Howarp University
WASHINGTON 1, D. OC.

December 20, 1969

SCHOOL OF MEDICINE
DEPARTMENT OF PHARMACOLOGY

Dr. Joshua Lederberg

Department of Genetics

Stanford University Medical Center
Stanford, Calif.

94305

Dear Dr, Lederberg:

I have followed your newspaper columns for some time and have found them
refreshing for their sober and reasoned comments and for their absence of shrill
invective. Consequently, I would like to take sober and reasoned issue with
you on your remarks on the cancer clause in the Delaney Bill, of which you
wrote in the guest editorial in C & EN 47:5:1969.

The basic premise of the Delaney Bill, widely used in pharmacology and
toxicology, is the dose-response relationship, and therefore of a no-effect
level for populations and individuals, In a population, this no-effect level
may statistically be at a very low effect level; say 0,00005%. TIt™is prob-
lematical that this low level effect would occur or be detected, Among other
things the small number of susceptibe individuals could well have been elim-
inated for other reasons, It is just as reasonable to ct this as to expect
the complete carcinogenic effect in 100 individuals in a population of 200
million. More likely, the hazards of life would have eliminated some signi-
ficant fraction of the susceptible individuals, and considerably less than
100 would have ben affected; but of course, to each of the few individuals,
it would have wen extremely important, But the hazard to one or a few in-
dividuals in an entire population must be considered in the total context
of all other hazards, Is liver damage, kidney damage or nerve fibre demyelin-
ation less undesireable than cancer induction? Tolerances for food additives
and pesticides based on no-effect levels, and a safety factor, are estab-
lished for many substances which cause such damage, These no-ef fect and
safety levels may actually be greater than the 0,00005% level of which we
spoke, The law says that the use level should be at the no-effect level ani
a safety factor. In this context, the cancer clause is irrational and the
bill should be amended to place carcinogenic substances in the same category
with other pharmacologically active substances, or to ban all such substances
as food additives regardless of the lefel at which toxicity may occur. The
latter occurred before the passage of the Delaney Amendment to the Food and

Drug Act,

Whether the carcinogenic ED50 for the cyclamates was 50 times the use
dose is really not the basic issue, The basic issue is whether the tolerance
Which is granted is rationally determined and whether it represents the most
desireable allowable level of use. It las been common practice in the Food
and Drug Administration to set the tolerances for food additives and pesti-=
cidés at the no-effect level times an arbitrary safety factor of 100 (2 logs),
Recently a 2000 times safety factor has been under discussion, Regardless of
the figure, 100 or 2000, the use of an arbitrary factor is wrong, Consider
two substances, carcinogenic or otherwise, but with the same ED50, The slope
of the regression of one of these is 5, and of the other, 0,0001, One tenth
the ED50 (1 log) of the first would affect 8.5% of the population, The same
fractional dose of the second would affect 45%. In one instance 100X the no-
effect level would be quite safe; in the other, it would be anything but safe,
2
It is in the disregard of such differences that the greatest danger lies.

In the case of cyclamates, cancers have been induced at doses of only
50X greater than use levels (1.7 logs), and I have been told that statistically
significant increases in cancer have been found at ‘bwer doses and at loca-
tions other than the bladder. The case is strong for the discontinuance of
the use of cyclamates, since the assumption is that carcinogenesis is equal
among the mammalian species (which, of course, may not be true),

It would be better to amend the cancer clause in the Delaney Bill, It
would also be better to eliminate the GRAS list. In its place I would suggest
a reimposition of the “prior sanctions" category, with no time limit, such
as the original category had. In this category, those compounds with a long
history of use, or t mse which might be used with no apparent hazard at any

level, would be subject for study by producers if the FDA found an reasonable
cause for such study,

Some of the answers must be found in better legislation and in rational
administration of such legislation... And it is such as you, who must lead
the way.

If this letter is too long, accept my apologies, It has been pared
considerably from what I originally wrote,

My best wishes of the Season,

  

ad slr
Frederick Spérli

Associate Profesdor of
Pharmacology