Cancer of the tonsil

CANCER OF THE TONSIL

HAYES MARTIN, M.D.

Attending Surgeon, Memorial Hospital

AND

EVERETT L. SUGARBAKER, M.D.

RockefelIer FelIow in Cancer Research, Memorial HospitaI

NEW YORK, NEW YORK

ANCER

C of the tonsi represents about the average of the

severa anatomic forms of maIignant growths of the pharynx from the standpoints of frequency and marignancy. Tonsillar

tumors are of particuIar interest since they comprise a we11 defined anatomic and histoIogic group in a region which is readily accessibIe to either surgery or radiation, so that the possibiIities of these two methods in this area may be fuIIy expIored and compared.

The present report is based upon 157 unseIected consecutive cases of cancer of the tonsi1, including all patients in all stages of the disease who appIied to the Memorial HospitaI during the years 193 I to 1935, incIusive. Duffy l5 has reported a preceding series (1920 to 1930) from our cIinic.

Definition. In the present report the term, “cancer of the tonsil,” designates those growths which arise in the paIatine tonsil itself, at its periphery in the tonsiIIar fossa, in the tonsiIIar piIIars (gIossopaIatine arch, pharyngopaIatine arch), or in the pIica tri- angularis. The term must not be confused with “cancer of the nasopharynx ” (pharyngea1 tonsi or adenoid) or “cancer of the

IinguaI tonsil ” (base of the tongue). Anatomy of the Tonsil. The palatine tonsiIs are two large

ovoid masses of Iymph tissue which are embedded in the side waIIs of the ora part of the pharynx between the gIossopaIatine and pharyngopaIatine arches. The Iatter two structures and the adjacent edge of the tongue outIine a trianguIar area with its apex above, which is known as the tonsiIIar fossa. For purposes of cIassification, the tonsi shouId be considered a pharyngea1 structure, the Iine of demarcation between the ora cavity and the pharynx being the free edge of the soft paIate and the anterior tonsiIIar pillar. The tonsi occupies the inferior position in this triangIe, and above it near the

MARTIN, SUGARBAKER-CANCER OF TONSIL 159

apex is the fossa supratonsiIIaris. From the lower margin of the anterior tonsiIIar piIlar, a foId of mucous membrane, caIIed the plica trianguIaris, passes backward between the tonsi and the edge of the tongue.

I \ I nt- car&-id

Ext. carotid

/ JUqU laf-/

Chaih

tonsil

FIG. I. The lymph drainage of the palatine tonsil. From the

paIatine tonsil the Iymph vessek pass directIy outward

to the subdigastric Iymph nodes of the jugular chain

which are invoIved first in practicaIIy a11 cases of metastasis from the tonsil.

The tonsils themseIves consist of masses of Iymphoid foIIicIes with a deIicate connective tissue reticuIum. MesiaIIy the surface is covered by a thin Iayer of mucous membrane and is very irreguIar because of deep crypts formed by infoIding of the epithelium. On the Iateral surface the Iymph tissue is invested with a rather firm connective tissue capsuIe which Iies directIy on the superior constrictor of the pharynx.

Z%e Lymphatic Drainage of tbe Palatine Tonsil. The coIIecting trunks of the paIatine tonsiIs are a part of a network which drains the entire pharynx. (Fig. I .) They perforate the capsule of the tonsi and the superior constrictor, passing successiveIy behind the styioglossus, the styIohyoid and the posterior 6eIIy of the digastric muscIes to reach the externa1 carotid artery. Most of the trunks then

a4

160 MARTIN, SUGARBAKER-CANCER OF TONSIL

pass behind the externa1 carotid artery and end in the upper, deep cervicaI group of the juguIar chain of Iymph nodes. The principIe node of this group lies just above the IeveI of the bifurcation of the common carotid artery and is generaIIy caIIed the subdigastric or tonsiIIar node. From this node, the Iymph drainage flows to the middle and thence to the inferior nodes of the juguIar chain. It seems probabIe that there are onIy a few by-passes in this area, for metastases from the tonsil seIdom appear first beIow the IeveI of the subdigastric node.

ETIOLOGY

General Incidence. From an anaIysis of the admission records of the MemoriaI Hospital, cancer of the tonsi comprises 8 per cent of a11 cancer of the upper respiratory and ahmentary tracts and about 2 per cent of a11 human cancer. Of the structures of the pharynx, the tonsi (21 per cent) is second onIy to the extrinsic Iarynx (37 per cent) in frequency as the site of origin of malignant growths. These figures do not agree with those of Despons13 who stated that the tonsi is the most frequent site of pharyngea1 cancer. It is diffIcuIt to expIain the findings of SchaI144 who reported that at the CoIIis B. Huntington HospitaI cancer of the tonsil made up 9 per cent of 24,437 cases admitted for aII forms of cancer.

Age and Sex. In the present series the average age was about fifty-seven years on admission and about 30 per cent of the patients were in the sixth decade. These figures are almost identica1 with most other anatomic forms of cancer of the upper respiratory and aIimen- tary tracts in our cIinic. The oIdest patient was ninety-four, the youngest eIeven. As one of us (M.)2g” has previously reported, cancer in children is not uncommon in the nasopharyngea1 tonsi but is apparentIy rare in the paIatine tonsi1.

In the present group, 86 per cent of the cases occurred in maIes and 14 per cent in females, a sex distribution which is almost identical with that of cancer of the tongue as observed in our clinic. There was no significant difference in the ages of maIes and females. AI1 ob- servers have reported a preponderance in maIes of cancer of the tonsil: Despons 92 per cent,13 Nathanson 93 per cent,38 SchaII 8g per cent44 and Mattick 91 per cent.31

Position of the Growth. Since the primary Iesion is usuaIIy Iarge when first examined, the exact site of origin within the tonsiIIar fossa cannot be determined in a11 cases. In over go per cent of the

f[3n

MARTIN, SUGARBAKERXANCER OF TONSIL 161

present series, the diameter was more than 2 cm. on admission, with an average of 4 cm. for the whole group. Less than 3 per cent of the tonsiIIar growths were I cm. in diameter or Iess and couId be cIassified definiteIy as “early.” The right side was invoIved pri- marily more often (60 per cent} than the Ieft (40 per cent). Such an irreguIar distribution has been noted previousIy by SchaI144 who found 55 per cent on the right and Duffylj who found 57 per cent on the left. In our chnic, several forms of intraora1 cancer have displayed simiIar inequalities in distribution; but since the dispro- portion aIways tends to be greater in the smaIIer series, we believe that it can be accounted for entireIy by the Iaws of chance and, therefore, that it has no cIinica1 significance.

Causative Factors. In growths of the paIatine tonsil as in other forms of pharyngea1 cancer, there appears to be no outstanding etioIogic factor. Obviously the pharyngeal waIIs are not subject to the trauma of mastication nor to irritation from sharp or irreguIar teeth, which pIay definite raIes in the causation of cancer of the ora cavity itseIf. Hot foods and drink, which are heId at Ieast momen- tariIy in the ora cavity, pass rapidly through the pharynx during the act of swaIIowing. Even when smoke is inhaIed, the irritating effects of tobacco are confined IargeIy to the oraI cavity. In consider- ing others possibIe etioIogic factors, repeated attacks of acute or chronic tonsihitis wouId appear, theoreticaIIy, to be a IikeIy form of chronic irritation in this area, but in our cases, as in the series reported by Berven, 2 the histories failed to revea1 any unusual tendency in this regard. The diffuse stomatitis CharacteristicaIIy found in avitaminosis, syphiIis, the PIummer-Vinson syndrome and the anemias, may extend to the pharynx and might aIso theoreticaIIy play a part in the etioIogy of tonsiIIar cancer, aIthough we have no supporting statistica evidence. LeukopIakia, though occasionaIIy found on the paIate and anterior tonsiIIar piIIars, has never in our experience occurred on the tonsiIs, but CiteIIig has reported two cases of precancerous IeukopIakia on the tonsiIs themselves.

In our series, about 70 per cent of the patients admitted the use of tobacco, usuaIIy smoking; but as we have previousIy pointed out in discussing other forms of intraora1 cancer, at least this percentage of addiction is found in norma maIe aduIts of corre-

sponding age. It seems to us that practicaIIy a11 statistical data offered in proof of a connection between smoking and intraoral cancer remain inconcIusive. NevertheIess, in our cIinic isolated cases

%4

I62 MARTIN, SUGARBAKER-CANCER OF TONSIL

of cancer of the base of the tongue, the tonsil and the soft paIate occur with sufficient frequency in heavy cigar smokers (ten to fifteen cigars daily) to Iend strong support to the theory of a direct causa1 reIation.

In our group, only 3 per cent of the Wassermann tests taken gave positive reactions, a figure which is identica1 with the findings of SchaII.44 This proportion is Iess than the average for the male popuIation of corresponding age (6 per cent).j2 Since the number is not Iarger, it is undoubtedIy onIy a coincidence that the percentage of syphilis in our series of tonsiIIar cancer is less than the genera1 average. In any event, it is obvious that syphiIis is not a factor in the etioIogy of cancer of the tonsil.

Poor dentaI and ora hygiene was observed in the majority of our patients; but since most of them were from the Iess fortunate economic group, such deficiencies are to be expected and, in the present series, were no more prevalent than among the skin cancer patients in our chnic.

SYMPTOMS, MORBID ANATOMY AND CLINICAL COURSE

In our series the average duration of symptoms (usuaIIy pain or “ soreness ” of the throat) before admission was seven months, as compared to fifteen months in cancer of the Iip, ten months in cancer of the nasopharynx and five months in cancer of the tongue and of the floor of the mouth, respectiveIy. Berven2 reports an interval of four months between the beginning of symptoms and admission to the Radiumhemmet, whiIe Nathansons in his series reports an average delay of one year. In interpreting such figures it is essentia1 to distinguish between the duration of the symptoms and the possibIe duration of the disease itseIf. In growths of such areas as the lip or anterior portion of the tongue, where the tactile sense is acute and the areas are under daiIy observation by the patient, the duration of the disease and the duration of the symptoms are more nearly equal. In the tonsihar area, on the other hand, there is IittIe or no tactile sense and the discovery of the primary Iesion by the patient must ordinarily await the deveIopment of pain, which may be Iong deferred. There is no way to determine how Iong the growth existed in these cases before pain f&t deveIoped.

In contrast to growths of the ora cavity itself, the most common first compraint in our series of tonsihar cancer was IocaI pain or sore throat (60 .per cent). This finding might at first suggest that

us1

MARTIN, SUGARBAKER-CANCER OF TONSIL ‘63

this is an earIy symptom, except for the fact that the average diameter of the primary Iesions was almost 4 cm. on admission. An ulcerated infected Iesion of the tonsi of this size obviousIy must be painfu1, whether or not it produces any other subjective symptoms. OccasionaIIy the growth in the tonsi is associated with such a degree of Iocalized sepsis or ceIIuIitis that the patient complains of pain radiating to the ear before admission, a symptom which is char- acteristic of advanced and uncontroIIed stages of this disease.

After pain in the throat, the next most common symptom was the discovery of an enIarged Iymph node in the upper deep cervical region (33 per cent in this series). In forty-four cases in which the patients compIained onIy of enlarged cervica1 nodes, the primary lesions discovered first by the examining physician averaged 3.5 cm. in diameter. In one of these cases the metastatic node (proved by aspiration biopsy) was treated and permanentIy controIIed by irradiation, and the patient was observed at monthIy intervaIs for aImost three years, with repeated examinations of the upper respiratory and abmentary tracts, before the primary lesion {about 5 mm. in diameter) was discovered in the tonsi and proved by

biopsy. The tactile sense of the pharynx is slight in comparison to that

of the oral cavity itseIf. In Iess than IO per cent of our cases did the patients state that they first became aware of a “lump in the throat,” a disturbance in swaIIowing or a cough. The primary Iesions in this small group were of about average size (3.5 cm.). In three cases of more buIky primary Iesions (over 4 cm. in diameter) the first compIaint was of earache, a we11 known form of radiating pain in inflammations of the throat. A few patients first noted such vague symptoms as cough, hoarseness, nosebleed, etc., too varied to be considered of real significance. It is probable that cancer of the tonsi must aIways reach a size of at Ieast I .5 to z cm. before sufficient infection occurs to cause pain or soreness or any other IocaI symptom. In our series the patients with Iesions Iess than r.5 cm. in diameter sought medica advice onIy because of enlarged cervica1 nodes, and the primary Iesions were discovered frrst by the examining physician.

An earfy tonsiIIar growth usually consists of a superficia1 granuIar ulcer either in the tonsil itself or in the groove between the tonsi and one of the pilIars. (Fig. 2~2 and b.) The Iesion may arise in a crypt and present onIy a smaI1 portion on the surface. As it enIarges the


tumor usually tends to fungate from and erode the surface of the ton&I, aIthough occasionally, as in lymphosarcoma, it may invade deeply and produce a buIky submucous mass with little or no surface ulceration.

b

c d FIG. 2. Cancer of the tonsil begins either in the tonsi itself (a) or on one of the tonsilIar

piIlars (b). SmaII Iesions (a and b) or moderately advanced lesions (c) are the exception.

In the average case the growth on admission has a diameter of about 4 cm. (d), so that

the exact site of origin cannot be determined, and in these cases the growth has usuaIIy invaded the soft paIate, both tonsillar piIIars and the adjacent edge of the tongue.

By the time the Iesion has reached a diameter of about 2 cm., beginning invasion of neighboring structures has taken pIace, usuaIIy in about the following order: the tonsiIIar piIIars, the soft paIate, the base of the tongue, the pharyngea1 walls, the aIveoIar ridges and the mucosa of the cheek (Fig. zd.) In its advanced stages the disease may invoive the extrinsic Iarynx or the ffoor of the mouth and consist of a bulky, irreguIar, partIy necrotic tumor, 3 to 6 cm. in diameter. At this stage intractabIe pain irradiating to the ear is a fairIy constant symptom with trismus as well as dysphagia and pain on movement of the tongue. Hemorrhage due to erosion by the tumor

U7R

MARTIN, SUGARBAKER-CANCER OF TONSIL r65

it.seIf or as the result of radionecrosis is a frequent Iate symptom. Cervical metastasis is practicaIIy always a part of the clinical picture even in the earIier stages of the disease. If the growth is uncontrolled, death occurs from a combination of IocaI causes: exhaustion from pain, sepsis, malnutrition, etc. In the present series, the average Iength of Iife in the unsuccessfully treated patients was about nineteen months. Other authors report survival periods varying from six to twenty-eight months: Berven,* Cristo- fordis, Despons l3 and New. 3g

i%fetastases. In cancer of a11 parts of Waldeyer’s tonsiIIar ring, metastases piay an early and prominent r6Ie. In the present series about one-third of the patients noted enIarged cervica1 nodes as the first symptom. On admission cIinicaIIy demonstrable cervical metastases were present in 76 per cent of the cases and subsequentIy developed in an additional 3 per cent.* These figures are aImost identica1 with those of Mattick. Other reports in the Iiterature vary from 65 per cent to 92 per cent: Burnam, Scha11,44 Costolow,‘” Leviez7 and New. 3g In thirty-six cases (25 per cent of the determinate group) there was no cIinica1 evidence of metastasis on admission, and the disease was permanently controlled. In seventeen cases (I I per cent of the determinate cases) the patients died of uncontroIIed cancer in the tonsil without the development of metastases after an average Iength of life of nineteen months.

* The question might be raised as to the vaIidity of the diagnosis of cervica1 metastasis in this series. At the present time, on the Head and Neck Service at the Memorial

Hospital al1 cases of cIinicaIIy demonstrabIe cervica1 metastases are subjected to aspira-

tion biopsy as a matter of routine. In the earlier cases of the present series, not a11 cIinicalIy positive nodes were checked by this means, ahhough of the whole group of

I r7 cases, in seventy-five (64 per cent) the diagnosis was confirmed by aspiration biopsy

or histologic examination of excised nodes. In forty-two cases histoIogic confirmation

was not obtained, and in this group only couId there be any doubt. The possible margin of error can be estimated by a comparison of the average accuracy of cIinica1 diagnosis in

our ciinic from date in another series.

Between the years rg3o and rg38, incIusive, 144 neck dissections were performed under the cIinica1 diagnosis of metastatic cervical cancer (a11 neck dissections for primary cancer of the submaxiIIary glands, for carotid body tumors, and for benign diseases such

as tuberculosis are excIuded). In onIy tweIve (8 per cent) of these did the histoiogic diagnosis fai1 to support the clinica diagnosis of metastatic carcinoma. Since in this operabIe group the disease was earIy, the chance or error would have been greater than in the

average case of metastatic cancer from a primary Iesion in the tonsi1, which is usuaIIy far advanced and readiIy diagnosed cIinicaIIy. If we concede a possibIe error of 8 per cent in the cIinica1 diagnosis of forty-two unproved cases (or three cases), the total number with actuaI cervica1 metastases wouId stiI1 be I 14, a variation from our reported figure of I 17 which is too smaII to affect seriously the vaIidity of our caIcuIated figures.


