New Multiple Myeloma Study

["Ron L. Kathren" <rkathren@tricity.wsu.edu>]

While I make no brief for or against the "new" multiple myeloma study by
Wing et al., since I have not read it, it might be instructive to look at
the already published literature with respect to radiation and multiple
myeloma.   Gilbert was not the only one who saw an apparent association
between multiple myeloma and low level irradiation.  BEIR V (pp. 327 ff.),
citing Cuzik, notes that such an increase in multiple myeloma was observed
in 12 of 17 irradiated populations, with an overall O/E of 2.25 (50 cases
observed to 22.2 expected).  NCRP Report 121, published 5 years after BEIR
V, tabulates no less than 42 studies (Table 3.4, pp. 35-36) with various
populations and exposures (including those cited by Cuzik, except where
later analyses were available for the same populations) coming up with an
observed:expected ratio of 1.17 for the combined studies (503 cases
observed; 431 expected).  Somewhat more than half of the studies showed an
O/E ratio > 1, with by far the largest ratios (5.2, 5.7, 6.2) occurring in
the three studies of Thorotrast patients.  If these three relatively small
Thorotrast studies are excluded, the O/E ratio changes only slightly, to
1.14, 491:429.   Interestingly, the 42 studies tabulated in NCRP 121 did not
include the 1979 paper by Gilbert cited by Otto Raabe but instead included a
later (1993) publication regarding the Hanford cohort which showed no excess
of multiple myeloma (O/E = 0.9, 25 cases observed vs. 27.4 expected) but did
show a marginally positive slope dose -response.

While the Wing et al. study may (and likely does) have its problems, I think
there are enough data from earlier studies to reasonably conclude (or at
least strongly suggest) that there is likely an association between low dose
irradiation and multiple myeloma.  Whether this is age dependent remains to
be seen.  But in any case, my first impression is that the Wing et al. study
does not appear to be too far afield from what has been observed in a number
of other studies.  According to the local newspaper reporter who just
called, Wing et al. have apparently indicated that based on their study the
exposure limits should be reduced.   Whether or not the recommended exposure
limits should be lowered on the basis of the study by Wing et al. is at this
point an interesting speculation, but it is just that -- speculation that
begs additional study to verify that the existing limits provide an
appropriate level of safety.

Ron Kathren






----- Original Message -----
From: Otto G. Raabe <ograabe@ucdavis.edu>
To: Multiple recipients of list <radsafe@romulus.ehs.uiuc.edu>
Sent: Monday, April 10, 2000 3:33 PM
Subject: Re: New Multiple Myeloma Study


> April 10, 2000
> Davis, CA
>
> Steve Wing's UNC study of 4 labs was discussed in RadSafe in detail in
> 1998. Here is a copy of one of the messages I sent in February, 1998:
> *****************************************
>
> February 3, 1998
> Davis, CA
>
> Dear All:
>
> Yesterday Steve Epperson posted the abstract of a new study by Steve Wing
> about risks to radiation workers. Dr. Wing is an epidemiologist at North
> Carolina who personally believes that the nuclear industry is evil.
>
> In the 1970's it was noted by Gilbert et al. (Rad. Res. 79:122-128, 1979)
> that there was an apparent association of some extra cases of multiple
> myeloma among radiation workers at Hanford. This was considered a random
> observation, since it was not found in other studies. Also, remember that
> if you look at every possible type of cancer, ignore those that are lower
> in exposed groups, and look only for those that are elevated, it is
> possible that you will find at least one type of cancer that seems to be
> associated with radiation exposure. That observation is expected by pure
> chance. Then if you ignore everything else, you could claim that this
> random observation proves that radiation is responsible even at the low
> levels of exposure received by radiation workers. This the the method used
> by some epidemilogists who really want something to report.
>
> Okay. What Wing did was get money from NIOSH to study myeloma cases at
> Hanford but incorporating data from other laboratories including LANL,
ORNL
> and SRS. In his abstract he notes that there also were confounding
> exposures of the radiation workers to many other agents "including
> solvents, metals, welding fumes, asbestos,....and non-ionizing radiation."
> However, only exposures to ionizing radiation were quantified well enough
> for him to include in his analysis!
>
> This was a case control study where all the cases of myeloma deaths were
> matched with control persons who lived to the same age, but did not
develop
> myeloma. When radiation exposure was considered he found: "Total
cumulative
> radiation doses were similar between cases and controls." NO RADIATION
> EFFECT! Well, that wasn't the right answer, so now he started subdividing
> the data and found that if he only considered people whose exposures
> occurred at ages 45 or older, there was a significant association of risk
> of myeloma with radiation exposure after age 45 "adjusted for age, race,
> sex, facility, period of hire, birth cohort, monitoring for internal
> radionuclide contamination, and external radiation received prior to age
> 45." This observation is the whole basis of his report. He reports an
> increased incidence of a factor of 4.3 for workers receiving doses greater
> than 5 rem. [Review of the A-bomb survivor data shows a dose of above 100
> rem delivered instantaneously was required to yield this large an increase
> in multiple myeloma cases and simultaneously yielded about a seven-fold
> increase in leukemia cases (BEIR V).]
>
> Wing claims that increased cancer sensitivity in older workers is expected
> theoretically and found in some other studies. Actually, my reading of
BEIR
> V shows a consistent reduction in risk of radiation induced cancer of all
> types with increased age at exposure! Also, if this is a radiation effect
> rather than a chance observation, we would expect that other cancer types
> would be more readily observed, such as leukemia and lung cancer, but
> apparently not.
>
> Wing's results show no overall effect until selectively grouped to achieve
> an observed level of significance. You can guess that he tried every
> conceivable combination of age grouping to harvest a significant
> observation from the study so that he could declare that his "findings and
> other studies of nuclear workers have implications for radiation
protection
> standards for workers and the general public."
>
> Finally, the observations by Wing that there is no overall effect, but one
> remarkably appears when considering exposures of people older than 45,
> suggests that there must be a protective effect for exposures that occur
at
> ages younger than 45. This is because overall he found that: "Total
> cumulative radiation doses were similar between cases and controls." NO
NET
> EFFECT!
>
> Otto
>
> *****************************************************
> Prof. Otto G. Raabe, Ph.D., CHP
>               Institute of Toxicology & Environmental Health (ITEH)
>    (Street address: Building 3792, Old Davis Road)
> University of California, Davis, CA 95616
> Phone: 530-752-7754  FAX: 530-758-6140
> E-mail ograabe@ucdavis.edu
>               *****************************************************
> ************************************************************************
> The RADSAFE Frequently Asked Questions list, archives and subscription
> information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html

************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html