FREE AMINO ACIDS IN BLOOD. II 383 
illustrates the findings with another normal individual (A. Knu.). Evidently the 
normal urine pattern is moderately constant, but large differences in the excretion 
of taurine and the methylhistidines are noted in particular. 
Fig. 109 shows a fairly typical urine sample from a chronic granulocytic leukemia 
patient (J. Hay.), while Fig. 110 shows a urine sample from a patient with polycy- 
themia (L. Gre.). No consistent differences were noted for the untreated patients as 
compared to normal controls even when plasma levels deviated from normal levels. 
The patients showed a somewhat greater variability, but the cause of these varia- 
tions was not ascertained since an extensive study of the urine was not undertaken. 
It was felt that the uncertainties related to the method of studying urine by paper 
chromatography were too great to warrant an extensive investigation. The major 
problem is one of choice of sample size. Frequently sample size is controlled by apply- 
ing an amount of urine that contains a constant amount of total nitrogen, urea 
nitrogen, or creatinine. These methods presuppose that free amino acid excretion is 
related to the excretion of these other substances, an assumption that does not seem 
justified to the present investigators. We have observed that most urine samples 
(excluding obvious diuresis with very dilute urine) could be spotted with a constant 
aliquot (such as 0.5 ml) to give useful comparative results despite differences in 
urine volume. This is well illustrated by the findings shown in Figs. 105-107. Moder- 
ate variations in urine volume do not appear to make major differences in the con- 
centration of any particular free amino acid. The variations and uncertainties make 
it difficult, if not impossible, to demonstrate mild amino acidurias or relatively small 
deviations of a few amino acids from the normal levels by paper chromatography. 
Gross changes can be observed as have been reported for certain types of metabolic 
diseases. Such gross deviations have not been encountered in patients with leukemia. 
The urine may show a sequence of progressive changes when therapy is instituted, 
or when therapy is stopped. Figs. 111-114 show paper-chromatographic results 
obtained from a patient (E. McG.) with chronic granulocytic leukemia who had 
been treated with myleran over an extended period until toxic symptoms developed. 
The leukocyte count fell for a time to 500/mm# and the patient suffered a loss of hair. 
The drug was withdrawn one week before the first urine sample was obtained. Figs. 
I1I—114 show the free amino acids in urine samples obtained at weekly intervals. 
The leukocyte count at the time of the first urine sample was 25 000/mm? and changed 
to 63 000, 39 555, and 103 000 on the days when the other three urine samples were 
obtained. The rise in the leukocyte count after withdrawal of the drug was accom- 
panied by a steady decrease of the excretion of taurine and cystine—cysteine without 
any marked changes in levels of other amino acids except the methylhistidines that 
appeared to decrease with time. Although this is evidently associated with the with- 
drawal of drug in this particular case, the variations in the excretion of both taurine 
and cystine-cysteine (as cysteic acid on the chromatograms) are sufficiently great 
in both normal individuals and untreated patients with chronic granulocytic leukemia 
that distinct deviations from the normal levels are difficult to ascertain with accuracy. 
On the basis of the relatively small number of urine samples examined in the present 
study, it is our general impression that the excretion of cystine-cysteine may be 
somewhat increased at certain stages in chronic granulocytic leukemia while both 
taurine and cysteic acid appear less frequently on the chromatograms in other stages 
of the disease. An extensive examination of urine samples from patients studied at 
References p. 447/448 
