Misleading information about proton beam therapy?

There are good data suggesting a significant potential value of proton beam radiotherapy (PBRT) in the treatment of prostate and other forms of cancer. It is therefore unwise of its advocates to “over sell” this treatment by publicizing outdated and potentially misleading information as if it were current.

An ASTRO media release distributed last Monday carries the heading “Proton Therapy Lowers Chance of Later Cancers” and reports that, “Patients who are treated with” PBRT have only half the risk of developing a secondary cancer, compared to those who are treated with standard photon radiation treatment. However, there is a major fly in this bottle of ointment.

The actual study compared data from 503 patients treated at the Harvard University Cyclotron Laboratory before it was closed in 2002 to 1,591 patients treated with photon radiation therapy who were identified using the SEER database and and who were treated between 1974 and 2001! So for starters all the data under discussion is based on treatments carried out a minimum of 7 years ago.

The claim in the heading of the press release and the introductory text of the release are not factually justified by the data actually reviewed. The treatments compared in this study predate most of the current improvements in the use of proton and photon radiation, including 3D-CRT, IMRT, etc. Just as importantly, we do not know whether the patients treated were strictly comparable because of the retrospective nature of the comparison. A justifiable claim would have been that patients who were treated with PBRT prior to 2002 may have a significantly lower incidence of secondary cancers than those treated with photon beam radiation over a similar timeframe.

This may have been an intellectually interesting analysis to carry out. It supports the hypothetical concept that PBRT may have a lower risk for development of secondary cancers than standard forms of photon beam radiotherapy. However, it did not deserve a media release that is potentially misleading to patients who may not be aware of the significant changes in the delivery of radiation therapy over the past 30 years.

To be fair to the study’s authors, the release does carry a quotation from the lead author that states, “Since this is a retrospective study, however, we will need additional studies to further prove this hypothesis.” The problem is that this is all they actually have, an hypothesis, which ASTRO unfortunately  packaged as a fact.

We should point out clearly that the original Harvard Cyclotron facility has now been replaced by the recently completed Francis H. Burr Proton Therapy Center at Massachusetts General Hospital.

9 Responses

  1. The recent advancements in radiation therapy, most notably IMRT as you mentioned, would be expected to INCREASE the rate of second cancers versus 3-D conformal RT, not decrease it. In addition, the proton system used at the old Harvard Cyclotron Lab was much more primitive than what any center in the US uses today, including Harvard themselves. One of the many advances in the current proton systems used today is placing the proton producing accelerators further away from the patient, which significantly reduces the amount of neutrons which is a suspected carcinogen produced in both XR machines and proton machines, especially the old proton machines.

    Thus, the radiation oncology community would actually view the abstract presented at the annual meeitng as the “worst case scenario”, since current XR machines are likely to be at least as, if not more, carcinogenic while current proton machines are likely to be less carcinogenic. The comment just made regarding current proton systems may or may not apply to single room proton systems, however.

  2. Thank you very much for this additional comment, which would seem likely to confuse the average patient even further.

    Are you able to refer us to published data on the relative risks of secondary cancers based on older and newer forms of photon-based radiotherapy? We would be surprised if most patients are aware that newer forms of external beam radiotherapy came with a higher risk for secondary cancers than older forms of such therapy.

    On September 4, we noted a report by Rapiti et al. from the Geneva Cancer Registry that documented a significant increase in risk for colon but not rectum cancer after earlier external beam radiotherapy (EBRT) for prostate cancer. However, this report was based on patients treated in the time period 1980-1998.

  3. The report is interesting and encouraging with regard to proton beam therapy. A large part of its theoretical advantage is the low integral dose to surrounding tissues, which would predict fewer secondary cancers. Issues about neutrons brought that into question.

    Here’s my question about the report, though. What kind of cancers were treated in the groups retrospectively examined? Prostate? Maybe not. The releases about the report don’t say!

  4. Dear Dr. Klein: Perhaps you could e-mail Dr. Tarbell, the senior author of the study at Harvard, and see if you could get a copy of the actual poster.

    Most of the media releases issued by ASTRO during the course of the recent meeting were distinctly “skimpy” in their content (IMHO), making it difficult for anyone not attending the meeting to get a sound understanding of what the scientific presentations did and didn’t really state.

  5. I did it. We’ll see what I get.
    Dr. Eric Hall was the guy who published concerns about excess cancer-promoting neutrons in both IMRT and standard garden-variety conformal proton beam therapy. The latter has been refuted, I should say at least to the point where the magnitude had probably been overstated. About IMRT, I think perhaps Hall’s observations still stand and I assume that is what Florida RT Doc is referring to.
    As for protons, intensity-modulated proton therapy should reduce whatever cancer-inducing neutron burden exists even further, and, overall, I am very hopeful regarding this development. Whatever the Tarbell study can be said to show, the theoretical advantage regarding secondary cancers has always seemed to me to be one of the most cogent points in favor of proton beam.
    It is surely hard for anyone to know what to make of the Tarbell study without some more details. I would make a small bet that the cases are pediatric and neurological and that the entrance doses for the protons were far less than with prostate (exit doses being neglible).

  6. Here is the Hall reference about IMRT (and protons):

    Dr. Tarbell replied (instantly!) that “we should wait until data is submitted for publication.” This did not surprise me.

  7. OK. I pushed, specifically asking if there were prostate cancer patients in the study, and I got this reply:

    “It did. Prostate, brain tumors, head and neck cancers, base of skull (chordomas) and sarcomas (near spinal cord and others). It will be of interest to break this out as Christine finalizes her paper(s). I have not specifically asked for the prostate cancer patients and their outcome. Since many prostate patients are older, I doubt they contributed significantly to these findings– but, if this (unpublished) data is being used this way, all the more important that we finalize this manuscript.”

  8. Thanks so much for your efforts, Dr. Klein. I think we are all in agreement that knowing exactly which categories of patients these data apply to is critical. Let us hope that Dr. Tarbell can get an appropriate paper published quickly in a suitable journal!

  9. Click here to see more on this, including a video if you can access it. (If you are not a member of OncologySTAT you will need to sign up, but it is free.)

    Median age at treatment was 62, follow-up was about 6 years. Dr. Anthony Zietman’s critique is included. What amazes me is simply the large total number of cancers (6.4% and 13.1%) I don’t see the question addressed of how many second (not secondary) cancers would occur with no radiation. I recall a figure of 1.7% for secondary cancers in 10 years after radiation for prostate cancer. I’m confused, but I don’t see that the study was particularly biased in favor of protons (in agreement with Florida RT doc), so there is some reassurance about protons here. Again, the whole paper must be awaited and would need to be critically analyzed.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: