THE AMERICAN JOURNAL OF CANCER A Gntinuation of The Joudof Cancer Research
VOLUME XIX OCTOBER,1933 NUMBER2
THE ETIOLOGICAL AND PATHOLOGICAL ASPECTS OF SQUAMOUS-CELL CARCINOMA OF THE PENIS AMONG THE CHINESE
AN ANALYTICALSTUDY OF 107 CASES H. K. NGAI, M.D. (Deportittent oj l’uthology, Peiping Union Yrdicul Collrgc, Peiping) INTRODUCTION Between 1921 and 1931,87 cases of squamous-cell carcinoma of the penis from the Peiping Union Medical College Hospital and 20 cases from various other hospitals in China have been studied by the Surgical Pathological Laboratory of the Department of Pathology, Peiping Union Medical College. Although the literature on the subject of penile carcinoma is voluminous in foreign countries, and although some investigators (7, 10, 26, 29, 66) have drawn attention to the fact that the fre- quency of penile cancer is much higher among the Chinese than among Westerners, no close analytical study of this condition has been made in China. The present series of 107 cases repre- sents one of the largest single groups of cases yet reported. INCIDENCE Various statistics give the average frequency of penile cancer as 18.3 per cent in China (7,10,26, 66), 18.9 per cent in the other Far Eastern countries (57,68,81), and 4.9 per cent in Continental Europe (9,52,83,94). Wolbarst (99), in America, reported penile cancer as representing 2 per cent of all cancers in males, and Andrews (2), in Great Britain, found it to constitute only 1.27 per cent of 7881 cases of primary cancer. It appears, therefore, that there exists a wide variation in the frequency of penile cancer in different localities, the highest figures noted being in the Far East 269 260 6. K. NGAI and the lowest in Great Britain and America, with Continental Europe in between the two (Chart I). ETIOLOGY Age: The age distribution of our patients is shown in Chart 11. The average age is 43.74 years, the two extremes being twenty- three and seventy-four ycars. Forty-four patients, or 41.12 per cent, were below forty years of age and 63, or 58.88 per cent, above forty. Bercovitz (7) in Hainan reported, among other tumors, 29 cases of penilc cancer; 18 of his series, or 62.07 per cent, were
CHART 1 (‘HART I. hEQUEKCY OF I’JCNILE CARCTNOMA AMONG CANCERS OF TliE MALEIN DIBPERENT LOCALITIES CIIART11. AQE COMPARISOXBETWEEN CHINESEAND WESTEEN CASESOF PENILE CARCINOMA Abscissa(%rcprcsciit ago ; ordinates, nunibcr of eases. The author ’s cases (Chineee) are designntcd by the solid line (scale 1: 1) ; thc reported cases (Westerners) by the brokcn line (scale 1: 10).
above, and 11, or 37.93 per cent, were below forty ycars of agc. hiendelson and Ellis (68) reported an average age of 38.1 years among their 13 cases in Siam. An analysis has been made of the age distribution in the cases reported among Westerners, and a comparison with our present series is shown in Chart 11. It will be observed that the greater number of our cases fall between the ages of thirty-five and fifty- four, while in the Western cases the peak is between forty-one and seventy. From these figures it appears that penile cancer develops about a decade earlier among Chinese than among Westerners. Although Barringer and Dean (6) included one case and Dayal (24) another case occurring below twenty years of age, and al- though Kiittiier (52) reported two cases, Hcimann (46) one case, CARCINOMA OF THE PENIS AMONo THE CHINESE 26 1 and the author two cases in patients less than twenty-five years of age, such findings are infrequent. The frequently quoted case of “ cancer of the penis ” reported by Creite (21) in a boy two years old was proved by Schultze (86), on re-examination of the original material, to have been hemangio-endothelioma. Civil State: Of the 87 individuals in the series from this hospital 8, or 9.19 per cent, were single, 77, or 88.51 per cent, married and 2, or 2.30 per cent, widowers. No conclusion can be drawn from the high incidence of married men in this series, since most men in China over the age of thirty are married, although no statistical data on the percentage of married men in the general population are at present available. It has been repeatedly stated that carcinoma of the penis is the result of contact with a cancerous cervix (12, 18, 19, 27, 34, 41, 68). That actual implantation of tumor by coitus, ‘‘rmaccr & deux,” may take place is, however, hard to prove. In our present series, no inquiries were made as to this point. Occupatiorz: From Table I it appears that penile cancer is most prevalent among farmers and coolies; in other words, among the economically poor. Dormanns (29) also drnws this conclusion from a study of data obtained from questionnaires sent to private practitioners in South China, who, as a rule, handle almost ex- clusively wealthy and intellectual patients and who report having seen no cases of penile cancer. This might be explained by the fact that ignorance and negligence of personal hygiene on the part of the poor would favor the growth of penile tumor, but it must be remembered in any case that the bulk of the hospital patients, as of the population in general, consists of the poorer class of people. In the absence of a comprehensive statistical study, any conclusion drawn in this light is apt to be misleading.
TAriLE I: Distribution of Occiqwtioa in Cases of Penile Carn’nonia Farmer ...... 41 cases Coolie ...... 11 cases Merehalit ...... 11 cases Peddler ...... Tcacher ...... 2 case8 Military officw ...... 1 case fitudcn t ...... 1 cam Cook ...... 1 case hpprcrl tice ...... 1 case Shoemaker ...... 1 case Official ...... 1 case Story teller ...... 1 ease No occupation ...... 4 cases Uiikiiowii ...... -20 cases Total ...... 107 cases 262 8. K. NGAI In certain occupations which give rise to repeated irritation to the genital organs, of long standing, carcinoma of the penis has’ been reported (26,72,100). Here, again, without a comprehensive statistical study, no definite conclusion is warranted. Heredity: In the present series there is only one case with a history of cancer in the family, the father of one patient having also suffered from tumor of the penis. Barney (5) reported a fiimilar occurrence among his 100 cases. Venereal Disease: In 85 cases questions were asked regarding the history of venereal exposures. Forty patients, or 47.06 per cent, admitted such exposure, while 45, or 52.94 per cent, denied it. Eighty-five patients also mentioned gonorrheal infection, 32, or 37.33 per cent, admitting a history of gonorrhea and 53, or 62.67 per cent, denying it. A blood Wassermann test was done in 72 cases, 65 of which, or 90.27 per cent, were negative; 3, or 4.17 per cent, were doubtful ; and 4, or 5.56 per cent, were strongly positive. In one case the tumor growth is said to have started from an unhealed chancre. Similar observations have been made by a num- ber of investigators (5,23, 27, 30, 35, 43, 65, 90). These data seem to show that venereal diseases have no im- portant bearing on the occurrence of tumor. Trauma: Only 4 patients gave a history of injury to the penis. The first was a man of forty-seven, who had had phimosis since childhood and began at the age of thirty-seven to have urethral discharge, although he denied venereal exposure. For treatment, he had applied hot onion, and then a sharpened brick, heated red hot, to the meatus of the urethra. Ulceration followed this pro- cedure and growth started. One patient had been hit on the penis and another kicked by a horse. In the fourth case the nature of the trauma was not stated. A small number of cases in which previous injuries of various kinds were closely related to the onset of penile tumor have been recorded in the literature (31, 44, 51, 56, 61). The relationship between trauma and tumor formation is, how- ever, very ditiicult to determine, since it is often impossible to ascertain whether or not the injured part was already in a date of malignant change before the injury. One cannot exclude the possibility that trauma and tumor may occur together merely as a matter of coincidence. The scarcity of cases reported suggests, indeed, that trauma plays a very minor rijle, if any, in the etiology of penile carcinoma. Phimosis: Under phimosis, we include both the genuine con- genital tight prepuce and the long and redundant foreskin. Of the 88 patients giving definite statements regarding phimosis, 87, or 98.86 per cent, gave a positive history. This fignre is significant CARCINOMA OF THE PENIS AMONG THE CHINESE 263 in that it is the highest percentage of incidence of phimosis re- ported, the percentages reported by others ranging from 17.72 to 85. Many authors (6, 34,62,75,99) have pointed out the apparent immunity of the Jews to penile carcinoma. Some (81, 102) be- lieve that this immunity is racial, but the fact that one case has been reported in an uncircumcised Jew (99), while not a single case has been recorded among circumcised Jews, strongly suggests that the immunity is acquired through the ritual practice of early circumcision. On the other hand, it appears that circumcision during adult life does not always provide immunity. Our series includes the case of a fifty-three-year-old man in whom penile cancer developed fifteen years after circumcision done at thirty- five; also 7 cases in which cancer appeared from a few months to two or three years after the same operation on account of papillary or ulcerative lesions within the preputial sac (see below). Other authors (25,45,73,89,96) have also reported cases of penile cancer in circumcised individuals, but failed to specify whether the cir- cumcision was done during infancy or adult life, a point of great importance since it has been emphasized (99) that no cancer of the penis has occurred in persons undergoing early circumcision. Efect of Circumcision and Excision. of Tumor 0% Growth: In our series, circumcision by “local doctors” had been done in 7 cases, in all of which there had been either papillary or ulcerated lesions inside the preputial sac. In addition, there were 14 cases in which surgical intervention in the form of circumcision or excision of tumor had been done on account of tumor or “ulcer.” In all these cases tumor soon developed in the operative wound, and in some of the cases the recurrent tumors grew much more rapidly than the primary growths. PATHOLOGY Macroscopic Pathology Of the 107 cases, gross specimens from only 87 were available for the present study. The data on the other 20 were obtained from the records, when available. Size of Growth: For convenience of classification, the tumors are arbitrarily divided into three grades : small, medium-sized, and large, with diameters of less than 3 cm., 3 to 6 cm., and above 6 cm., respectively. The size is known in 89 cakes: 17 tumors, or 19.10 per cent, were small; 51, or 57.30 per cent, of medium size; and 21, or 23.60 per cent, large. Lewis (58) found among his 34 cases 7 tumors less than 1.5 cm. in diameter, 13 between 1.5 and 3.0 cm. in diameter 264 S. R. NGAI
with the shaft free of tumor involvement, and 14 measuring 3.5 em. or more in diameter with destruction of the shaft. Barringer and Dean (6) reported 10 cases with tumors of 5 sq. em. or less, 3 with tumors of 15 sq. cm., 8 with tumors of 15 to 30 sq. em., and 13 with tumors of 20 sq. cm. or more; the largest growth being the size of a base-ball, involving the distal three-fourths of the penis, and covering about 40 sq. em. It will be observed that the tumors of the present series arc comparatively larger than in cases previously reported, which is probably explained by delay in seeking medical aid.
FIQ. 1 FIQ. 2 FIQ.3 FIG. 1. TIGHT FORESKIN,THROUGH THE ORIFICE OF WHICH PAPILLARY GROWTHS ARE VISIBLE;MULTIPLE ULCERATIVE FIS~JLAE. GRADE I1 FIG. 2. PIlIbIOSIS AND l
In the leukoplakia-like form, one or more areas of the glans penis or of the inner surface of the foreskin show a whitish, firm, rough, granular plaque, with or without suggestion of a papillary tendency. These plaques arc usually secondarily infected and cov- ered with a small amount of purulent exudate. The margin is slightly raiscd and sometimes overhanging. Occasionally the whole glans penis and the entire inner surface of the foreskin show this type of change (Fig. 8). As the name suggests, this type shows a marked resemblance to leukoplakia.’ The tumor, however, is well formed and has passed beyond the stage of
FIO.4 FIQ.5 Fro. 6 FIQ.4. INVOI.VF.MENT OF kvHOLE HIM OF THE I’KEPUCE, INN= SUHPACE OF FORE- SKIN, AND PARTOF THE GLANSPENIS. GRADE I Fro. 5. PAPILLARYG~LOWTII INVOLVINQ WIIOLE PENILE SIIAFT AND ENCROACHING UPON SCROTAL SKIN, WIiICH SHOWS A NIJNBEROF ~)ISCIIARQINQ 81NUSES; IKOUINAL LYMPHNODES FREE FROM METASTASIS. GRADE I FIG.6. PAPILLARYT~JMOR INVOLVING PUUIC AKD SCROTALSKIN WITH COMPLETE DESTRUCTIONOF PENILEMIIAFT; INGUINALLYNPII NODES FREE FROM METASTASIS. GUDE I ordinary leukoplakia. 1)iffcrcntiation between the two is often difficult macroscopically, so that microscopic examination should be done. The papillary form (Figs. 5 and G), usually known as a “cauli- flower-like” growth, is by far the most frequent and constitutes 50.58 per cent of our cases. Although the growth may have a single site of origin, it usually arises from more than one focus; * Precancerous lesions of the pcnis, sucli as leukoplnkia, Icukokcrntosis, kraurosis, or tho older “psoriasis pruclputii” hnve been nientioned by a number of authors (22, 28, 34, 39, 40, 48, 80, 85). Lesions similnr to the so-cnlletl ‘ ‘ I’aget ’Y disense” and ‘ ‘ Bowen ’8 disease” occurring on the glans penis linve been dcscribed (3, 17, 87, 92, 93, 101), and the rare form of erythroplakia on the glans penis is rqortcd by sonic investigators (60, 76, 79). Carcinomntous change following nll the conditions mcntioncd above hns been observed, and these lesions should be considered ns precancerous. 266 8. E. NffAI but owing to the close proximity of the papillae the multiplicity of origin is obscured (Fig. 3). In cases of tight foreskin, the underlying tumor may protrude from the small preputial orifice (Fig. l), or it may break through and appear in one or more places on the surface of the foreskin, with fistula formation (Figs. 1 and 2). . Fistulae were observed in 23 cases in the present series. In such cases, a mere slit through the overlying foreskin will expose the tumor mass. Occasionally one sees cases, such as described by Kuttner, in which the tumor growth spreads between the skin of the penile shaft and the corpora cavernosa toward the root of the penis, so that the organ is uniformly thickened and cylindrical in shape, composed of a stiff skin shell which is ulcerated in its inner surface and encloses the corpora cavernosa like a piston. Often the glans penis is still visible, a condition which Kiittner (52) described as similar to “a turtle that sticks the head out of its shell.” Secondary infection by pyogenic microorganisms, causing a layer of purulent exudate over the growth, is invariably present, so that the surface of the growth is soft and moist. In cases where infection is not so marked, however, and in areas that are not covered by the remaining foreskin, the tumor surface is rather hard and dry. In such cases, needle-shaped, brittle, spi- nous, hard, dry structures may be present. These may be so grouped as to appear like a hard brush, as in a case reported by Maresch and Chiari (64). The shape and size of the growths vary not only in different tumors, but also in different parts of the same tumor. The stroma may be thin and delicate, or it may be thick and abundant. In addition to the secondary infection, retrogressive changes of the tumor itself set in sooner or later, leading to necrosis and eventual loss of tumor tissue. As these processes go on, simul- taneously with tumor invasion of the penile shaft, gross mutilation and eventual loss of the organ may occur (Figs. 5 and 6). In late cases, tumor masses may extend directly and creep on to the pubic and scrota1 skin (Figs. 5 and 6). The nodular type of growth (Fig. 2) forms 28.74 per cent of our cases and is, therefore, not so uncommon as originally claimed by Kuttner (52), who considered it even rarer than the infiltrative type. This type of tumor is generally of medium size, rarely larger than a tennis ball, having a knobby appearance, with a broad base to each nodule, and arising usually from the glans penis but occasionally from the prepuce. It has a suggestion of papillary character, but, in general, the resemblance is slight. The growth is almost always covered by a layer of purulent exudate. The infiltrative or ulcerative type is the least frequent of all, CARCINOMA OF THE PENIS AMONG THE CHINESE 267 forming about 10 per cent of our cases. It rarely attains a size larger than 3 or 4 cm. in diameter and consists of crater-like ulcers on the glans penis with granular, often superficially hemor- rhagic surface and roughly nodular and firm overhanging borders. Atypical papillary or nodular protuberances may be present at the edge. Secondary infection is invariably present and, strangely enough, mutilation of the organ is not so conspicuous as in the other types, although it may be present in cases of long standing. In both the nodular and infiltrative growths, tumor masses may be seen in the corpora cavernosa (Fig. 7) without appearing to have any direct connection with the primary tumor in gross.
Fxo. 7 Fro. 8 Fro. 7. INVASIONOF CORPORA CAVEENOSA BY TUMORTISSUE FIo. 8. ~ROSS-SECTIONOF A LEUKOPLAKIA-LIKEGROWTH INVOLVINQ THE LININ~OF THE WHOLEPFIEPUTIAL SAC. GRADEXI When the blood spaces of the spongy bodies are invaded, tumor tissue may fill up the entire corpora cavernosa of the penis. Although the above seemingly distinct types of growth have been distinguished, transitional forms occur, the exact classifica- tion of which may be difficult. Edent of Growth: In more than 60 per cent of our cases the tumor was confined to the prepuce and glans, about 40 per cent had involvement of the shaft, and about 5 per cent involved also the pubic and scrota1 skin. The latter group, however, are always cases of long standing. Initial State of Growth: In Table 11, it will be seen that the initial state of the growth has been variously described ; the most common forms are “nodules” or “pimples” and “warty growths.” Starting Place of Growth: Positive information was obtained in only 89 out of our 107 cases regarding the starting place of the 268 S. K. NOAI growth (Table 111). It will be observed that the glans penis was mentioned slightly more frequently as the starting place of the growth than the prepuce. The growth not infrequently arose from the coronary snlcus and occasionally from the meatus. According to Demarquay (27), growth starts most frequently on the glans penis. Barney (5) and Leighton (56) agree to this, and Thomsen (95) further specifies that the locality is frequently on the dorsal aspect of the glans near the corona. Kaufmann, however, found that of 33 cases, 20 started on the prepuce and 13 on the glans. TABLE11: Initial State of Growth in Penile Carcinoma Nodule or pimple...... 49cases (51.58y0) 28 cases (29.48%) Ulcer ...... 8 emea (8.42%) ...... 3 case8 (3.16%) IJnaccountahle swelling of prepuce...... 1 case (1.050/,) Small abrasion ...... 1 case (1.05%) Vissurc ...... 1 case (1.05%) r.I rnumntic wound...... 1 case (1.05%) ...... 1 case (1.05%) Patches of granules on prepucc...... 1 case (1.05%) Swelling of mid-portion of shaft...... 1 case (1.057,)
TOTAL...... 95cases(100.0070) * Thcse data are compiled according to the patients’ own statements, but in sornc cases modifications were made after a careful questioning 80 as to put the initial state of growth into the correct class.
TABLE111: Slurling Place of Growth in Penile Carcinoma 39 cases (43.82%) 37 cases (41.58%) 11 cases (12.36%) 1 case (1.1270) ...... 1 case (1.12%) - 89 (100.00%) Tumor arising from the urethra is rare; when it does occur, the site is usually the meatus. Leighton (56) reported one case in which the origin was claimed to be in the urethra at the peno- scrota1 junction. Such a case is undoubtedly rare. ImpinaE Lgmph Nodes .- Inguinal lymph nodes were received for examination in only 52 of our caws. Although some of the other types of tumor give rise to very bulky secondary growth (33), carcinoma of the penis rarely pro- duces bulky metastases, as shown by the cases in this series. Histopat hology Microscopically, penile carcinoma is composed of squamous epithelial cells, or the prickle cells of the epidermis, in different FIO. 8. WELL I)IFFERENTIATFD PRICKLE CELLR WITH INTERCELLULAR RRIDOES IN TUMOR MASSES; Nrrnim0~8MITOTIC FIOURES AT THE GROWINOMAROINS
Fro. lo. ARUWDANTFORMATION OF PEARLS IN TUMOR MASS
Fro. 11. ELEIDIN PIOMEXT APPEARINOIN THE FORMOF SMALL GRANULESIN THE CYTOPLASMOF TUMORCELLS 269 270 13. K. Xt3b.I degrees of cell differentiation. The growth practically always takes its origin from the epidermis covering the glans penis and inner surface of the foreskin, although some authors have claimed an occasional origin from the urethral mucous membrane. In well established cases of epithelioma, invasion of the under- lying stroma is the important part of the histological picture. One sees many irregular masses penetrating into the subcutaneous tissue, the dartos, and the soft areolar tissue, but they cease to proceed further when they reach the deep fascia, which is a thick layer of tough and hyalinized fibrous tissue. In cases of long standing, however, this barrier is eventually broken through, so that tumor cell masses appear in the cavernous tissue.
FIG. 12. TUMORCELLS IN PEUINEWALLYMPHATIC VEBSELS In papillary growths, the formation of papillae is the outspoken feature. These papillae are generally covered with fairly well differentiated and hyperplastic squamous epithelial cells, with moderate or marked hyperkeratosis. The stroma stalk may be composed of very thin or thick strands of loose and vascular connective tissue infiltrated by small round cells and other types of wandering cells. In addition to these outgrowing papillae, downward projections of epithelial massea are invariably present -a point of distinction from the usual benign papilloma or warts which are sometimes present on the external genitalia. These masses vary greatly in size and shape and are irregularly ar- ranged, but the individual cells that make up the mass are rather uniform in size, larger than the normal prickle cells, polygonal in shape, having small or large vesicular nuclei and distinct acido- philic nucleoli and finely granular acidophilic cytoplasm (Fig. 9). Intercellular spines are often remarkably evident. In some cells, lamellated epithelial or keratin pearls may be present (Fig. 10). CARCINOMA OF THE PENIS AMONG THE CHINESE 27 1 In some cases, eleidin pigment can be demonstrated in the layer of cells surrounding the pearls (Fig. 11). Mitotic figures are present in moderate numbers and are mostly typical and sym- metrical. As secondary infection and ulceration are invariably present in these cases, leukocytes in varying numbers are often met among the tumor masses. In the periphery of these cells, the epithelial cells, whose nuclei are dark and round or oval, are generally cuboidal in shape and are often arranged parallel to each other, resembling the cells of the malpighian layer of the normal epi-’ dermis, but without a definite basement membrane. In rare cases, the tumor cells may so arrange themselves as to suggest adenoid
FIQ. 13. TUMORMASS IN A VEIN. v. ELASTICTISSUE STAIN or cystic carcinoma. In such cases, small or large cyst-like spaces lined by a layer of squamous epithelial cells may contain necrotic cells and keratin material. While tumor cells may be well differentiated, as above de- scribed, they may also become irregular in shape and size, have many atypical mitotic figures and hyperchromatic nuclei and less acidophilic cytoplasm, with loss of intercellular spines. A few cells, however, may be sufficiently differentiated to form typical pearls. In very rapidly growing tumors, the epithelial cells may be totally undifferentiated and may even be represented by masses of densely packed spindle-shaped cells. In the rare infiltrative growth, the tumor cells may not form masses at all but be scattered about, infiltrating the underlying tissue. In all these rapidly growing tumors, multinucleated tumor giant cells are frequently seen. Degeneration and necrosis of isolated tumor cells and cell 272 8. K. NGAI masses are frequently seen. In the cell masses, necrosis may involve either the central portion or the entire mace. Different grades of cell degeneration are also to be noticed. The stroma is always infiltrated by a large number of small round cells, a few mononuclear wandering and plasma cells, and occasionally a moderate number of eosinophiles, which in some cases may be abundant. In cases with infection, polymorpho- nuclear neutrophiles may be present in varying numbers. Occa- sionally foreign-body giant cells may be seen surrounding necrotic 'tumor cells. The blood vessels are usually dilated and may, in rare cases, contain tumor cells in the lumen (Fig. 13). Tumor cells in the lumina of the lymphatics at the growing margin are not infrequently encountered. Perineural lymphatics are occa- sionally seen to contain tumor cells (Fig. 12), which may encroach upon and press against the nerve fibers. Gradiig: According to the method of Broders, our cases are graded as follows: GredeI...... 55caees (51.40%) GradeII...... 23casea (21.49%) GradeIII...... 17caees (15.89%) GradeIV ...... 12aaeee (11.22%) - 107 caw (100.00%) It will be observed that most of the cases fall into Grades I and 11, a fact which is interesting in comparison with epitheliomas of the lip, the majority of which are of Grades I1 and I11 (lla) and with those of the cervix, which are usually of Grades I11 and IV (llb). TABLE1V: Oroae Appearance and'Grade of Penile Carcinomn
Leukoplakia- Infiltra- Total Grade like Papillary Nodular tive Unknown Caeee
~~ ~ ~~ ~~ ~~ I ...... 7 37 1 0 10 55 11...... 2 7 11 0 3 23 I11 ...... 0 0 10 2 5 17 IV ...... 0 0 3 7 2 12 TOTAL...... 9 44 25 9 20 107
It is always advisable to examine a number of sections taken from various regions of a tumor, for sections taken from different regions of the same tumor often show different degrees of cell differentiation. The impression that we have obtained from' our cases is not in agreement with Broders' conclusion (llc) th'at, as a rule, there is practically the same grade of malignancy through- out a carcinoma. The variation of the microscopic picture in CARCINOMA OF THE PENIS AMONG THE CHINESE 273 different parts of the same tumor necessitates a more guarded conclusion as to the degree of malignancy when only a single piece of biopsy material is examined. The correlation of the gross appearance of the tumor with the grade of malignancy was analyzed, and the result is shown in Table IV. In the light of this study, there would seem to be some relationship between the gross character of the tumor and the grade of malignancy. Most of the papillary forms were of Grade I or 11; the nodular forms of I1 or 111; and the infiltrative forms of Grade I11 or IV. The recognition of this relationship is of help to the clinician in estimating the approximate degree of malignancy of the growth from the gross appearance alone. A study of the age of the patient in relation to the grade of the tumor (Table V) demonstrated no important relationship be-
TAB= V: Age oj Patient and Grade of Penile Cardm
Grade Grade Grade Grade Total Age I I1 I11 IV cases
~ 20-20...... 8 1 1 0 10 30-39 ...... 15 10 0 3 34 40-49 ...... 14 4 6 3 20 60-59 ...... 11 8 4 6 28 60-69 ...... 7 0 1 0 8 70 ...... 0 0 0 1 1 TOTAL...... 55 23 17 12 107 tween the two. The number of cases is probably too small, how- ever, to be of real value in a correlative study of this type. Mitosis: Mitotic figures vary in number as well as in form. Quantitatively there is a direct relationship between the number of mitotic figures and the grade of tumor according to cell differ- entiation, as shown in Table VI.
TAB- VI ; Number of Mhea and Crade of Pen& Carcinoma
Mitosis Total Grade f + ++ +++ ++++ ca8ea I ...... 12 37 6 0 0 55 I1 ...... 0 5 10 2 0 23 111...... 0 0 1 13 3 17 IV...... 0 0 1 4 7 12 TOTAL...... 12 42 24 19 10 107
Mitotic figures, as observed in these cases, vary considerably, but the majority of them are typical, either symmetrical or bipolar, 13 274 8. K. NOAI although atypical and abortive types are also seen in rapidly growing tumor masses. Gametoid mitosis is occasionally seen, for the most part at the growing margin of the tumor masses. Amitotic division of tumor cells is apparently present in tumors of Grade I11 and IV, as suggested by the multinucleated tumor giant cells. Cellular Imfiltratiolt ilt Stroma: Infiltration of the stroma by Rmall round cells, which are generally regarded as lymphocytes, is invariably present in carcinoma of the penis. Murphy and others (69) showed, in their experimental studies, that lymphocytic infil- tration signifies a reaction of immunity, limiting the growth of cells and establishing local tissue immunity. A correlation of the degree of lymphocytic infiltration in the stroma and the grades of the tumor was made (Table VII). Although our result is not
TABLEVII: Lymphocylic I4l&&m in Stroma VB. Grade oj Pen& CaPcinma
Lymphmytic infiltration - Totsl Grade +++ ++++ +++++ CaeeS I...... 4 (7.27%) 11 (20.00%) 40(72.73%) 66(100.00%) 11...... 1 (4.34%) 2 (8.70%) 20 (88.88%) 23(100.00%) 111,...... 2 (11.76%) 8 (47.08%) 7 (41.18%) 17 (100.00%) IV...... 3 (26.00%) 6 (41.67%) 4 (&3.33%) 12 (100.00%) TOTAL...... 10 (9.34%) 26 (24.30%) 71 (68.36%) 107 Cl~.oO%)
~ striking, it suggests that as the tumor becomes more malignant the degree of lymphocytic infiltration in the stroma is lessened. In the presence of chronic secondary infection and necrosis, however, the significance of this finding is considerably decreased. Plasma-cell infiltration in the stroma has also been regarded as a sign of immunity in the latter part of the tumor process. No relationship between the degree of plasma-cell infiltration and the duration of the tumor could be demonstrated in this series. Eosinophil infiltration in the stroma was present in over half of our cases. From a correlative study it seems quite clear that there exists no definite relationship between the degree of eosino- phi1 infiltration in the stroma and the malignancy of the tumor. The significance of eosinophil infiltration is not understood, and the results of the present study do not seem to throw any new light on the subject. Certainly they do not support the view that its presence is a sigi of favorable prognosis (78), nor the popular theory that its presence is due to hemorrhage in the tumor. Such infiltration does often represent a late manifestation of the inflam- matory reaction, but even here we do not know what are the factors in the body which determine the proliferation of the eosino- phils. CARCINOMA OF THE PENIS AMONG THE CHINESE 275
Imvasion of Corpora Cavernosa: Among 84 cases examined, invasion of the corpora cavernosa by the tumor tissue occurred in 45. Invasion takes place more readily in tumors of higher malignancy, even though the average duration is shorter than in the more benign growths. Invasion of the corpora cavernosa is usually checked by the thick sheath surrounding the spongy tissue. After penetrating through the tunica, the tumor spreads either in the tissue (49) or in the blood spaces (53). In our cases, spreading in the tissue, with compression of the blood spaces, was the more common. Extension in the tissue takes a longer time to involve the whole penile shaft than does extension by way of the blood spaces, which results in a rapid involvement of the entire shaft. In such cases priapism is likely to occur, due either to the occlusion of the blood spaces by solid tumor masses (21) or to thrombosis (36). Pria- pism in one of our cases is explained on the former basis.
MODESOF EXTENSION In addition to the usual method of direct spread, extension of carcinoma of the penis is accomplished by invasion of the lym- phatics, or of the blood stream, or of both. Evidence of lymphatic invasion was present in 21 of our cases, of blood invasion in 4, and of both in 3. Of the 7 cases in which the blood stream was invaded, blood vessels were involved in 2 and the cavernous spaces in 5. In the lymphatics the tumor may spread in two ways: by direct extension (so-called lymphatic permeation) and through embolus formation. In direct extension a hard cord of tumor tissue may be present on the dorsal surface of the penis along the median line, marking out the main dorsal lymphatic trunk (5, 37, 49, 74, 95). We agree with Kuttner (52) that such a finding is rare, not a single case of this type appearing in our series or in Kiittner’s. On the other hand, tumor emboli in the lymphatics are frequently encountered and are, therefore, much more important. They give rise to isolated secondary tumor nodules a considerable distance from the primary growths. According to the statements of different investigators (23, 42, 52, 71), it is clear that the lymphatics of the foreskin, the glans penis, and the skin of the penile shaft empty directly into the inguinal lymph nodes. But there is a lack of agreement as to whether or not the lymphatics of the deeper portion of the penis empty directly into the pelvic lymph nodes. That some of the lymphatics actually drain directly into the pelvic nodes seems to be proved beyond doubt by the finding in them of metastatic penile tumors while the inguinal nodes remain uninvolved (28, 32, 52). Extension of penile carcinoma by way of the blood stream is 276 S. K. NOAI not a frequent occurrence. It usually follows invasion of the blood vessels or the blood spaces of the corpora cavernosa. Occasionally it follows the invasion of a large femoral vessel by a metastatic tumor in the femoral lymph nodes, in which case a widespread internal metastasis will occur (49, 50, 52, 54, 59, 77, 98). In one of our cases the metastatic tumor in a femoral lymph node almost eroded through the wall of the femoral vein. METASTASES Imguiltal Lymph Nodes: The inguinal lymph nodes were exam- ined in 52 cases, of which 35 showed no metastatic tumor and an enlargement due solely to inflammatory process. The other 17 cases, or 32.69 per cent, showedmetastatic tumor growths. Kiittner (52), in a series of 56 cases, stated that the inguinal nodes were enlarged in 71 per cent, although only 32 per cent were proved to be cancerous. Barney (5) stated that of a series of 66 cases, 75 per cent showed metastases. Kaufmann (49), in a series of 48 cases, found that 40 had carcinomatous metastases in the in- guinal nodes and that 30 out of these 40 were bilateral. Such wide variation in different reports regarding the frequency of inguinal node involvement is due probably to the difference in the clinical criteria by which the inguinal lymph nodes are removed. The results of correlative studies as to the frequency of meta- stasis and the physical character of the primary tumor (size, and grade of malignancy) are presented in Tables VIII and IX. While
TAB- VIII: Freipnq of MecaStaeiS in Relation lo Sits of Tumm
~ ~ ~ Size No Metastaeia Metastasis (+) Total
~ ~- Small 1C~(100%) 0 cases (0.00%) lam (100%) MediW 19 cam (67.86%) 9 casea (32.14%) 28 cam (100%) Large 8 cw(80.00%) 2 68888 (20.00%) 10 c88BB (100%) Unknown 7 caaea 6 casea 13 casea ~- TOTAL 35 cmee 17 CWB 52 cam
TAB- IX: Frequency of Melaekreie in Relation lo (3rd of Turn
~ ~ Grade No Metastasis Metastasis (+) Total I...... 21 cases (95.48%) 1 cm (4.64%) '22 cases (lOO~o) 11...... 10 cm(71.43%) 4 cam (28.67%) 14 cam (100%) 111...... :, ... 2 casea (25.00%) 6 cam (76.00%) 8 cam (100%) IV...... 2 casea (26.00%) 6 CBBBB (76.00%) 8 cam (100%) TOTAL,...... 36 cwea 17 case8 52 caaea
these studies permit no definite conclusion as to the relationship between the frequency of metastasis and the size or extent of the CARCINOMA OF THE PENIS AMONG THE CHINESE 277 primary growths, it is clearly shown that the frequency increases in the tumors of the more malignant grades, I11 and IV. As to the metastatic growths, it seemed of interest to determine whether they showed any deviation in structure from the primary tumor. Our study (Table X) indicates that, as a whole, the
Grade of Tumor Pearla in Tumor Primary Metastatic Primary Metastatic I 1 (deg.1 + - I1 I1 ++ +++ I1 I1 +++ +++ I1 (dq.1 +++ + I1 I1 +++ +++ I1 111 + - 111 IV + + 111 IV + + I11 I11 + + 111 IV + - I11 I1 +++ +++ 111 I1 +++ +++ IV IV +++ - IV I11 - + IV IV + + IV IV ++ - IV IV - - structural type in the metastatic tumor is fairly well preserved, although there is a tendency for it to be slightly more anaplastic and to show more retrogressive changes, such as degeneration and necrosis. Irzternnal Metastasis: Among the fatal cases in our hospital, two were examined at autopsy, but neither showed any internal meta- stasis. Barney (5) stated that over 15 per cent of his cases had metastases in the vital organs, this conclusion being based on clinical manifestations, but not on autopsy. Winiwarter (98) saw a case with one pulmonary tumor nodule ; Ricord and Ressel (77) saw a case with many nodules in the lungs; Lebert (54) reported 2 cases with metastasis to the lung and liver; Legallois (59) one case with metastases to the heart, lung, and liver ; and Kaufmann (49) one case with tumor metastases in the heart, lung, pleura, spleen, liver, and kidneys. Keller (50) found metastases in liver and peritoneum, and Kuttner (52) found a rather generalized metastasis in 2 cases. It aeems, then, that internal metastasis is not a common complication in carcinoma of the penis. Testis: Testes were examined in 26 cases. In 25 cases they were normal; one case showed fibrosis and atrophy of the tubules. None showed metastatic tumor. 278 8. K. NGAI
