Does diagnosis/treatment for prostate cancer increase risk for colorectal cancer too?

A newly published study in the journal of Cancer Control has again suggested the possibility that men treated for prostate cancer may be at elevated risk for a secondary diagnosis of colorectal cancer by comparison with similar men never diagnosed with prostate cancer.

This paper by Lu et al. (available as a full text article) is based on a careful analysis of data from the nationwide Swedish Cancer Registry of patients diagnosed between 1961 and 2008 and related Swedish registry data. And the question of whether there is a link between diagnosis and treatment for prostate cancer and a secondary diagnosis of colorectal cancer has long been controversial.

What Lu et al. did was to identify all patients diagnosed with prostate cancer in Sweden during the period 1961 to 2008 who subsequently received a secondary diagnosis of colorectal cancer. For the men diagnosed with prostate cancer from 1981 to 2008, they divided the cohort into three subcohorts:

  • Men treated by orchiectomy
  • Men treated by radical prostatectomy
  • Men treated by any other means

Prior to 1981, most Swedish men with prostate cancer were diagnosed with relatively advanced disease and were treated with estrogen therapy, which has been associated with a decrease in risk for subsequent colon cancer.

Here are the basic results reported by Lu et al.:

  • The study included 149,743 men diagnosed with prostate cancer from 1961 to 2008
    • 33,373 men diagnosed prior to 1981 (most of whom received estrogen therapy)
    • 116,370 men diagnosed between 1981 and 2008, of whom
      • 21,917 received a bilateral orchicectomy
      • 16,521 received a radical prostatectomy
      • 77,932 received other forms of treatment
  • There were a total of 601,542 person-years of follow-up.
  • 1,698 cases of colorectal adenocarcinoma were identified among the 149,743 prostate cancer patients (1.1 percent) during that follow-up.
  • Compared to the incidence of colorectal cancer among the general male population of Sweden who had not been diagnosed with prostate cancer,
    • There was no increase in risk for colorectal cancer among men diagnosed with prostate cancer between 1961 and 1980.
    • There was a significant increase in risk for colorectal cancer among men diagnosed with prostate cancer between 1981 and 2008.
      • For men treated by bilateral orchiectomy, the standardized incidence ratio (SIR) = 1.30.
      • For men treated by radical prostatectomy, the SIR = 1.22.
      • For men treated by other means (including LHRH agonist therapy and radiation therapy, but excluding estrogen therapy), the SIR = 1.37.
  • The increase in risk appeared to be more apparent in cases of adenocarcinoma of the distal colon and rectum than in the proximal colon.

Lu et al. conclude that:

Patients with prostate cancer undergoing bilateral orchiectomy, prostatectomy, or other treatments, including anti-androgen therapy and radiation, may be at increased risk for colorectal adenocarcinoma.

Now we need to be extremely careful about how we interpret these data. For starters,  the risk of a secondary diagnosis of prostate cancer was low. Of the 149,743  men diagnosed and treated form prostate cancer, 98.9 percent did not have a secondary diagnosis of colorectal cancer. Secondly, the vast majority of men diagnosed and treated for prostate cancer in Sweden between 1961 and 1995 were being diagnosed with clinically significant, symptomatic prostate cancer that was at least locally if not regionally advanced or micrometastatic or metastatic disease. From that perspective, these men are difficult to compare to the average man being diagnosed in most of the developed world today, who is being diagnosed with localized disease.

However, the data from this study does seem to suggest the very real possibility that treatment for prostate cancer — and in particular treatment for prostate cancer by any form of androgen deprivation therapy or by radical prostatectomy — could increase a man’s risk for colorectal cancer by as much as 30 percent (from 1.1 percent to 1.4 percent in real terms). It will become easier to assess the seriousness of this risk when we have data from other large patient registries that can be used to confirm the findings from the Swedish data. There is no other truly comparable registry at the present time.


8 Responses

  1. This is another complication I was not warned about. I learned about this risk from reading. So far so good, though fully open doctors would have helped.

  2. Dear George:

    This is an association. It has yet to be proven that there is a real risk.

  3. Dear Sitemaster,

    I know. I should not have used “complication.” Rather some weaker term.

  4. It’s unclear to me how the numbers in your summing-up were calculated. In particular, the following statement — “increased risk … (from 1.1 percent to 1.4 percent in real terms)” — doesn’t seem borne out by the numbers in the article as excerpted.

    Clearly the 1.1% is 1,698/149,743, where 149,743 = 33,373 + 116,370. But that would mean that 1.1% is not the baseline for men prior to 1981, but rather the averaged overall result across all men.

    The numbers seem to an indicate a baseline of well below 1%. In fact, I would guess 0.43%, as follows:

    0.43% = 143/33,373 (pre-1981: “no increased risk”)
    1.30% = 285/21,917 (bilateral orchiectomy)
    1.22% = 202/16,521 (radical prostatectomy)
    1.37% = 1,068/77,932 (other, excluding estrogen)

    giving us an average of 1.13% = 1698 / 149,743 in total.

    If so, then the unadjusted hazard ratios would seem to be much much higher than the “30%” of 1.4 vs 1.1 that the summing-up cites. In fact, they would seem to be as follows:

    300% = 1.30/0.43 for bilateral orchiectomy
    285% = 1.22/0.43 for radical prostatectomy
    320% = 1.37/0.43 for other means (excluding estrogen)

    Hazard ratios this large would be headline news, so I’m guessing I’ve misunderstood something important. Can you help me out?

    Also, regardless of my misunderstanding, it would be particularly interesting to calculate the adjusted hazard ratios of men treated with/without radiation.

  5. So if you have high-risk prostate cancer it seems to me that you have no choice but to treat one problem at a time. Can’t be worried about bridges that you may never need to cross.

  6. Dear Paul:

    The 30% increase in risk is based on the SIR values (which range from 1.22 to 1.37) and that risk increase is confirmed by a statement at the bottom of p. 268 in the full text of the article.

    If one looks at the actual numbers of cases of colon cancer before and after 1980 (i.e., before and after the initiation of use of androgen deprivation therapy) one finds the following — using the data from Table 1 in the paper:

    — In the estrogen-treated patients prior to 1981 there were 198 cases of colorectal cancer among 33,373 prostate cancer patients (0.59%)
    — In the patients treated from 1981 to 2008 there were 247 + 156 + 1,097 = 1,500 cases of colorectal cancer among 21,917 + 16,521 + 77,732 = 116,370 prostate cancer patients (1.29%)

    That’s a 188% increase in risk. … But the authors don’t come close to suggesting that.

    Now I’m not a statistician so I can’t make any more sense of detail of these data than you can. My point was exclusively that if there was only a 30 to 40% relative increase in risk, and the overall level of risk was only 1.1% ro begin with, that’s not a big absolute increase in risk.

  7. And of course there is the possibility that the increased risk had nothing whatever to do with treatment, but was simply a manifestation of some patients’ immune systems’ inability to fight off cancer cells of any sort, whether prostate or other.

  8. I agree that going from 1.1 to 1.4%, although its a 30% increase, is still very low risk. I don’t see why so much is made of this type of increase. And again, are you supposed to not treat prostate cancer because you’re worried about the tiny chance that it will result in other types of cancer? Come on!

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 )

Google photo

You are commenting using your Google 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: