Here's an old report in Pubmed from 1987.  I have no idea
how accurate it is, or whether current techniques are better
or worse than they were then.

The authors are claiming pretty precise knowledge of how
often this occurs, 1/3 of 1% of the time.

---------------------------------------------------

Seeding and perineal implantation of prostatic cancer in the
track of the biopsy needle: three case reports and a review of
the literature.

Haddad FS, Somsin AA.

Several months (an average of 12.86 months) after perineal
needling of the cancerous prostate for the purpose of obtaining
tissue for biopsy, a tumor nodule becomes clinically evident in
the subcutaneous tissue of the perineum, at the site of the
needling in 0.34% of the cases. This nodule presents the same
histological picture as the biopsy of the prostatic tumor. This
is a review of 15 such cases (12 collected from the literature
and an additional three unpublished cases, two of which are
personal observations). At the time of needling, no metastases
could be clinically detected in any of the patients; the serum
acid phosphatase was normal in 73% of them. The average age of
the patients was 65.66 years. The perineal nodule was tender in
40% of the cases; its average size was 2.5 cm. Excision of the
nodule was the most frequently employed form of management. At
the time of reporting, 60% of the patients were living and well,
for an average of 18.56 months after excision. In order to
prevent perineal implantation, especially in patients who are at
risk, it is suggested that a fine needle be employed to obtain
prostatic tissue for biopsy, and that every possible therapeutic
effort be made.

According to Walsh in Guide to Surviving Prostate Cancer, prostate
cancer cells escape the prostate all the time.   The issue is whether or
not such cells have the capability to survive distant from their point
of origin.  That is called metastatic capability, and it is what we all
fear since in that case early treatment won't cure the disease.  Some
recent research confirms that cancer cells are found in the blood of men
with prostate cancer, and at least for early cancers the incidence drops
significantly after treatment.   Also note Alan's answer.   It seems to
me that having the biopsy probably doesn't significantly change the risk
of recurrence after treatment, and that the risk of metastasis is much
much greater if the cancer is undiagnosed and untreated than from any
cells which may escape during biopsy.  Of course, whether that risk is
high enough to merit treatment depends on a variety of factors such as
the life expectancy of the patient and the degree of aggressiveness of
the cancer.  There are those who argue that for many men treatment or
even screening is not generally helpful, but I think the issue of
whether or not the biopsy is a significant factor in metastasis, is a
red herring.

thanks Alan, Leonard, ron for illuminating that.

My fear about this was aroused initially in reading Scardino's book
where he counsels patients not to even bother with a biopsy unless
they intend to undertake treatment soon after.  As a gleason6 (2 years
ago, now GL7) I opted for WW.  Just wondered what those loosened
cancer cells might have been up to in the meantime.