An Opinion

A patient asked:

"Ten years ago they were not using 79-82 grays in proton. They are now. A much better result. Maybe you should try to see the picture more clearly."

And you are right.  Higher doses of external beam radiation do give better results - at least in disease freedom in the relative short studies that we have.

But since we are talking about proton - let look at a study with proton that shows exactly that - one that I have quoted from before.

A study (JAMA. 2005 Sep 14;294(10):1233-9) PMID: 16160131 was done between MGH and LomaLinda in which 393 Patients were randomized between Jan 96 and Dec 99.  Whereas medium follow-up was 5.5 years the range was 1.2 to 8.2 years.  Those treated for 1.2 years have not obviously had time to fail.

The low dose group received a total of 70.2 cGy (a combinations of proton and photon) and the high dose group received 79.2 (a combination of proton and photon).  Although the stats in the two groups were close to being the same the high dose group actually had more low grade risk groups, less intermediate and high grade risk groups.  The staging was similar but the high dose group Gleason scores had less Gleason 8-10 and unknown Gleason.  The high dose was an average of 1 year younger (younger men have better prognosis), fewer blacks (blacks have a worse prognosis) and the PSA was about the same.  Obviously those receiving the high dose overall had less serious cancer.

They actually figured 4 definitions of failure in their study.  They used as the primary definition the ASTRO definition which is 3 consecutive rises in the PSA with the date of failure back dated to half way between the last PSA before the first increase and the first increase.  They used the ASTRO definition with no back dating and to some extent they use a definition of a PSA nadir of 0.5 plus yet another one of a PSA nadir of below 1.0

Is there a difference - lets see what they had to say.

On high dose

ASTRO - back dating - 80.4%

ASTRO - no back dating - 

Nadir 1.0 - 81%

Nadir 0.5 - 44.7%

Medium time to nadir in high dose was 39.6 months (remember some had only been treated for 1.2 years).

From other studies we know that we can deduct some 35 points difference between ASTRO and 0.5 - lets see 80.4 - 35 = 45.4 strangely agreeing with what they found.

Now this is at 5 years subtract an additional 1.5 to 2.0 points for each hear before 10 in the cases of the nadir figures (ASTRO changes very little).  This now brings the nadir 0.5 definition to maybe 35 points.

Overall survival showed no difference.  Acute and Late Genitourinary (GU) and Gastrointestinal (GI) morbidity was slightly higher in the high grade group - as would be expected.  84% had Acute Grade 1 and Grade 2  GU problems and 82% had Acute Grade 1 and 2 GI morbidity.  The Late morbidity was again only slightly higher with a total of 63% Grade 1 GU and 60% for GI.  These are very decent figures but I would not say they are a whole lot better than other forms of high dose radiation using IMRT by experts.

One of the telling stories of any study are the patients at risk when the report is made.  The high risk group started with 195 patients and at 5 years they still had 111 of them they were following.  However at year 6 and 7 it was only 53 and 20.

Studies (including proton studies) have routinely made the statement the lower the PSA following treatment the higher the chance of success.  Some studies have shown that almost all men who do not nadir at below 1.0 will fail with a rising PSA.  This figures of course reduces as the PSA gets lower and lower.  Studies will show if we accept the PSA nadir of 0.5 at five years and try to balance that out to 0.2 at 10 years we would come up with someplace between 30 and 35% success rate.

Other forms of treatment including surgery will certainly do much better than this.

To me this study only verifies what we have seen in others - proton beam has not come of age yet even using an escalation of cgy's.  It is the points made above that makes it essential that one knows how to read and reads the whole study - not just an abstract and not just a news release or an internally generated paper.  In order to make the above comparisons one must have knowledge of many other studies.  It took me 5 months of 14 hours days 7 days a week to make my treatment decision.  I realized early on that proton beam did not match up to the others despite many who swore by it (they have all failed now).

Review of a study on: Conformal Proton Therapy for Prostate Carcinoma

By: Jerry D. Slater et al. Loma Linda University Medical Center

International Journal of Radiation Oncology Biol. Phys., Vol 42, No. 2, pp. 299-304

This is a study of 640 men with localized prostate cancer that were treated between December 1991 and December 1995. The treatment was proton beam with or without photon beam. Patients ranged from stage T1 to T3. 119 patients were T1 and 484 were T2 and the balance of 37 were T3. There were 232 patients with Gleason 2 - 5, 324 with Gleason 6-7 and 54 with a Gleason of 8 - 10. (My comment - this shows a lot of patients with Gleasons of 5 and under - maybe more so than most studies.)

Patients with a higher risk were treated with 45 Gy of conventional photon beam whole-pelvis therapy and an additionally 30 CGE (cobalt Gray Equivalents) for a total of 75 CGE of proton beam therapy to the prostate with a 1.2 cm margin around the prostate. The proton was given before the photon treatment when both treatments were included. Where low risk subclinical patients, proton beam was used exclusively and were given 74 CGE. In all cases the proton beam was given from both sides and included the seminal vesicles.

Failure was assessed as three consecutive rises in the PSA of greater than 10%. This was only for the patients who had treated at least 24 months prior to evaluation.

Biochemical disease-free survival, patients who had been followed for at least 24 months (555 patients) at 5 years was 79%. Pre-treatment PSA and disease free rate is as follows:

Less than 4 100% - this was actually 4.5 year

4.1 to 10.0 89%

10.1 to 20.0 72%

over 20.0 53%

Using PSA nadir following treatment, the following were found at 5 years:

Less than 0.5 91%

0.51 to 1.0 79%

Over 1.0 40%

The study concluded with the following statement. "PSA nadir following proton therapy is the most important predictor of long-term disease-free survival, as has been shown in other studies (Ref. no. 1, 21, 25, 26, 33), wherein 91% of the patients who achieved a PSA nadir of less than or equal to 0.5 had no clinical or biochemical evidence of disease at 5 years."

My Opinion:

1. I see the biggest problem with this study is the definition of failure together with the short term of the study. If it takes 18 to 24 months to reach nadir, following radiation treatment, and then three rises of 10% or more above that nadir (whatever it is) to be ruled a failure, the time could easily extend out close to and maybe over 5 years. Until they had the three rises of 10% or more they would be counted a success. The mean and medium time for treatment was 43 months. Then the patients who had not been treated for 24 months were not evaluated for failure so we have only a certain number of patients who were included – the study did not say how many. This is troublesome to me.

2. For reasons above I would have to question the disease-free rates using the pre-treatment PSA. Using only 5 years with the failure definition that they used makes 5 years a little tight for reliable figures. Another three years will give us a little better information. For the same reasons the lower PSA nadir figures could be off if they had not had enough time to rise 10% three times above the nadir.

3. If you compare the lower Gleason stages in this study you find that 38% of the patients were Gleason 5 and below. In the Grado study the numbers were the patients with a Gleason 4 and below were 23%. In addition there were only 9% of the patients with a Gleason 8 - 10 whereas in the Grado study with a Gleason of 7-10 their were 22% of the patients. Not directly comparable but one would think that the Grado study had a higher grade of patients.

4. Since this was done at Loma Linda, a Seven Day Adventist facility, the majority of their patients could be from the church. The members of this church are mostly vegetarians and probably live a better life than the normal run of patients in other institutions. Certainly if diet makes a difference these figures would reflect that. This is an ongoing study that is being done with members of The Seven Day Adventist Church. The members of the church do no smoke, low consumption of alcohol, low consumption of meats, high incidence of vegetarians. What effect this might have would be pure speculation.