approve for that (which is rather silly, b/c you can keep dose low to those structures using traditional techniques or using breathing control/gating or a prone board).

When I focused on those trials, she asked if I was using 2D and a wedge like the control arms on those trials, I told her I was not, but that even with a CT and wedging, the concept is the same, that the hot spot couldn't be reduced. Then she said they would approve it for that indication.

-S

 

[qwert]

Simul, what are using now for motion control for breast IMRT?
If nothing, using IMRT constitutes fraud and IMHO malpractice.

 

[deleted4401]

Qwert, I'm not sure if you are understanding how typical breast IMRT works. It is nothing like completely inverse planned, multi-field treatments like for prostate or head and neck cancer. It is not multi coplanar beams. I think you should watch what you say before you accuse me and other practitioners of malpractice and fraud. That is a very serious thing to say to another physician and acquaintance.

I'm not going to describe the entire process to you, but it is a series of tangential beams that "blurs" out hot spots. The minimal motion that is involved actually helps smooth out the hot spots. There is an excellent paper on it that describes the technique, and when I read it, I realized I did not understand how IMRT is delivered for breast cancer, and I'm not sure you do, because you would understand that it isn't malpractice at all. Even if the charge was the same as 2D, I'd prefer to do it on all patients - it's worth the extra time to me, but at this point it's not and I make it a point to use medical dollars wisely.

The basics is that you set up tangents as you would for 2D breast. You open the beam and then use an isodose line to delineate the breast tissue target. We use the 85-90% isodose line and turn that into a contour. To be thorough, you would then modify the volume - carving out chest wall, muscle, rib, lung, axilla that you aren't interested in treating. This becomes your target. Some people add a margin to this to make it PTV, others use this as the PTV without a margin. I'm unsure of whether this is relevant, b/c the traditional field design that this is based on would cover a similar volume of tissue, so I don't add. Then you look at the various hot spots, and use smaller inverse planned tangential segments to smooth it out. 80% of the dose is delivered through traditional tangents. 20% of the dose is delivered through optimized "segments". These segments are just smaller, modulated tangents that balance out the dose creating a homogeneous dose distribution. A basic plan would include the 2 medial and lateral fields, and then 2-4 (or more) tangent segments. To say that this is malpractice is to say that tangent beam radiation is malpractice.

Purists would say that this isn't really IMRT, and I can't say I disagree. I think some experts (see Beryl McCormick's review in Seminars in Radiation Oncology this month) are grouping this and multi-coplanar techniques as "more conformal techniques" to distinguish between 2D/wedges.

The purely inverse planned techniques are useless. They are able to be much more conformal (reducing dose to heart, lung, contra breast), but the hotspots are ridiculous (130-140%), thereby making it worthless in this scenario. The multi-field techniques (adding anterior oblique fields to tangents) are a little bit better with homogeneity, but terrible in terms of dose to other structures (high V20 for lung, high heart dose).

Read this article that I've attached, read the actual randomized trials, and then you will understand why motion control doesn't make sense for what you are talking about, and why this technique is not fradulent or malpractice. I think JCO and the Green Journal typically refrains from publishing randomized controlled trials proving the superiority of a fradulent technique. Motion control benefits you in breast cancer when you are doing something like active breathing control to get the heart out of the field, not when decreasing inhomogeneity.

S

 

[Gfunk6]

Excellent post Simul.

Another way the technique is described is "field within a field" which are smaller "segments" that the dosimetrists manually add to the opposed tangents to iron out dose inhomogeneities.

IMRT is technically both aperture and intensity modulation. IMRT for the breast is really more focused on the former. If you have a woman with really large breasts, inverse-planned aperture modulation may be the best way to compute solutions rather than the hours of labor it would require for a dosimetrist to do it manually.

 

[qwert]

I understand forward-planned breast perfectly well. Three important points.
a) If one indeed follows accepted guidelines and administers 80% of dose through open tangents, it is probably safe from standpoint of missing tumor and committing malpractice.
b) However: improved dose distribution seen in planning does not correspond to dose that is actually delivered. To support this point, data is plentiful.

Song T, Suh CO, Lee I, Jeong K, Keum K, Lee CG, Seong J, Cho JH. The effect of
respiratory motion on forward intensity modulated radiotherapy for breast cancer.
Technol Cancer Res Treat. 2008 Jun;7(3):207-15. PubMed PMID: 18473492.

Liu Q, McDermott P, Burmeister J. Effect of respiratory motion on the delivery
of breast radiotherapy using SMLC intensity modulation. Med Phys. 2007
Jan;34(1):347-51.

c) Did anybody here actually read Pignol study?? It's a disaster in every regard. His stat. analysis is garbage. Donovan is irrelevant because he used compensators not MLC.

 

[medgator]

Another way the technique is described is "field within a field" which are smaller "segments" that the dosimetrists manually add to the opposed tangents to iron out dose inhomogeneities.

That's my understanding of what "IMRT" was in those studies. I use FIF quite often to reach homogeneity in the breast.

 

[radiaterMike]

Just to add fuel to the fire ...

There is absolutely no necessity to use "motion control" with IMRT for any disease site. Sure it can help augment the benefit of IMRT, but they are separate techniques. IMRT is simply a way to modulate the beam to shape the dose contours. IMRT can generate a concave dose distribution that allows you you to better spare lung regardless of chest wall motion. The other obvious benefit is homogneous dose delivery.

Sure- without "motion control" the target needs to be larger, and there is inter and intra- fraction movements which can result in day-to-day variation in dose distribution- but this should average out, and have no clinical impact on breast treatment.

I agree that it is unfounded to make accusations of malpractice.

Simul, what are using now for motion control for breast IMRT?
If nothing, using IMRT constitutes fraud and IMHO malpractice.

 

[deleted4401]

There are 2 randomized controlled trials and multiple, large, retrospective series from multiple countries with hundreds of patients indicate the same outcome - less acute and late toxicity. Yet, we are focusing on: 1) a series of 10 patients from S. Korea that raises "theoretical" (read: likely clinically insignficant) concerns while not looking at clinical outcomes and 2) a study looking at an anthromorphic phantom that raises these same "theoretical" concerns. Seriously - 10 patients and a dummy trumps thousands of patients and 2 controlled trials?

There are randomized data (with flaws - what trial doesn't have them?), loads of supportive multinstitutional data (Pittsburgh, Beaumont, Canadian group, MSKCC, Fox Chase, Emory, a large Italian series, and that's just a few that I can think of) with some looking at toxicity, some at oncologic outcomes, some at both. The NCCN guidelines suggests utilizing it (yes - they most certainly do), and I've seen outcomes with my own eyes. I think it is insulting and ridiculous to call it fraudulent or malpractice. My caveat at the end of the initial post was that I was uncertain that it should be billed at such a high rate, but it is technically more challenging and should probably be billed somewhat higher than a 2D plan prescribing to midplane.

I just don't think that Beaumont or MSKCC or the good people in Italy or I have been malevolently mistreating patients, intentionally producing poor oncologic outcomes with no decrease in toxicity for financial gain. I can't even believe I'm having this discussion. Unreal. Wait until you get out there and tell a colleague that they are fraudulent and commiting malpractice... It doesn't go over well.

And, btw, TCRT will publish a Xerox of my butt cheeks if I have meet their formatting criteria. It's in April 2008.

-S

 

[qwert]

Man, the establishment sounds wounded here No need to resort to"I see the results myself" argument. Bottom line for me, we need use "field within the field" with care, utilizing breath hold when available. Those billing it as IMRT should stop before getting the whole profession in trouble. Those using inverse-planning IMRT for intact breast should stop too.

 

[TarHeelRadOnc]

Simul- I agree with your argument for IMRT in this patient. Control arms of Pignol and Donovan were poor, but that fact does not eliminate the dosimetric benefit of IMRT in select patients. I wish we had as much success with IMRT approval with the large private insurer in my region.

One question: Are you using daily image guidance with IMRT in that patient by kV imaging of one of the tangent fields? I only ask because of the challenges with setup reproducibility in large breasted patients. I do daily kV imaging of at least one of the tangent fields in patients with large breast volume to confirm setup accuracy. Interesting the frequency and magnitute of the required shifts from standard 3-point setup. I think that this is relevant to the homogeneity issue. FIF segments with small beam appetures are used to reduce hotspots and improve homogeneity, but if setup accuracy is poor then there is considerably less return on this investment. I think that image guidance is useful in patients treated with IMRT. While it may sound blasphemous, I also think that very lightly weighted anterior oblique segments are useful for patients with large breast volume. Additional heart/lung dose is minimal, provided dose contribution from this field is small. Only so much can be done with tangential fields/segments and low/mid energy photon fields in patients with central axis separation >25cm!

Qwert- Those are serious accusations. In the real world (ie outside of the impermeable bubble of residency), those kind of statements will have you looking for a second job shortly after you take your first. That is a more important lesson than any of the academic lessons gained from this thread... Good luck to you...

 

[deleted4401]

Hmm... Does kV imaging add much more than port films? I assume we are matching to the breast contour and the ribcage? Maybe it is something to add, I'd have to see how much variance as compared to MV ports. Yes - I think in the very large breasted women, an anterior field or two may allow for some benefit, but set up gets even trickier.

Going to have to look into that.
-S

 

[TarHeelRadOnc]

Hmm... Does kV imaging add much more than port films? I assume we are matching to the breast contour and the ribcage? Maybe it is something to add, I'd have to see how much variance as compared to MV ports. Yes - I think in the very large breasted women, an anterior field or two may allow for some benefit, but set up gets even trickier.

Going to have to look into that.
-S

No, I don't think that kV adds to port films in this setting... rather, I just don't want to port daily due to the added dose associated with daily mV imaging. For patients treated with anterior segments I do a daily kV-kV isopair in addition to imaging a tangent field to assure setup accuracy...