Radioactive phosphorus uptake test
An in vitro analysis of choroidal melanoma
and ocular tissues
Samuel Packer, Jerry A Shields, David R. Christman, Alfred P. Wolf,
and Harold L. Atkins
The concentration of radioactive phosphorus in uveal melanoma and normal parts of the eye
was determined in vitro in 14 eyes. The eyes were enucleated after a positive 32P uptake test.
Portions of the melanoma as well as normal choroid, retina, sclera, lens, and vitreous were
analyzed. The 32P uptake test had been performed at various intervals after intravenous administration of :!2P from 24 to 556 hr. The in vitro uptake of 32P was compared to cell type, tumor
volume, time of testing, percent uptake measured clinically, and specific activity. The only
positive correlation was between percent uptake measured clinically and 32P concentration
(dpm/gm). A higher concentration of phosphorus in melanoma resulted when carrier-free 32P
was used. A negative correlation existed between number of hours from injection to clinical
measurement of percent uptake, although melanoma to normal choroid ratios did. not change
from 24 to 72 hr. No correlation was found between uptake and tumor volume. The sample was
small; however, we saw no correlation between 32P uptake and degree of malignancy.
Key words: radioactive phosphorus, choroidal, melanoma, uveal, malignancy
R
adioactive phosphorus (32P) concentrates
in malignant tissue to a greater degree than
in normal tissues. This fact has been known
since 19411 and is the basis for its use in ophthalmology as an aid in the diagnosis of malignant melanoma of the choroid. Some auFrom the Medical Research Center, Brookhaven National Laboratory, Upton, Long Island, N. Y. (S. P.,
D. R. C , A. P. W., and H. L. A.) and Division of
Ophthalmology, North Shore University Hospital and
the Department of Ophthalmology, Cornell University Medical College (S. P.), Manhasset, Long Island,
N. Y., and the Oncology Unit, Retina Service (J. A. S.),
Wills Eye Hospital, Philadelphia, Pa.
Research carried out under the auspices of the U.S. Department of Energy under Control No. EY-76-C-020016, NIH grant CA 16216 and a grant from the
Mildred Werner Cancer League.
Submitted for publication Nov. 6, 1978.
Reprint requests to: Samuel Packer, M.D., Medical Research Center, Brookhaven National Laboratory, Upton, New York 11973.
thors believe that a high 32P uptake correlates
well with the presence of an intraocular
malignancy.2"4 However, 32P uptake has not
correlated with histologic features,5' 6 or
prognosis7 and more recently has given false
results. 8 " 10 The results with choroidal
hemangioma have also proven to be variable. 11 ' 12 The uptake of 32P by the suspected
tumor is measured against a normal area of the
same eye and expressed as a percentage. How
positive the 32P uptake should be to rely on it
for enucleation remains unanswered, although most would agree to greater than
100%. The range between 50% and 100% remains a gray zone with different authors using
different percent uptake for positives, 50%, 6
60% 12-14 and 85%. 15
The purpose of this study was to measure
phosphorus concentration by in vitro analysis
of 32 P. We then sought to correlate this with
32
P uptake in vivo as well as a number of
386
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017
Volume 19
Number 4
Radioactive phosphorus uptake test
i
i
i
MIXED AND
PRED SPINDLE B
500 400 10
i
Q. 300 -
500
-
450
-
400
-
350
-
300
-
i
387
I
EPITHELIOID
•
•
-
•
-
200 250
100 -
-
#
•
200
24
48
72
INTERVAL (h)
-
*
150
556
100
Fig. 1. Radioactive phosphorus uptake at various
intervals after injection.
(556)
50
I
I
I
24 48 72
1
1
1
1
1
1
1
1
1"II
1
3.0 -
1
1
I
24 48 72
TIME (h)
-
Fig. 3. Radioactive phosphorus percent uptake at
various times according to cell type.
2.5 -
D
X
D
2.0 -
X
1.5 -
-
•
-
X
Q
1.0 -
-
D
A
•
24 HOURS
D 48 HOURS
0.5 -
X
72
HOURS
A 556 HOURS
0.1 ~
1
I
1
80
160
1
1
1
1
240 320
% UPTAKE
1
I
400
I
I
"
480
Fig. 2. Percent dose per gram of 32P in melanoma
in vitro and corresponding uptake measured
clinically.
prognostic factors. This was done in an attempt to explain previous failure to correlate
32
P uptake in vivo and degree of malignancy 2 ' 5
and to determine the reason for the wide
range of uptake seen with similar tumors. We
examined eyes at different intervals after injection of 32P to find an optimal time for per-
forming the test. We believed that by examining the melanoma and the normal tissues we might find a time of optimum tumor
to background ratio.
Materials and methods
Fourteen patients were given radioactive phosphorus (10 ju-Ci/kg) intravenously. Measurement
of the radioactivity from the suspected melanoma
and a control area was performed with the standard surgical techniques at varying intervals after
injection (24 to 556 hr). Most testing was at a 24,
48, or 72 hr interval. Eight patients were women
(age range, 52 to 86) and six were men (age range,
51 to 67). All testing was done in the operating
room, with the control site being in the same eye
in a quadrant opposite the area of tumor. The lesion was localized with indirect ophthalmoscopy
and transillumination. Radioactivity was measured
with either a silicone solid state probe or a
Geiger-Mueller probe. A minimum of three series
of 60 to 100 sec accumulated counts were taken
over the tumor and over the control area. Percent
uptake was determined as follows:
average counts over tumor —
,
average counts over control
,~~
% uptake =
r x 100
average counts over control
m
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017
Invest. Ophthalmol. Vis. Sri.
April 1980
388 Packer et at.
Table I. Phosphorus-32 uptake in human choroidal melanoma
Specific
Patient Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
79
67
69
52
61
63
63
57
86
59
59
60
51
76
Sex
F
M
F
F
M
F
F
M
F
M
F
M
M
F
Time
(hr)
72
72
48
48
72
24
48
24
556
48
48
24
48
48
32p
uptake
activity
(mCi/funol)
207
214
167
260
257
266
171
400
114
209
155
494
450
Mgof
phosphorus
%dose/gm
dpm/gm
Melanoma
cell type
NA
NA
466
268
680
680
345
NA
250
311
*
*
*
*
2.26
1.99
0.70
1.12
1.47
1.91
0.93
2.80
0.82
2.03
0.71
34,865
16,635
10,884
28,220
22,693
29,470
14,368
43,287
10,389
31,279
122,861
140,465
255,020
350,231
Mixed
Epithelioid
Mixed
Epithelioid
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Spindle B
Epithelioid
Epithelioid
Spindle B
NA
NA
0.183
0.269
0.124
0.124
0.244
NA
0.275
0.275
*
*
*
*
88
NA
2.81
2.27
NA = not available.
•Carrier-free, less than 3.0 x 10" 3 fig/mCi.
Table II. Percent dose per gram of phosphorus-32 in ocular tissues
Patient
Time
(hr)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
72
72
48
48
68
24
48
24
556
51
48
24
48
48
Tumor
Choroid
Retina
Vitreous
Cornea
Lens
Muscle
2.26
1.99
0.70
2.12
1.47
1.91
0.93
2.80
0.82
2.03
7.09
0.23
0.33
0.39
0.67
0.24
0.24
0.16
0.48
0.44
0.07
1.56
0.77
0.64
0.21
0.22
0.38
0.63
0.16
0.42
0.27
0.16
2.80
0.48
0.53
0.38
0.37
0.09
0.16
0.17
0.10
0.11
0.27
0.03
0.06
0.85
0.44
0.26
0.20
0.14
—
—
—
2.81
2.27
0.31
0.30
—
—
—
0.26
0.42
0.73
The minimum percent uptake was 88% in this
series. Enucleation was promptly performed following the positive 32P test. Immediately after
enucleation, the eye was dissected, and a small
portion of the tumor and normal eye parts (10 to
100 mg samples) (cornea, lens, vitreous, retina,
and choroid) were placed into sealed vials. These
were autoclaved and then analyzed in the Chemistry Department at Brookhaven National Laboratory. All specimens were digested, and beta counting was done in a liquid scintillation detector. At
no time were these specimens exposed to formalin, and all specimens were analyzed in their original vial. Laboratory data concerning activity were
expressed as disintegration of 32P per minute per
gram of tissue, specific activity (/aCi 32P//u,mol P),
—
0.04
0.14
0.18
0.17
—
—
0.08
0.09
0.19
0.24
0.13
0.44
0.03
0.69
0.18
0.26
0.59
0.19
0.14
0.81
1.20
—
—
—
—
amount of phosphorus (/Ag), and percent dose per
gm. Percent dose per gram was obtained by the
following formula:
tissue activity QxCi)
tissue weight (gm)
X 100
total activity injected (fid)
Clinical data included age, sex, percent uptake
P, number of hours between 32P injection and
measurement of 32P uptake, cell type, pigmentation, mitotic activity, and size (maximum diameter
and volume). No correlations were made with
fluorescein angiography, scleral invasion, optic
nerve invasion, integrity of Bruch's membrane, or
length of time of symptoms before enucleation.
Histology, was performed in the routine man-
32
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017
Volume 19
Number 4
Size
L x W xH
15 x
9 x
12 x
8 x
12 x
7 x
6 x
12 x
14 x
12 x
15 x
12 x
20 x
10 x
13 x 7
Radioactive phosphorus uptake test 389
Vol (mm3)
(L x W x H)
1365
9x5
405
12 x 8
1152
6x3
144
960
294
126
10 x 8
7x6
7x3
12 x 8
12 x 7
12 x 8
12 x 9
12 x 6
15 x 12
8x3
Vol (mm3)
(geometric
analysis)
1344
1176
1152
1620
864
213
889
85
503
208
464
554
380
3600
240
ner; that is, the eyes were placed into formalin
after the dissection described above had taken
place.
The eye, except for the small sample taken for
in vitro radioactivity analyses, was placed into
formalin and processed for histopathologic diagnosis. The specimens for radioactivity analysis
were taken from the apex of the tumor whereas
histologic parameters were from the base of the
tumor. Degree of pigmentation was expressed as
average number of pigment cells per 10 highpower fields (hpf) and mitotic activity as average
number of mitotic figures per 10 hpf.
Results
The clinical and laboratory data are given
in Tables I and II. The patients were selected
at random from those scheduled for the 32P
uptake test. The 32P uptake test was performed at the following intervals after injection: 24 hr (three patients), 48 hr (seven patients), 72 hr (three patients), and 556 hr (one
patient). Fig. 1 is a scattergram of percent 32P
uptake vs. interval after injection.
Scattergrams were also made for specific
activity, micrograms of phosphorus, and tumor base diameter vs. percent 32P uptake.
No correlation seemed to exist between these
factors. Both percent dose per gram vs. percent 32P uptake (Fig. 2) and dpm per gram vs.
percent 32P (not shown) had a positive correlation. Cell types were divided into spindle A
and B and epithelioid. Fig. 3 relates the cell
type and percent 32P uptake. Cell type was
taken from those cells near the base of the
tumor. Tumor volume was determined by
two methods: (1) length times width times
height and (2) a geometric analysis16 when
the shape of the tumor made this reasonable.
Fig. 4 shows the lack of correlation and large
variation between tumor volume and percent
32
P uptake.
Tumor to background ratios are critical for
detection. Fig. 5 shows that there is no variation in the ratio of uptake by tumor to choroid (dpm/gm) vs. time after injection.
Table I shows an obvious increase in disintegrations per minute per gram in patients
who received carrier-free 32P (Patients 11 to
14). This did not vary with cell type but did
result in the two patients with epithelioid
melanomas having a higher percent uptake.
This differentiation (between epithelioid and
spindle cell) was not apparent when noncarrier-free 32P was used (Patients 3 to 7, 9,
and 10). We assured ourselves that the former preparation was near carrier-free by
analyzing for phosphorus. By our testing
there was no more than 0.2 jug/ml phosphorus which is substantially less than the range
of the non-carrier-free 32P (250 to 680 fig);
although to be carrier-free there would have
to be less than 0.003 /ug/mCi.
Fig. 6 shows the lack of correlation between percent 32P uptake and the number of
pigmented cells per 10 hpf. These cells were
at the base of the tumor within 1 mm of the
choroid. Fig. 7 correlates percent 32P uptake
and numbers of mitoses per 10 hpf at the base
of the melanoma. There is wide variation in
each group.
Table II lists the uptake of the tumor and
various ocular tissues that would contribute
to background. Lens, cornea, and vitreous
had low uptake compared to the melanoma.
Similarly 32P uptake by choroid and retina
were consistently lower than tumor. The uptake by muscle may be significant and one
should therefore not allow muscle between
probe and tumor when testing. Table II and
Fig. 5 are evidence that the 32P uptake test
would not be expected to vary significantly
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017
Invest. Ophthalmol. Vis. Sci.
April 1980
390 Packer et al.
1
1
1
1
1
1
1
1
1
30
-
20
-
10
-
8
-
6
-
4
-
2
-
X LWH
o
500
-
-
o
GEOMETRIC ANALYIS
-
X
LJ
X
I
400
X
-
-
Q.
u
n
IO
8«
300
200
-
0
xox
_-
0
o
xj9
X
x
100
--
)<
0
X
0
X
X
_
X
X
-
X
1
400
I
800
1
1
1200
1
1600
//
>2000
VOLUME (mm3)
Fig. 4. Radioactive phosphorus percent uptake
and corresponding tumor volume.
at any time between 24 and 556 hr from injection.
Discussion
Certain correlations seemed unnecessary
because of the physical constraints of the
probe and of 32 P. Phosphorus-32 is a pure
beta emitter with an average beta energy of
0.7 MeV. The physical half-life is 14.5 days
and the average range in tissue is 2 mm. For
example the 32P probe has a 7 mm sensitive
area, and therefore a tumor whose base is
greater than this would not result in a significant increase in the count rate. Perhaps 2
mm to 7 mm tumors would have a progressive increase in count rate with increased
size. Most tumors under 7 mm (~4 disc
diameters) should not be subjected to the
surgical procedure required for the 32P
test since the probability of malignancy is
low.17- 18
Recent studies have correlated the uptake
of radioactive phosphorus with various clinical parameters.2* 4> 5i 17~21 Some of the variability in uptake may well be due to the vagaries of measuring radioactive phosphorus
or to lack of attention to certain parameters of
24
Fig. 5. Ratio of melanoma to choroid (dpm/gm) at
various times after injection.
radioactive materials. The specific activity of
P varies from batch to batch and as a consequence the absolute amount of phosphorus
that each patient receives will vary. The
standard dose of 10 /xCi/kg maintains the
amounts of radioactivity per patient, but not
the amount of phosphorus. If carrier-free
isotope is given, the amount of phosphorus
(mg) differs from non-carrier-free 32 P. It appears from our data that where non-carrierfree 32P was used there is binding of free
phosphate in the melanomas and therefore
less apparent uptake of the radioactive phosphorus. Our results indicate a difference in
uptake when carrier-free material was used;
however, more patients need to be studied.
In addition, the transfer of materials for analysis as well as storage in formalin may alter
results.
The limited tissue penetration of 32P (2
mm)10 means that only radioactivity from the
base of the tumor (1 to 2 mm) is responsible
for the counts obtained with an eye probe
that is placed against the sclera beneath the
tumor. Other tumor-localizing radiopharma-
32
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017
Volume 19
Number 4
500
Radioactive phosphorus uptake test
391
500-
400
LU
300
400-
-
a
a.
300-
CVJ
rO
200 -
200100 20
60
100
140
180
NUMBER PIGMENT CELLS IN 10 hpf
Fig. 6. Percent 32P uptake correlated to number of
pigment cells in 10 hpf.
100-
I
2
3
4
5
6
7
8
9
10
NUMBER MITOSES PER 10 hpf
ceuticals may allow for a more accurate assessment of malignancy than 32p.22> 23 We
found no correlation with many of the prognostic factors of melanoma17' 18 such as size,
pigmentation, and mitoses. It was our intention to correlate 32P uptake with the various
histologic parameters which we examined at
the base 1 to 2 mm of the melanomas. This
should correspond to the tissue penetration
of 32 P. Similarly, samples for in vitro analysis
were from the base of the tumors. We realize
that melanomas are often not uniform in histologic characteristics and therefore feel that
aside from this study it is illogical to make
clinical statements relating 32P uptake to the
malignant potential of the entire tumor unless it is less than 2 mm in height. However it
was our purpose to establish any correlation
given the above limitations. The fact that we
were able to positively correlate 32P uptake
with 32p concentration in vitro adds validity
to our other negative correlations.
The appropriate time to measure 32P uptake clinically has never been determined.
Our data support doing the test at any time
after 24 hr from injection. This is not surprising since phosphorus stays bound to malignant tissue1 and the half-life of 32P is 14.3
days. More patients have to be studied with
carrier-free 32P to confirm these findings.
Our findings agree with those of Rao et al. 5
Fig. 7. Percent 32P uptake correlated to number of
mitoses in 10 hpf.
and disagree with those of Char et al., 2 who
believed that there was a correlation between
degree of malignancy and the 32P uptake. The
main difference is that we did in vitro determination as well as in vivo whereas Char
et al. did only the latter. This is difficult to
understand, since our studies show a good
correlation between 32P uptake measured
clinically and 32P concentration (Fig. 2).
However, both studies lack sufficient numbers of patients to be considered definitive,
and there was a large variance in the previous
work2' 5i 6 as there is in ours.
Donald Margouleff, M.D., Division of Nuclear Medicine, North Shore University Hospital, helped with the
manuscript. Dennis Greenberg and Joan Briggs provided technical assistance. Dr. Merlyn Rodrigues at
Wills Eye Hospital and Dr. Madilyn Kahn at North
Shore University Hospital performed the ophthalmic
pathology.
REFERENCES
1. Erf LA and Lawrence JH: Phosphorus metabolism
in neoplastic tissues. Proc Soc Exp Biol Med 46:694,
1941.
2. Char DH, Crawford JB, Irvine AR, Hogan MJ, and
Howes EL Jr: Correlation between degree of malignancy and the radioactive phosphorus uptake test in
ocular melanomas. Am J Ophthalmol 81:71, 1976.
3. Shields JA, Sarin LK, Federman JL, et al: Surgical
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017
Invest. Ophthalmol. Vis. Sci.
April 1980
392 Packer et al.
approach to the 32P test for posterior uveal melanomas. Ophthalmol Surg 5:13, 1974.
4. Shields JA: Accuracy and limitations of the 32P test in
the diagnosis of ocular tumors. An analysis of 500
cases. Trans Am Acad Ophthalmol Otolaryngol
85:950, 1978.
5. Rao NA, Gamel JW, McMahon RT, and McLean
IW: Correlation of in vitro 32P counts with histologic
features of malignant melanoma of choroid and
14.
15.
ciliary body. INVEST OPHTHALMOL VISUAL SCI 16:98,
1977.
6. McLean IW, Zimmerman LE, and Evans RM:
Reappraisal of Callender's spindle A type of malignant melanoma of choroid and ciliary body. Am J
Ophthalmol 86:557, 1978.
7. Hagler WS, Jarrett WH II, and Killian JH: The use
of the 32P test in the management of malignant melanoma of the choroid: A five-year follow-up study.
Trans Am Acad Ophthalmol Otolaryngol 83:49,
1977.
8. Feldon SE, Sigelman J, Albert DM, and Smith TR:
Clinical manifestations of brawny scleritis. Am J
Ophthalmol 85:781, 1978.
9. Gass JDM, Guerry RK, Jack RL, and Harris SG:
Choroidal osteoma. Arch Ophthalmol 96:428, 1978.
10. Shamas HF, Burton TC, and Weingeist TA: Falsepositive results with the radioactive phosphorus
test. Arch Ophthalmol 95:2190, 1977.
11. Hoskins JC, and Kearny JJ: A case report of a negative 32P test in a histologically proven choroidal
hemangioma. Arch Ophthalmol 95:438, 1977.
12. Jarrett WH II, Hagler WS, Larose JH, and Shields
JA: Clinical experience with presumed hemangioma
of the choroid: radioactive phosphorus uptake studies as an aid in differential diagnosis. Trans Am Acad
Ophthalmol Otolaryngol 81:862, 1976.
13. Shields JA, Hagler WS, Federman JL, Jarrett WH
16.
17.
18.
19.
20.
21.
22.
23.
II, and Carmichael PL: The significance of the 32P
uptake test in the diagnosis of posterior uveal melanomas. Trans Am Acad Ophthalmol Otolaryngol
79:297, 1975.
Goldberg B, Tabowitz D, Kara GB, Zavell S, and
Espiritu R: The use of 32P in the diagnosis of ocular
tumors. Arch Ophthalmol 65:196, 1961.
Ruiz RS and Howerton EE: 32P testing of posterior
segment lesions. Trans Am Acad Ophthalmol Otolaryngol 79:287, 1975.
Kramer SK and Lapiana FG: Quantification of intraocular malignant melanoma volume: a preliminary report. Ann Ophthalmol 8:781, 1976.
McLean IW, Foster WD, and Zimmerman LE:
Prognostic factors in small malignant melanomas of
choroid and ciliary body. Arch Ophthalmol 95:48,
1977.
Shammas H F and Blodi FC: Prognostic factors in
choroidal and ciliary body melanomas. Arch Ophthalmol 95:63, 1977.
Packer S and Lange R: Radioactive phosphorus for
the detection of ocular melanoma. Arch Ophthalmol
90:17, 1973.
Ruiz RS and Howerton EE Jr: Choroidal melanomas
size and accuracy with the radioactive phosphorus
test. Am J Ophthalmol 78:794, 1974.
Ruiz RS and Pernoud JM: Time and dose for optimum radioactive phosphorus uptake measurement
in rabbit uveal melanoma. Am J Ophthalmol 82:218,
1976.
Winchell HS: Mechanisms for localization of radiopharmaceuticals in neoplasms. Semin Nucl Med
6:371, 1976.
Packer S, Lambrecht RM, Christman DR, Ansari
AN, Wolf AP, and Atkins HL: Metal isotopes used
as radioactive indicators of ocular melanoma. Am J
Ophthalmol 83:80, 1977.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933091/ on 06/17/2017