GENERALSYSTEMIC EFFECITS OF GROWTH Loss of Body Weight: In 79 cases mention of the general sys- temic effect was made. Loss of body weight to the extent of emaciation occurred in only 7 cases, or 8.86 per cent, in our series, while 62 patients, or 78.48 per cent, suffered no loss of weight, and 10, or 12.66 per cent, only slight loss. This is in contrast to carcinoma of the gastro-intestinal tract, in which loss of body weight is frequent. BEood Picture: Routine blood counts were done in 80 hospital cases. From these studies, it may be said that moderate secondary anemia is present in only a small percentage of cases, but leuko- cytosis is more commonly seen. In practically all these cases both anemia and leukocytosis can be satisfactorily explained by the presence of varying degrees of infection. Urine: Routine urinary examination was done in 79 hospital cases of our series. To summarize the findings, the urine in these cases is usually of light yellow or straw color, rarely bloody, slightly cloudy or turbid, having a normal specific gravity, mostly acid in reaction, with albumin varying from a faint trace to two plus in about two-fifths of the cases, no sugar unless diabetes mellitus is also present, frequent pus cells, and not infrequently red blood cells in the sediment. The fact that over 70 per cent of our cases showed an acid reaction is probably of no significance, since litmus was frequently used in earlier years and acid reaction to this indicator was regarded as normal. The presence of pus cells and red blood cells in the sediment is not at all surprising, when an ulcerating tumor mass is present on the external genitalia. DEATH Of the patients who have been followed up, 11 have died: 6, or 54.5 per cent, died of postoperative complications, 4, or 36.4 per cent, of “recurrence” (?), and onb of “ulcer of the neck.” As the follow-up study of these cases is exceedingly dificult, the result obtained is not at all instructive. However, from the reports received, postoperative complications seem to occupy an important place in the cause of death. Intercurrent diseases, such as pneumonia, sepsis, etc., are frequent causes of death, although some authors (52) rank these after metastasis and cachexia. Death due to the latter cause is, in our opinion, rare. Copious hemorrhage may occur in cases of penile cancer, probably due to the erosion of the larger blood vessels, such as the femoral, and generally leads to a fatal outcome. Few cases of such nature have been reported (50, 52, 74, 98). CARCINOMA OF THE PENIS AMONG THE CHINESE 279 DISCUSSION In a consideration of the etiology of the present series of cases, one is struck by the fact that in practically every one there is a his- tory of phimosis, while in the literature no case of penile carcinoma has been reported in a circumcised Jew. Thus there is reason to believe that certain conditions associated with or resultant from phimosis may have some bearing in the production of the tumor. These conditions are (a) the retention of the urinary products, (b) the accumulation of desquamated epithelium and secretion of Tyson’s glands, and (c) lack of cleanliness of the preputial sac which favors (a) and (b). The urinary products usually contain, among other things, indol, skatol, indigo, and some lipolytic and proteolytic ferments. Although it is not definitely known whether the above mentioned substances are present in all of our cases, the importance of their presence must be recognized. In view of the fact that marked epithelial hyperplasia with downward growth into the stroma has been produced experimentally by subcutaneous injection of 5 per cent indol and 5 per cent skatol in rabbit fat (91),and by similar injection of indigo suspended in water or gum arabic (16), one must seriously consider the relationship between the presence of these substances in the preputial sac and penile carcinoma, espe- cially in view of the fact that these substances may be accumulated there for long periods of time. The secretion of Tyson’s glands may also have some direct influence on the production of tumor. This secretion, according to Lehmann (55), contains a high percentage of fatty material.’ It is unlikely that these fatty substances can themselves cause any important change in the epithelium, but under the action of the ferments and others, they may in time decompose, become rancid, saponify and produce other substances which may directly stimu- late epithelial growth. That this may be so is suggested by the formation of tumors under somewhat similar, though by no means identical, conditions, such as the production of carcinoma of the breast in animals following the ligation of the main ducts of the mammary glands (4). That the tumor is actually the direct result of the stimulating action of the accumulated or retained materials from the urine or Tyson’s glands is proved by the fact that practically all of the tumors have their origin in the epithelium directly exposed to the aIn analyzing human amegma, he found that it contains 52.8 per cent of ether extract, 7.4 per cent alcohol extract, 6.1 per cent water extract, 9.7 per cent salt, 5.6 per cent albuminate, and 18.4 per cent residue. Of the fatty material, the acid number is 18.4, eaponiflration number 142, and iodine number 41.2. The unaaponifled part con- tains very little cholesterin. 280 8. I(. NOAI smegma, arising from the glans penis, the inner surface of the prepuce, or the sulcus. There seems to be reason to believe that the effect of the smegma may be latent for a considerable length of time, which probably explains the finding of penile carcinoma in cases where circumcision has been done several months or years previous to the appearance of the new growth. The frequency of penile cancer in the Far East, as previously stated, is much higher than in Europe and America. Dormanns (29) explains this high frequency by the popular practice of ap- plying highly irritating plasters for the treatment of gonorrhea, chancre, and chancroid lesions. There is no satisfactory proof of this contention. In our patients, about 63 per cent denied gonorrheal infection and none of those admitting gonorrhea or syphilis mentioned the previous use of any plaster. The only explanation that we can offer is that the lack of personal hygiene and the limited practice of circumcision among Asiatics favor the prolonged irritating action of the smegma on the susceptible epithelium. It has been pointed out by various investigators, as well as by the present author, that cancer occurs earlier in Chinese than it does in Westerners. This may be a matter of racial difference, although the underlying cause is not clear to us.
It is a pleasure to acknowledge the author’s indebtedness to Dr. J. R. Cash, formerly Professor of Pathology, Peiping Union Medical College, for suggestion of this work; to Drs. C. H. Hu, J. W. Spies, F. R. Dieuaide, and W. P. Ling for valuable suggestions and aid in the preparation of this paper, and last, but not least, to the Department of Surgery of Peiping Union Medical College for per- mission to use a part of their clinical material.
BIBLIOGRAPHY 1. ALLENBACH,E. : Primiires Urethralcaroinom mit priapismusiihnlichen Folgen, Dtsche. Ztschr. f. Chir. 138: 152, 1916. 2. ANDREWB: Cited from Barney. 3. ABAHI,K. : On precancerous dermatosis, Jap. J. Dermat. 27 : 20, 1927. Abst. in Zentralbl. f. Hautkrkh. 26: 379, 1927-8. 4. BAQQ,H. J.: Experimental studies concerning the functional aotivity of the breast in relation to mammary carcinoma in mice, J. Cancer Rea. 9: 498, 1925. 5. BARNEY,J. D. : Epithelioma of the penis. AJI analysis of 100 cases, Ann. Surg. 46: 890, 1907. 6. BWINQER,B. S., AND DUN, A. L., Js. : Epithelioma of the penis, J. Urol. 11 : 497,1924. 7. BERCOVITZ,N. : Cancer in Hainan, China M. J. 34: 119, 1920. 8. BIEBL, M. : Seltene Form von Peniskrebs (papillares qlindromatiises Carci- nom. Beiteilung eines paraurethralen Ganges. Urspriinglische spitze Kondylome), Beitr. e. klin. Chir. 137: 228, 1926. CARCINOMA OF THE PENIS AMONG THE CHINESE 281
9. BILLROTH, TH. : Chirurgische Erfahrungen, Zurich, 1860-1867 : Epithelial- Carcinome des Penis, Arch. f. klin. Chir. 10: 549, 1869. 10. BRAABZADT,L. H.: Carcinoma, China M. J. 41 : 7, 1927. 11. BRODERB,A. C.: (u) Squamous-cell epithelioma of the lip, J. A. M. A. 74: 656, 1920. (b) Epithelioma of the gentto-urinary organs, Ann. Surg. 75: 574, 1922. (c) Practical poiiits on the microscopic grading of carcinoma, New York State J. Med. 32 : 667,1932. (Bull. Am. SOC. Control of Cancer, 14 : 1, 1932). 12. BRUCE,A. : Epithelioma of the penis, Trans. Path. Soc. London 19 : 288, 1868. 13. BRUCKHAUS,P. A.: Cited from Maresch and Chiari, p. 367. 14. BRWNING,M. : Beitrag zur Prognose des Carcinoma penis, Diss.-Qreifswald, 1892. Cited from Maresch and Chiari. 15. BUDAY,K.: Beitriige zur Kenntniss der Penisgeschwiilstc, Arch. f. klin. Chir. 49: 101,1895. 16. BULLOCK,F. D., AND RQHDENBURG,O. L.: A study of Scharlach R reaction and of allied forms of epithelial proliferation, J. Med. Research 33: 53, 1915. Experimental “ carcinomata ” of animals and their relation to true malignant tumors, J. Cancer Research 3: 227,1918. 17. BUBMAN,~O.J., AND WOODBURNE,A. R.: Paget’s discase of the glans penis with central carcinomatous degeneration, Arch. Dermat. & Syph. 24 : 396, 1931. 18. CALEF,C.: Tre casi di epitelioma del pene, Riv. osp. 15: 513, 1925. Abst. in Zentralbl. f. Chir. 53: 1665, 1926. 19. CORUZZI,C. : Su un caso di carcinoma del pene, Policlinico (sez. prat) 33 : 1285, 1926. Abst. in Zentralbl. f. Hautkrkh. 23: 143, 1927. 20. CRAYER,W.: On experimental carcinogenesis: The local resistance of the skin to the development of malignancy, Brit. J. Exper. Path. 10: 335, 1929. 21. CREITE: Peniscarcinom bei einem 2 jahrigen Kinde, Dtsche. Ztschr. f. Chir. 79: 299,1905. 22. CUBERO:Epitelioma de glande sobre leucoplasie, Actaa dermo-sif. 16: 84, 1924. Abst. in Zentralbl. f. Hautkrkh. 16: 272, 1924-6. 23. CUNNINQHAM,J. H., JR.:The operative treatment of carcinoma of the penis, Surg. Gynec. Obst. 19: 693, 1914. 24. DAYAL,S. : A case of epithelioma of the penis, Indian Med. Oaz. 59: 89,1924. 25. DE HABB: Twee gevallen van penis-carcinom, Oeneesk. ti jdschr. v. Neder1.- Indie 71: 1075, 1931. Abst. in Am. J. Cancer 16: 680,1932. 26. DE VOQEL,W.: Le cancer parmi lea ouvriers Chinois des mines d’6tain dam l’ile de Biliton (Indes Nberlandaises), Bull. de 1’0fice Intern. d’Hyg. 20: 1973, 1928. Cited from Wolbarst. 27. DEMARQUAY:Maladies chirurgicales du phis, Paris, 1877. Cited from Kiittner. 28. DON,A. : Cancer of the penis and its extirpation, Edinburgh M. J. 3: 14, 1909. 29. DORMANNB,E. A.: Ober das Peniscarcinom und seine besondere Hauflgkeit in Ostasien, Ztschr. f. Krebsforsch. 29: 435, 1929. 30. DRESOHER:Peniskarzinom, Schles. dermat. ges. Setzz. 14, 11, 1925. Abst. in Zentralbl. f. Hautkrkh. 17 : 270, 1925. 31. DUPUYTREN: Cited from Kaufmann. 32. ENCLISCH: Zum Peniscarcinom, Wien. klin. Wchnschr. 15: 430, 1902. 33. EWINQ,J.: Neoplastic Diseases, W. B. Saunders, Philadelphia, 1928, 3d ed. 34. FODERL,V. : Znr Klinik und Statistik des Peniskarzinomes, Deutsche Ztschr. f. Chir. 198: 207, 1926. 35. FORSTER:Cited from Maresch and Chiari, p. 367. 282 8. K. NOAI
36. FRON~,W. A., AND ALYEA,E. P.: Priapism of unusual etiology, J. Urol. 20: 135,1928. 37. FRIEDRICH,W.: Ein Fall von Penieaarcinom, Diss., Munchen, 1910. Cited from Maresch and Chiari. 38. GAROPALO,F.: Priapism0 da tumore metastatic0 dei corpi oavernosi del pene, Arch. ital. di dermat. e sif. 7: 20, 1931. ‘Abst. in Am. J. Cancer, 16: 446, 1932. 39. GENNER, V., AND NIELSEN,J.: Leukokeratosis and kraurosis of the penis, Acta Dermato-venereol. 12: 195,1931. Abst. in Cancer Rev. 7 : 313,1932. 40. QILLIAM,D. B.: Carcinoma of the Penis with Report of Two Cases, Urol. & Cutan. Rev. 28: 390, 1924. 41. GUELLIOT:Union Med. du Nord-Eat. 16: 106, 1801. 42. GUY, H.:Anatomy, Lea & Febiger, Philadelphia, 1913. 43. GREEN,T. M. : Charlotte Med. J. 30: 182, 1907. Cited from Leighton. 44. GREEN,L. : Three cases of quamous-cell epithelioma of the penis, J. Urol. 13 : 598, 1925. 45. QUEENSFELDER:Internat. J. Surg. 27: 163, 1914. Cited from Leighton. 48. HEIMANN,Q.: Die Verbreitung der Krebserknmkung, die Hauflgkeit ihres Vorkommeris an den einzelnen Korpertheilen und ihrer ohirurgischen Behandlung, Arch. f. klin. Chir. 58: 31, 1899. 47. HEY: Cited from Maresch and Chiari, p. 367. 48. IWASAKI,K. : Leukoplakia penis mit carcinomatoser Umwandlung, Deutsche Ztschr. f. Chir. 119: 127, 1912. 49. JCAUPMANN,C. : Verleteungen und Krankheiten der mannliohen Harnrijhre und des Penis, Deutsohe Chir., Lief. 50a: 263, 1886. 50. KELLER, H.: tfber die Amputation des miinnlichen Gliedes und die Anwendung veraenkter Etagenniihte hei derselben, Beitr. e. klin. Chir. 4: 254, 1889. 51. KR~NLEIN:Cited from Martin. 62. K~~TNER,H.: Zur Verbreitung und Prognose des Peniscarcinoms, Arch. f. klin. Chir. 59: 181, 1899. tfber das Peniscarcinom und seine Verbreitung auf dem Lymphwege, Beitr. e. klin. Chir. 26: 1, 1900. 53. KYRLE,J.: Histo-Biologie der menschlichen Haut und ihrer Erkrankungen, J. Springer, Berlin, 1925, Vol. 1, p. 15. 54. LEBERT:Cited from Kaufmann. 55. LEHMANN,C. 0.: Ber. sicha. Akad. Wiss. 2: 1889. Cited from Oppen- heimer, C.: Handhuch der Biochemie des Menschen und der Tiere, Jena, 1925, Vol. 5, p. 414. 56. LEIOHTON,W. E.: Carcinoma of the penis, with a report of 67 cases, Am. J. Cancer 16: 251, 1932. 67. LEROYDES BARRES,A., AND HEYMANN,P.: Le cancer de la verge au Tonkin, J. radiol. et cleotrol. 11 : 89, 1927. Abst. in Zentralbl. f. Hautkrkh. 25 : 379, 1927-8. 58. LEWIS,L. G.: Young’s radical operation for the cure of cancer of the penis. A report of 34 oases, J. Urol. 26: 295,1931. 59. LEGALLOIS,E. : Cancer de Ia verge, Bull. SOC. Anat. Pans. 6 : 99, 1830. 60. Lousm, CAILLAUD,AND MARASSI: Un cas d’Brythroplasie g6nitale d’apparence chancriforme aveo BpithBlioma, Bull. 800. franc. derm. et syph. 32: 227, 1925. Abet. in Zentralbl. f. Hautkrkh. 17: 927, 1925. 61. WWENTHAL,C.: tfber die traumatisohe Entstehung der Qeschwulste, Arch. f. klin. Chir. 49: 1, 1895. 62. MAJANZ,A. J.: tfber kareinomiihnliche Erkrankungen des Penis, Ztschr. f. Urol. 22: 620,1928. CARCINOMA OF THE PENIS AMONG THE OHINESE 283
63. MANNTEUFEL:Ein Beitrag zur Statist& des Penis Karzinoms, Diss., Breslau, 1901. Cited from Maresch and Chiari. 64. MARESCH,R., AND CHIARI,H.: Krebs des Gliedes, in Handbuch der speziellen pathologischen Anatomie und Histologie (Henke, F. und Lubarsch, 0.) 1931. Vol. VI/3, p. 352. 65. MARTIM,E.: Epithelioma of the penis, J. Cut. & Gen.-Urin. Dis. 13: 95, 1895. 66. MAXWELL,J. L. : Cancer among the Chinese, China Med. J. 42 : 69, 1928. 67. MAOFARLAND:Cited from Martin. 68. MENDELEON,R. W., AND ELLIS,A. 0.:Cancer as a public health problem in Siam, J. Trop. Med. 27: 274, 1924. 69. MURPHY,J. B., AND OTHERS: Studies on x-ray effects. VI. Effect of oellular reaction induced by x-rays on cancer grafts, J. Exper. Med. 33 : 299, 1921. MURPHY,J. B., AND STURY,E. : Conditions determining the transplantability of tissues in the brain, J. Exper. Med. 38: 183, 1923. 70. NOBUJI, H.: Carcinoma penis, Inaug. Diss., Munchen, 1909. Cited from Maresch and Chiari. 71. NICOLL, J. H.: The operative technique of caroinoma of the penis, Ann. Surg. 49: 240, 1909. 72. OPPENHEIM,M. : Krebsentwicklung des Priiputiums bei einem Metallschleifer, wahrscheinlich durch Schmierol bedingt, Wien. klin. Wchnschr. 42 : 249, 1929. 73. PACHA,D.: Cited from Kretschmer, H. L.: Surg. Clinic. 2: 269, 1918. 74. PARTSOH: Das Carcinom und seine operative Behandlung, Kohler, Breslau, 1884. Cited from Maresch and Chiari, and Kiittner. 75. PATTEXSON:Univ. of Penn. Med. Bull. Vol. 14, 1901-2. Cited from Barney. 76. QUEYRAT:Cited from Louste. 77. RICORDAND RE~SEL:Arch. gbn. de m6d. 1: 560, 1864. Cited from Kiittner. SOC. d’anstomie. 1866, p. 320. Cited from Kaufmann. 78. ROOADE VIRALE, R.: Epithelioma penis con eosinofilia local y circulante, An. Hosp. de Santa Crue y San Pablo. 5: 168, 1931. Abst. in Am. J. Cancer 16: 680, 1932. 79. ROFBY),A. H.: Erythroplaaie der Eichel, Rev. Dermat. 11: 315, 1926. Abst. in Zentralbl. f. Hautkrkh. 23: 459, 1927. SO. SAINZDE AJA, E. ALVAREZ:Sindrome precanceroso del pene, Actas dermo-sif. 18: 146, 1926. Abst. in Zentralbl. f. Hautkrkh. 22: 137, 1927. 81. SAYPOERNO,F. : Peniskarzinom und Beschneidung, Deutsche Ztscbr. f. Chir. 201: 282, 1927. 82. SAWTELL&H. W.: Epithelioma of the penis, New York M. J. 57: 527, 1893. 83. SOHIOK,F. : Zur galvanokaustischen Amputation des Penis, Beitr. z. klin. Chir. 9: 346, 1892. 84. SCHMIDT:Zur Atiologie des Carcinoma Penis, Dim.-Erlangen, 1889. Cited from Maresch and Chiari. 85. SOHUCHARDT,K. : Beitriige zur Entstehung des Carcinome aus chronisch Entziindlichen Zustiinden der Schleimhaute und Hautdecken, Volkmann’B Samml. klin. Vortr. No. 257: 16, 1885. Cited from Kaufmann. 86. SCHULTZE,W. H.: Cited from Maresch and Chiari, p. 376. 87. SCOMAZZONI,T.: Contributo alla conoscenza del morbo di Bowen del glande. Gior. ital. dermat. 68: 1505, 1927. Abst. in Zentralbl. f. Hautkrkh. 26: 863, 1928. 88. SIBLEY:Cited from Kaufmann. 89. SHERRILL,J. G. : Carcinoma of the penis : Report of a case well 14 years after operation, Am. J. Surg. 38: 36, 19’24. 90. SHIELD,A. M.:Cancer of the penis, Lancet, 1: 75, 1900. 284 8. K. NGAI
91. STOEBER,H., AND WAOICER,L. : Ein weiterer Beitrag eur Erzeugung atypischer Epithelwucherungen mit Eiweissfaulnisproduen, Miinchen med. Wchnsohr. 67: 947, 1910. 92. SUBMAN,M. P. : “ Paget’s disease ” of the glans penis, Brit. J. Surg. 15 : 635, 1928. 93. SZATHMARY,S.: Morbus Bowen genitalis, Zentralbl. f. Hautkrkh. 28: 24, 1929. 94. TAEGTME~:Ein Beitrag eur Lehre vom Carcinoma penis. 1naug.-Diss. Gottingen, 1904. Cited from Maresch and Chiari. 95. THOMSEN,A. : Carcinoma of the penis, British M. J. 2 : 1W, 1897. 96. Tsav~~s:Case of cancer of the penis commencing with paraphimosis, Lancet 1: 182, 1827. 97. VILLIER:Cited by Englisch in Eulenberg‘s Raalencyclopiidie 16 : 317, 1888. 98. WINIWARTFX:Beitriige mr Statist& der Karzinome, Stuttgart, 1893. Cited from Maresch and Chiari. 99. WOLBARST,A. L.: Is circumcision a prophylactic against penis cancer? Cancer 3: 301, 1926. Abst. in Zentralbl. f. Hautkrkh. 22: 906, 1927. Circumcision and penile cancer, Lancet 1: 150, 1932. 100. WOGLOM,W. H. : Experimental tar cancer, Arch. Path. 2 : 709, 1926. 101. YOBHIDA,S.: Ein dritter Fall von Pagetscher Krankheit am Penis, Acta Der- mat. 8 : 867, 1926. Abst. in Zentralbl. f. Hautkrkh. 23 : 460, 1927. Ein 2 Fall von Pagetscher Krankheit am Penis, Acta Dermat. 7: 772, 1926. Abst. in Zentralbl. f. Hautkrkh. 23: 460, 1927. 102. YOUNQ,H. H.: Practice of Urology, W. B. Saundere Co., Philadelphia, 1926. A radical operation for the cure of cancer of the penis, J. Urol. 26: 285, 1931. 103. ZIELEWICZ, J. : tfber die Amputation des Penis mit der galvanocaustischen Schneideschlinge, Arch. f. klin. Chir. 12: 580, 1871.