Is SBRT more toxic than IMRT in treatment of localized prostate cancer?

A new paper, published online today in the Journal of Clinical Oncology, suggests that stereotactic body radiation therapy (SBRT, commonly delivered using the CyberKnife system) has a higher risk for genitourinary (GU) toxicity than intensity-modulated radiation therapy (IMRT).

The paper by Yu et al. is based on a retrospective analysis of data extracted from the Chronic Condition Warehouse (CCW) — a comprehensive database of 100% of Medicare fee-for-service claims for patients with specific conditions, including prostate cancer. As a consequence, the data from this study needs to be interpreted with a degree of caution, because it takes no account of things like the experience of the clinical teams using the equipment, the precise details of the patient’s diagnosis, or the precise dose levels of radiation being delivered to the individual patients.

Here is a summary of the key findings reported by Yu et al.:

  • The authors identified 53,841 study-eligible patients who received IMRT and 1,335 who received SBRT.
  • SBRT patients were more likely to be white, younger, healthier, from higher income areas, and less likely have received androgen deprivation therapy (ADT).
  • The average (median) number of IMRT treatments was 42.
  • SBRT patients had higher levels of GU toxicity than IMRT patients at all time periods.
  • At 6-months after treatment initiation,
    • 15.6 percent of SBRT patients had a claim indicative of treatment-related GU toxicity.
    • 12.6 percent of IMRT patientshad a claim indicative of treatment-related GU toxicity.
    • Odds ratio (OR) = 1.29 (favoring IMRT)
  • At 12-months after treatment,
    • 27.1 percent of SBRT patients had a claim indicative of treatment-related GU toxicity.
    • 23.2 percent of IMRT patients had a claim indicative of treatment-related GU toxicity.
    • OR = 1.23
  • At 24-months after treatment,
    • 43.9 percent of SBRT patients had a claim indicative of treatment-related GU toxicity.
    • 36.3 percent of IMRT patients had a claim indicative of GU toxicity.
    • OR = 1.38
  • Increased GU toxicity associated with SBRT appeared to have been largely due to increased urethra- and bladder-related toxicities.
  • At 6 months after treatment initiation, SBRT patients also exhibited increased gastrointestinal (GI) toxicity compared to IMRT patients.
    • 5.8 percent of SBRT patients had a claim indicative of treatment-related GI toxicity.
    • 4.1 percent of IMRT patients had a claim indicative of treatment-related GI toxicity.
    • OR = 1.42.
  • At 12 and 24 months post-treatment, there was no significant difference in the apparent rates of GI toxicities.
  • There were no other differences in non-GU and non-GI toxicities at any time point.
  • The overall toxicity rate was significantly increased for SBRT vs. IMRT at 6 months, but was not significantly different at 12 and 24 months post-treatment.
  • The average (mean) per-patient cost to Medicare for a full course of SBRT was $13,645.
  • The cost of a course of IMRT was $21,023.

In evaluating the above information, it is very important to appreciate the following:

  • Patients who receive SBRT are generally given higher daily doses of radiation (e.g., 7.5 or 7.25 Gy/day as opposed to 1.75 – 2.00 Gy/day) over shorter time frames (e.g., 1 week as opposed to 8 weeks) than patients treated by IMRT.
  • SBRT, and particularly CyberKnife radiation, has been heavily and aggressively promoted by some hospitals and clinical practices.
  • Experience with SBRT over the past 5 years is still limited, with few practices having treated more than 100 prostate cancer patients.

Yu et al. themselves note that,

Given the rapid pace of innovation and the relative novelty of prostate SBRT, SBRT technique may have improved in recent years, potentially reducing the difference in toxicity between SBRT and IMRT.

A detailed set of editorial comments on this paper, by Dr. Anthony D’Amico, are also available on line (as of today, at least) in the Journal of Clinical Oncology.

The bottom line is … Patients should be aware of this new report, and — if they are interested in treatment using SBRT — they should very definitely be asking the radiation oncologist and his/her team about the risk for an elevated incidence of GU complications, and what is being done at that specific practice to minimize these risks. They should also be asking about the practice’s overall experience in using SBRT specifically for the treatment of prostate cancer. Practices with very high experience levels of using SBRT have probably treated as many as 1,000 prostate cancer patients by now, and even they would probably state that it might take 150-300 patients to learn to do this well and with minimal risk for complications and side effects.

Because SBRT uses higher daily doses of radiation over a far shorter time frame, there may well be risk for higher levels of GU and GI side effects — especially if appropriate precautions are not being taken to minimize such risks.

10 Responses

  1. Thanks for posting this. Something to consider, for sure.

  2. While the data from the 5-year combined SBRT consortium, the 6-year Katz study, and the 2-year Georgetown study all report similar patterns — a brief fall in genitourinary and gastrointestinal EPIC scores followed by a return to baseline — this Yale analysis at first glance seems to show mounting problems out to 2 years. How are they to be reconciled? Part of the problem may be that the Medicare data only includes visits to doctors for any reason, while the toxicity data in the studies are broken down by the seriousness of the issue that occasioned the doctor visit.

    I looked at the late-term urinary morbidity by grade as reported for high dose IMRT (80 Gy) at Memorial Sloan-Kettering Cancer Center (MSKCC, see Alicikus et al.) and compared those data to the QOL results reported by Georgetown after 2 years. The data aren’t easy to find because Grade 1 side effects are seldom reported in the literature.

    Georgetown reports the following late urinary symptoms for their patients 2 years after SBRT: 26% Grade 1, 17% Grade 2, and 1% Grade 3. For those unfamiliar with this terminology, Grade 1 is mild, Grade 2 is what I would call annoying (e.g., getting up to pee more than twice a night), Grade 3 is serious, requiring medical intervention, and Grade 4 is life-threatening. For IMRT, Alicikus et al. report late-term urinary toxicities of 23% Grade 1, 9% Grade 2, and 5% Grade 3.

    So at Georgetown (SBRT), 44% of patients had a reported visit to their doctor, while at MSKCC (IMRT), 37% saw their doctor with a urinary complaint — 16% less for IMRT. Yet IMRT had a five times higher rate of serious (Grade 3) urinary complaints.

    In other words, a Medicare claim for a visit to the doctor for a problem that is resolved by a prescription for Flomax is counted the same as a visit to the ER for catheterization for urinary retention.

    I agree that there may be a learning curve effect too, as 2008-2011 was a time when many new SBRT treatment centers started operating. It may not be only learning expertise, but also learning when a patient requires referral for treatment. My question is — are radiation oncologists who are relatively new to a procedure more likely to refer their patients to a urologist with even minor complaints? Maybe this occurs in the same way that new mothers are more likely to call pediatricians when their first baby sneezes, while by the third or fourth child, they seldom would. If either of these effects occurred, we would expect to see Medicare claims per physician due to SBRT reduced over time. Perhaps in a further analysis, the authors would be able to track SBRT patient claims per physician over time.

    While the cost comparison is eye opening, I agree with D’Amico that the only way to legitimately compare oncological and QOL outcomes between two treatments is through a prospective, randomized, clinical trial (RCT). In addition to the RCTs he mentioned, I know the PACE trial in Europe (NCT01584258) will compare SBRT to conventional IMRT, and SBRT to laparoscopic radical prostatectomy.

  3. Agree with Allen above. I’ve treated 650 cases of prostate cancer over the last 5 years with robotic SBRT. Control rate for Gleason 6 and 7 (3 + 4) is > 99%; 60-70% of patients back to pretreatment urinary function at 2 weeks; almost everyone at 6 weeks. The men are enjoying 95% erectile function preservation. 0 (zero) patients have required surgery for treatment of problems related to treatment. There will be a paper out in the next year of a prospective head-to-head comparison between IMRT and SBRT.

    Gregg Dickerson, MD

  4. Dear Dr. Dickerson:

    Would you be willing to comment on my original suggestion above that experienced practitioners of SBRT might state that “it might take 150-300 patients to learn to do this well and with minimal risk for complications and side effects.”

    I am interested in being able to provide this type of perspective for patients who are considering this type of therapy.

  5. Patients should always seek out the most experienced provider for their care (as is the case with robotic surgery employing the daVinci robot). The primary measure of competency for treating prostate cancer with SBRT is being familiar with SBRT, having experience with the tools, and being familiar with the results realized with the tissues being treated.

    I practiced SBRT for 5 years before beginning to treat prostate cancer cases. I have recommended that physicians who are new to tools such as the CyberKnife acquire a great deal of experience treating sites whose surrounding critical structures are tolerant to the doses being used before attempting to treat prostate cancer. I have proposed that fellowships be set up for radiation oncologists to spend a year after their residency with an experienced radiosurgeon to more rapidly achieve the experience and knowledge to enable them to make the transition from being a radiation oncologist that is familiar with conventional techniques and outcomes to being a radiosurgeon familiar with the techniques and outcomes associated with extreme hypofractionation. Acquiring that experience has certainly paid off for my patients and my practice.

  6. Thanks Dr. Dickerson … That’s very helpful for the patient community!

  7. You’re very welcome. My extremely accomplished chief physicist reminds me to say that it is also a team effort and that experienced physicists and therapists are very, very important to the outcome as well.


    (Yes, he is guilty of shameless self-promotion!)

  8. Well … We have to accept that in this case he is also guilty of being right! Will we also be hearing from the head of your nursing care staff?


  9. Indeed he is! The only reason you haven’t heard from my nurse yet is because she wasn’t looking over my shoulder when I wrote the note. I expect to hear from her as soon as she and the physicist speak.


  10. Update from Dr. Dickerson: I’ve discovered the cause and trigger of delayed toxicity after CyberKnife SBRT for prostate cancer. Over 220 patients patients treated in 2015 have zero delayed side effects and perfect control of their prostate cancer since realizing this and implementing changes in planning and clinical management. Hope to publish on this in the near future.

Leave a Reply

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

You are commenting using your 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: