Effectiveness of cimetidine
on the size of melanomas of 40 grey horses
Research project Veterinary Medicine
Nienke Helle Project tutor
3050602 University Utrecht: Joop Loomans
Research period: april-july 2011 Heilan International Equestrian Club:
Table of contents
Introduction ................................................................................................................................. 3
Aim of the study .....................................................................................................................................4
Materials and Methods ................................................................................................................ 5
The melanoma is one of the most common cutaneous tumors and represents 3,8% of the total
equine neoplasia (J.P. Sundberg, 1977) (Laus F., 2010). Other studies find prevalence’s between 4 and
15%. The true incidence may be higher, because these masses are easily identified by practitioners
and therefore infrequently submitted for histological examination (K.C. MacGillivary, 2002).
Melanomas are benign tumors, formed by melanocytes. They are slow growing, locally invasive
masses. Melanomas usually affect grey horses older than 6 year of age. Congenital tumors are rare.
However, they can be found in horses of any coat color, where it usually tends to be more aggressive
(Laus F., 2010). Grey horses are more likely affected when they progress in age. Many reports have
estimated that as many as 80% of older grey horses will develop dermal melanomas (C. Fleury F. B.,
2000; Valentine, 1995). There seems to be a race-related predisposition. Lipizzaner, Spanish Pure
Breed, Camargue horses and Arabians are prone to develop melanomas. The gender of the horse
isn’t related to the development of melanomas (C. Fleury F. B., 2000; M.H. Seltenhammer, 2003).
The most common localizations of the tumors are the perineal area, the perianal region, the external
genitalia, the ventral surface of the tail and the parotid region (C. Fleury F. B., 2000; M. Rodriguez,
1997). Melanoma is the third most common tumor on the penile and preputial surface (J.G.B. van
den Top N. d., 2008). Other sites were the tumors also can be found are the eyelids, the coronary
band, the vertebral region and the nasal cavity (Laus F., 2010). Depending on the location melanomas
can become ulcerated. (R.R. Pascoe, 1999) Common sites for metastasis of the melanoma are the
lymph nodes, liver, spleen, skeletal muscles, lungs, guttural pouch and surroundings, or within blood
vessels throughout the body. The majority of these dermal melanomas are usually benign in grey
horses. Of these benign tumors the thesis has been that they eventually will metastasize and are
therefore potentially malignant. Many of the horses with metastases had dermal melanomas for
years (K.C. MacGillivary, 2002). Metastatic spread is possible to almost all organs of the body and is
common in old aged grey horses, often without related clinical symptoms (Laus F., 2010).
The exact pathophysiology of the melanoma is not fully understood yet. It is believed that
melanocytic tumors are a result of a disturbance in melanin transfer from dermal melanocytes to
follicular cells. It is related to the depigmentation of the skin (W. gebhart, 1977). Depigmentation of
the skin starts around the eyes and the anal region, which explain the high incidence of melanomas
around this area (M.H. Seltenhammer, 2003). Some authors define it as a true malignancy, while
others see it as a storage disorder in association with the depigmentation process that occurs in grey
horses. The graying of hair seems a risk factor, associated with a gene mutation, which is not
recognized in humans (Laus F., 2010). Valentine described four different syndromes for melanoma’s;
1. Dermal melanoma affecting mostly mature grey horses. 2. Dermal melanomatosis affecting aged
grey horses. Melanomatosis is associated with metastases. 3. Melanocytic neavus affecting mainly
young horses. Melanocytic nevi are believed to arise from altered melanocytes within the basal layer
of the epidermis (Valentine, 1995). 4. Anaplastic malignant melanoma in non-grey horses with
histological features of malignancy (Laus F., 2010). Melanocytic nevus have some different
histopathologic features compared to dermal melanomatosis and dermal melanoma (Valentine,
1995). In non-grey horses melanomas can rapidly metastasize through the body and can give rise to
accumulation of large amounts of melanin and sickness. These malignant type tumors can cause
death in a few months (Leroy, 2005). Melanocytic nevi are distinguished from dermal melanoma by
their location in the superficial dermis or dermoepidermal junction, frequent epithelial involvement,
distinct groups of relatively large, frequently mildly pleomorphic, epitheloid to spindle-shaped tumor
cells with euchromatic nuclei, variable cytoplasmic pigmentation and occasional mitosis. Dermal
melanomas have a deep dermal location, are small, homogenous, indistinct, round, or dendritic
tumor cells with condensed chromatin and dense cytoplasmic pigmentation and no visible mitosis
(Valentine, 1995). Benign melanoma seems to share some features with human blue naevus and
with animal-type melanoma (Laus F., 2010).
In veterinary medicine the importance of melanoma lies in its high prevalence, its negative effect on
performance of the sport horse and the negative effect on equine health and wellbeing, mostly
regarding dysfunctions of the urogenital and the gastrointestinal tract (Laus F., 2010). Clinical
symptoms arise when the size of the tumors compromises the function of the involved organ. Other
examples are the Horner’s syndrome or paresis due to spinal cord compression (J. Tarrant, 2001).
Treatment of melanoma’s depends on the localization, the size and the number of the masses.
Surgery in an early development stage is possible, however if tumors fuse to form a tumor tissue
plate, removing becomes very difficult because of the extensive and invasive character of the tumor
tissue plate. Consequently, the risk of relapse is high (Laus F., 2010). Local chemotherapy or cryo-
necrosis can be used as well, individually or in combination with surgery. Results are doubtful
however (Laus F., 2010; J.G.B. van den Top J. E., 2010). Autogenous vaccines have also been
attempted, however without success (K.C. MacGillivary, 2002).
Cimetidine, an H2 receptor antagonist, is a thoroughly researched medicine and is approved by the
FDA for inhibition of gastric acid secretion. It is also used for numerous skin conditions like warts,
urticaria and mastocytosis (Scheinfeld, 2003). It has positive effects on colorectal cancer, renal cell
carcinoma and salivary gland cancer in humans (Laus F., 2010). Cimetidine is interesting for this
research because according to previous research cimetidine plays a role in the regression of tumors.
It has immunomodulatory effects that include blocking suppressor T cells and facilitating cell-
mediated immunity (Scheinfeld, 2003). It also increases the natural killer cell activity (Scheinfeld,
2003). Cimetidine acts probably by three principal mechanisms: 1. Direct inhibition of tumor cell
proliferation by antagonism of the H2 receptor. Histamine H2 receptors have been recognized on
human melanoma cells (Laus F., 2010). 2. Activation of the local immune response characterized by
interferon-gamma production by macrophages. 3. Blocking of stimuli that histamine normally exerts
on T-suppressor cells. Several other mechanisms have been demonstrated as well. Cimetidine has
been proven helpful in treating melanoma in horses (K.C. MacGillivary, 2002; T.E. Goetz, 1990), other
research however failed to obtain the same positive results (Laus F., 2010; J.R. Bowers, 1994).
Aim of the study
The aim of the study is to examine the effect of cimetidine medication on the size of melanomas in
Materials and methods
Forty grey horses with ages ranging between 5 and 19 years were selected for this clinical trial. These
horses were selected in the Heilan International Equestrian Club, located in Xinqiao China, which
owned at the time 68 white horses. Of these 68 horses 40 horses had masses during a clinical
examination identical to melanoma. The participating horses are all tested positive by Fine Needle
Aspiration Biopsy for melanomas and then included in this trial. The biopsies were considered
positive when they contained melanocytes. 24 Horses were older than 10. The group of horses
included 22 Andalusians horses, 10 Lusitanos horses and 8 gray horses who were crossbreeds with a
percentage of Spanish Pure Breed or Lusitano. The horses are used for show activities, all of them
were in training during the research project.
Of the horses the pedigree, the gender, the weight and the age of the horse were recorded before
the start of the experiment. The size, the location, stage of development, morphological
characteristics and the total number of the melanomas are recorded at this stage as well. This will be
done by measurements of the size, characterizing and describing of the location on a registration
form and capturing images of the melanomas. We made images of the entire horse, followed with
pictures specified by the localizations of the melanomas, which guarantees a close up image of the
melanomas. On the pictures the caliper to measure the melanomas and a number given to the horse
are visible. The size of a single melanoma will be determined using a caliper at the widest and the
smallest point pressed on the surface of the skin. Subsequently the surface area will be calculated
out of these two numbers by multiplying them. These numbers are added for each individual horse,
which give us one number for every measure moment. Adequate inspection of the penis is
accomplished through an intra venous injection of acepromazine (0,15 mg/kg BW), xylazine (0,5
mg/kg BW), domosedan (20 mcg/kg BW) or turbogesic (0,1 mg/ kg BW). A rectal examination is
performed to determine if there are masses in the palpable part of the abdomen. Melanomas
smaller then 0,4 cm x 0,4 cm were not included in the measurements.
Every 8 hours for 90 days the horses received 2,5 mg/kg cimetidine. Different studies investigated
the most effective dosage. Laus et al found no significant difference in dividing the dosage during the
day. In other research they found clinical effect with dosage ranging from 1.6 mg/kg PO SID to 7.5
mg/kg PO BID or TID. (Laus F., 2010) Duration of the therapy (90 days) was chosen on the basis of
scientific literature. Improvement of the condition could be detected during the first 4-8 weeks (T.E.
Goetz, 1990), in other research the therapy continued for 13 months (Laus F., 2010; K.C.
MacGillivary, 2002). The weight of the horse is used for determination of the amount of cimetidine
each horse will get administered. The weight of the horses will be classified in five groups, the scale
ranges from 400 kilos until 650 kilos. We have five groups: 400-450; 450-500; 500-550; 550-600; 600-
650. These groups got 4,5 ml, 5 ml, 5,5 ml, 6 ml and 6,5 ml three times a day. We dissolved 125 gram
The weight of the medication will be specified by an accurate and calibrated scale. After the correct
amount is determined the medication will be mixed with water and will be administered three times
a day, mixed with forage, at 08:00, 16:00 and 23:00. After 30 days the amount and size of the
melanomas were measured again. At day 60 and day 90 the third and fourth measurement of the
melanomas took place. In total there were five persons involved with the measuring of the
melanomas. Two veterinary students of the Utrecht University, one senior equine veterinary teacher
of the Utrecht University and two Chinese veterinarian students. During the research the horses
where under close surveillance by the Heilan veterinary team. Record were kept of the physical
condition of the horse.
Statistical significance was tested using the ‘linear mixed model’ method, carried out by the program
SPSS. This is a mixed analysis of a fixed effects model or a multiple variables linear regression model .
This model needs a linear relationship between dependent and independent variables. That’s why
the results of the surface area were transformed into 10Log and makes sure that the model is
properly fitted. The fixed effects that are put in the model include the time, the amount of
melanomas counted for each individual horse and the age.
The history of the horses reported the presence of cutaneous masses in most of the horses for at
least one year. The horses were no longer than three years in the possession of Heilan International
Equestrian Club, the older horses were bought including cutaneous masses. The horses were not
treated for the melanoma’s since the purchase of the horses. One horse had clinical symptoms
relating to the condition, expressed by difficulties extending his penis during urination and diarrhea
following obstruction of the rectum. Melanomas, classified as black, firm and raised, were present in
the 40 horses. The number of masses measured ranged from 1 until 23 with diameters ranging from
0,4 cm x 0,4 cm to 12,5 x 9,0 cm. The localizations of the masses for each horse are presented in
table 5. The most common sites were the ventral side of the tail and the perianal area. Melanomas
appeared mostly in a group. Rectal palpation was performed in all horses. In 3 horses masses were
palpable. In all cytological samples there was a homogenous population of strongly pigmented
epithelial cells with the presence of melanin granules, which confirmed the diagnosis melanoma. No
adverse reactions were observed on the administered cimetidine.
Table 1 shows the calculated total surface area of the counted melanomas of each individual horse
for the four measure moments. T0 is the first measurement moment and functions as a benchmark.
Visible in the table is the variation between the horses. To make the distribution between the horses
more clear, a boxplot was made. It clarifies the median, the minimum and the maximum values and
the lower and upper quartile. The dots are outliers, which means that the calculated value is
numerically distant from the rest of the data. The boxplot indicates that the surface area of the
horses mainly lies between 3,00 cm2 and 19,66 cm2. The second measurement shows more
distribution. The third measurement lies close to the second measurement, the fourth measurement
approaches the first measurement again.
Table 1 Calculated total surface area of the melanomas by horse and time interval
Horse T0 T1 T2 T3
1 19,98 21,91 22,63 21,12
2 8,97 17,28 17,62 11,47
3 15,93 18,91 19,62 15,79
4 6,26 10,98 8,01 8,22
5 5,62 8,11 7,24 3,04
6 14,92 17,22 18,53 10,74
7 0,36 0,49 0,49 0,49
8 12,41 14,59 14,88 11,39
9 11,53 12,15 13,86 11,77
10 8,76 27,31 16,77 24,32
11 3,56 5,2 4,51 4,81
12 37,46 37,39 22,15 55,24
13 2,6 4,02 3,82 0,42
14 4,69 7,76 4,89 6,17
15 14,67 17,13 10,21 13,37
16 4,94 5,17 4,39 3,36
17 15,34 15,52 11,26 7,1
18 0,97 3,16 1,05 0,25
19 7,06 9,22 3,62 5,91
20 97,29 107,32 82,52 101,9
21 152,16 154,87 159,51 135,81
22 36,98 28,8 26,26 8,06
23 13,08 17,78 18,3 18,35
24 30,32 40,07 43,96 48,28
25 28,32 34,29 28,96 27,85
26 6,88 12,11 8,32 6,91
27 9,65 9,06 11,13 9,19
28 3,85 4,06 3,06 3,11
29 19,68 18,27 19,08 8,84
30 1,08 1,82 0,8 0,49
31 38,94 45,78 42,74 45,93
32 3,68 4,86 2,85 4,45
33 3,52 4,45 3,48 4,46
34 2,86 2,57 1,94 2,41
35 3,1 3,57 2,22 1,56
36 1,05 1,04 0,85 1,18
37 0,36 0,36 0,16 0,28
38 2,19 2,06 1,51 1,92
39 0,49 0,36 1,21 0,56
40 27,79 39,29 24,32 33,12
Table 2 shows the means and the standard deviations of the measurements. By adding the mean and
the standard deviation we can calculate that 70% of the horses have melanomas with a total surface
beneath 50 cm2. It is interesting to see that T1 and T2 have higher values then T0. T3 is comparable
N Minimum Maximum Sd Mean Table 2
T0 40 0,36 152,16 28,00 16,98
T1 40 0,36 154,87 29,15 19,65
T2 40 0,16 159,51 27,91 17,21
T3 40 0,25 135,81 27,52 16,99
The output of the linear mixed model showed that the three factors (age, time and amount of
melanomas) are potentially important predictors of the surface area. They were all significant at the
0,05 level as we can see in Table 3.
Table 3 Type III Tests of Fixed Effectsa
Source Numerator df Denominator df F Sig.
Intercept 1 41,123 411,658 ,000
Time 3 119,591 3,690 ,014
Age 11 38,101 3,447 ,002
Amount of 20 126,524 20,531 ,000
a. Dependent Variable: logoppervlakte.
The question we would now like to answer is if there is a significant difference between T0 and T3.
The statistical analysis indicates that there is no significant difference between the mean of the first
and the fourth measurement. This can be seen in table 4. The table shows also that there is no
significant difference in the total surface area between the second and the first measurement and
the third and first measurement.
Table 4 Pairwise Comparisonsb
(J) Mean Difference (I- 95% Confidence Interval for Differencea
(I) Measurement Measurement J) Std. Error Df Sig.a Lower Bound Upper Bound
Second 10 ,004 ,025 122,042 1,000 -,057 ,065
Third 10 -,058 ,024 118,359 ,054 -,116 ,001
Fourth 10 -,047 ,024 117,992 ,177 -,106 ,013
Based on estimated marginal means
a. Adjustment for multiple comparisons: Bonferroni.
b. Dependent Variable: logoppervlakte.
If we compare T3 to T0, we can conclude that the total surface area of the melanomas staid the
same. The results show also that after the first measurement the total surface area increased on the
second and third measurement and decreased by the fourth measurement. However this change in
total surface area is not significant. The experiment didn’t include a control group. This is
unfortunate, because we can’t say that the rise of the surface area in T1 and T2 followed by the
decrease in T3, is solely a result of administering cimetidine. We have no comparison. However
following research of scientific literature it is generally accepted that melanomas don’t go in
regression by themselves and will continue with growing. In this experiment we cannot conclude that
cimetidine had a positive effect on the size of the melanomas. However if we compare T3 to T2 we
see a slight decrease in size. Therefore it would be interesting to follow patients for a longer period
to monitor if the size of the melanomas remained stable or decrease further and that this change in
size is significant.
We have to keep in mind that this experiment used a caliper as a measuring tool. Working with this
instrument requires extreme accuracy of the people performing the measuring. It is not guaranteed
that the same person always measures the same size of one melanoma. Because more people were
involved with the procedure it is likely that there is variation in the results caused by human handling
of the caliper. How big this difference is and if it is significant is difficult to say. This could be
responsible for the little differences of 0,1 mm or 0,2 mm between the size of the melanomas.
Another point we have to consider is the difference in numbers of melanomas of every single horse
each time point measured. Most horses differed in the counted melanomas during the four measure
moments. The research of a fellow student, C. Lettinga showed that there was a significant increase
in the amount of melanomas counted during the experiment. So if we combine the two results of the
experiments we can say that with a bigger amount of melanomas the total surface area of the
melanomas staid the same. This implies that at least some of the melanomas has become smaller. If
we look at the individual melanomas and not at the total surface area, we see changes of maximum
0,3 mm. However because of the arguments mentioned earlier we can’t see that there was a
significant decrease in size of the melanomas.
There are other possible reasons we have to consider as well. First there is the possibility that
melanomas weren’t measured every time because they decreased in size a few millimeters below 0,4
x 0,4 cm and therefore were not measured anymore. The opposite is also possible. Melanomas grew
in size and were large enough to measure. Human error is also a reason. The people involved
measured potentially different melanomas from each other and included more or less tumors. The
measurements from one researchers differed also in time. However the biggest difference was seen
between the first and the second and third measurements. The author of this script was present this
three times. The penile and preputile tumors were also difficult to assess. These tumors are so
localized that they are hard to reach and therefore hard to measure accurately. A last option is the
problem with differentiating between multiple melanomas which grow into each other.
The research contained mostly older horses. (J.R. Bowers, 1994) suggested that this could be a
possible reason for the non-responsiveness to cimetidine in comparison to the research of (T.E.
Goetz, 1990). In this research we included also younger horses. They showed not a different
response then the older horses. The hypothesis is also been made that active and faster growing
tumors are more responsive to cimetidine. Because we didn’t do a histological examination of every
tumor which made it clear if we were dealing with active tumors we cannot support this theory. This
is also a interesting research question.
There is need for further research to evaluate the long term effects of cimetidine. The experiment
should be carefully designed with a control group and a accurate measure instrument. Ultrasound I a
good option, because this also has the possibility to determine the volume of the tumor.
The conclusion of this experiment is that cimetidine did not lead to shrinking of melanomas.
I would like to thank the Heilan International Equestrian Club for their cooperation with this experiment.
Also I appreciated the help of Joop Loomans, Coby Lettinga, and the Chinese colleagues Sha and
C. Fleury, F. B. (2000). The study of cutaneous melanomas in camarque-type gray-skinned horses
(1): Clinical-Pathological Characterization . Pigment Cell Res , 13: 39-46.
C. Fleury, F. B. (2000). The study of cutaneous melanomas in camarque-type gray-skinned horses
(2): a epidemiological survey. Pigment cell rep , 13: 47-51 .
J. Tarrant, T. S. (2001). Diagnosis of malignant melanoma in a horse from cytology of body cavity fluid
and blood. Equine Veterinary Journal , 33 (5) 531-535.
J.G.B. van den Top, J. E. (2010). Penile and Preputial tumors in the horse: Literature review and
proposal of a standardised approach. Equine veterinary journal , 42 (8) 746-757.
J.G.B. van den Top, N. d. (2008). Penile and preputial tumours of the horse: A retrospective study of
114 affected horses. Equine Veterinary Journal , 40 (6) 528-532.
J.P. Sundberg, T. B. (1977). Neoplasm of equidae. J Am Vet Med Assoc , 170 (2): 150-2.
J.R. Bowers, P. H. (1994). Effifacy of cimetidine for therapy of skin tumours of horses - 10 cases.
Australian equine veterinarian , 12 (1) 30-32.
K.C. MacGillivary, R. S. (2002). Metastatic melanoma in horses. J. Vet. Intern Med , 16: 452-456.
Laus F., C. M. (2010). Evaluation of cimetidine as a therapy for dermal melanomatosis in grey horses.
Israel Journal of veterinary medicine , Volume 65 (2).
Leroy, B. E. (2005). Tail-base mass from a horse of a different color. Veterinary CLinical Pathology ,
Vol 34 (1).
M. Rodriguez, V. G.-B. (1997). Grey horse melanotic condition: a pigmentary disorder. Journal of
equine veterinary science , 17 (12) 677-681.
M.H. Seltenhammer, H. S. (2003). Equine melanoma in a population of 296 grey Lipizzaner horses.
Equine Veterinary Journal , 35 (2) 153-157.
R.R. Pascoe, D. K. (1999). Manuel of equine dermatology.
Scheinfeld, N. (2003). Cimetidine; a review of the recent developments and reports in cutaneous
medicine. Dermatology online journal , Volume 9; Number 2.
T.E. Goetz, G. O. (1990). Cimetidine for treatment of melanomas in three horses. Journal American
Veterinary Medical Association , 196 (3): 449-452.
Valentine, B. (1995). Equine melanocytic tumors: A Retrospective study of 53 horses (1988-1991).
Journal of veterinary internal medicine , Volume 9 Issue 5, 291-197.
W. gebhart, G. N. (1977). Beziehungen zwischen Pigmentschwund und Malanomatose am Beispiel
des Lipizzanerschimmels. Archives for Dermatological Research , 259: 29-42.
Horse Breed Age Tumor Localization
1 Andalusian 17 Perianal and perineal region, ventral surface of the tail
2 Andalusian 13 Ventral surface of the tail
3 Lusitano 10 Parotid region, perianal and perineal region, ventral surface of the tail
4 Andalusian 17 Shoulder, ventral and dorsal surface of the tail
5 Andalusian 15 Ventral surface of the tail
6 Andalusian 10 Perianal and perineal region, ventral and dorsal surface of the tail
7 Andalusian 10 Neck, perianal and perineal region, ventral surface of the tail
9 Lusitano 11 Mane, ventral surface of the tail
10 Andalusian 10 Ventral surface of the tail
11 Lusitano 14 Perianal and perineal region, ventral and dorsal surface of the tail,
preputium, penis, peritoneum
12 Lusitano 19 Parotid region, perianal and perineal region, ventral and dorsal surface
of the tail, preputium, penis,
15 Andalusian 15 Perianal and perineal region, ventral surface of the tail
16 Andalusian 15 Ventral surface of the tail
17 Lusitano 10 Parotid region, mane, dorsal and ventral surface of the region,
19 Andalusian 10 Shoulder, perianal and perineal region, ventral surface of the tail
20 Andalusian 10 Ventral surface of the tail
21 Andalusian 15 Perianal region, ventral surface of the tail
22 Crossbreed 11 Ventral surface of the tail
23 Andalusian 12 Ventral surface of the tail, preputium
24 Crossbreed 10 Perianal and perineal region, ventral surface of the tail
26 Andalusian 11 Perianal and perineal region, ventral surface of the tail
27 Crossbreed 10 Perianal and perineal region, ventral surface of the tail, preputium
28 Andalusian 14 Shoulder, perianal and perineal region, ventral surface of the tail
30 Andalusian 15 Shoulder, perineal region, ventral surface of the tail
35 Lusitano 10 Shoulder, perianal and perineal region, dorsal and ventral surface of the
37 Lusitano 11 Neck, shoulder, ventral surface of the tail
38 Crossbreed 15 Ventral surface of the tail, rectal mass
39 Andalusian 14 Ventral surface of the tail
41 Lusitano 8 Perianal and perineal region, dorsal and ventral surface of the tail,
45 Lusitano 10 Perianal and perineal region, dorsal and ventral surface of the tail
47 Andalusian 11 Perianal region, ventral surface of the tail
48 Andalusian 13 Perianal region, ventral surface of the tail
49 Lusitano 9 Ventral surface of the tail
50 Crossbreed 13 Ventral surface of the tail
52 Lusitano 7 Thorax, ventral surface of the tail
58 Andalusian 10 Ventral surface of the tail
59 Crossbreed 11 Perianal and perineal region, ventral surface of the tail
61 Crossbreed 11 Shoulder, neck, ventral surface of the tail
62 Crossbreed 5 Perianal region, ventral surface of the tail