The first node involved (in about 95 per cent of all cases) was the subdigastric, which Iies in the upper deep cervical or jugular chain just above the bifurcation of the common carotid artery. Occasion- alIy the first paIpabIe metastatic node appeared in the submaxiIIary region and, more rarely, in the middIe and lower parts of the deep cervical chain. Such variations are probabIy due to anomaIies in the cervical lymph system, so that the main Iymph stream is diverted past the upper deep cervica1 nodes through by-passes. After invoIve- ment of the subdigastric node, the disease tends to progress down the jugular chain to the clavicle and often into the posterior triangIe of the neck aIong the accessory chain or sometimes into the submaxiI- Iary and submental lymph nodes.

When the main lymph drainage channeIs are bIocked by radiation fibrosis or by surgica1 excision, subsequent Iymph drainage must take pIace by the way of anastomoses, usuaIIy through the superficia1 lymphatics and finaIIy across to the other side. Miliary metastases over large areas of the superhciaI Iymphatics of the neck and cheeks probabIy always take pIace in this manner. We have observed this phenomenon in a number of instances of cancer of the oraI cavity and pharynx. In one patient of the present series, a metastatic node in the upper deep cervical region appeared to regress compIeteIy with the primary lesion under radiation therapy. About three months Iater there was a recurrence in the tonsil which aIso responded to irradiation treatment. SeveraI months Iater a subcutaneous metastasis appeared in the scaIp near the occiput, foIIowed shortIy by another subcutaneous metastasis on the other side of the occiput with a Iater spread to the upper deep cervical nodes of the side opposite to that of the primary Iesion and the original metastasis.

In nineteen cases {about 15 per cent of those which showed metastases) there were biIatera1 metastatic nodes on admission, and in one additional case biIatera1 involvement deveIoped Iater. One of these patients, in whom the bilateral involvement was proved by aspiration biopsy, survived free of disease for more than five years.

It is a curious fact that dissemination beIow the cIavicIe to the viscera appears to occur onIy rarely. In the autopsy records of the Memorial Hospital there are twenty-nine cases of patients dead of tonsilIar cancer with onIy three instances of visceral metastases (mediastinum, stomach, bones). Despons l3 has aIso caIIed attention to the rarity of visceral metastases in cancer of the tonsi1; and we b&eve the onJy reasonable expIanation of this fact is that patients

u9ll

MARTIN, SUGARBAKER-CANCER OF TONSIL I67

with this form of cancer do not survive long enough, on the average, for the wide dissemination of the disease, although the histoIogic type of cancer in the tonsi would favor it.

TABLE I

HISTOPATHOLOGI’ IN CANCER OF THE TONSLL

Epidermoid carcinoma_ .......................... Squamous ceI1, grade I ...... .................... ... Squamous ceI1, grade rr ............... ........ ........ Squamous cell, grade III ... ...................... .... Squamous ceII, grade rv .............. .................

Squamous cell, ungraded. .............................. Transitional ceII, grade II .............. ........... .... Transitional cell, grade III ..... ................. ...... Transitional ceIi, grade IV ........................... ..

Lymphoepithehoma ...................................... Lymphosarcoma, ......... .......... .............. .... UncIassiIied”. ............................ ........ ......

* This incIudes cases of patients referred for treatment of cervical metastatic cancer after the primary growth in the tonsi1 had been controIIed.

HISTOPATHOLOGY”

A histoIogic classification of the growths in the series herein reported is given in Table I. The epidermoid carcinomas (84 per cent) and the lymphosarcomas (16 per cent) present about the same reIa- tive distribution of these two tumors as in a series of nasopharyngeal

* Dr, Fred Stewart, pathologist to the MemoriaI HospItaI, has recently reviewed a11 histdogic shdes in this series and his analysis is as follows:

TonsiIIat cancer comprises sevcraI more or Iess we11 defined histoIogic entities. The terminology employed in the diagnosis of cancer of the tonsil at the Memorial Hospital is the foIIowing:

The designation, “ squamous carcinoma,” serves to distinguish forms of epidermaid cancer in which keratinization and the occurrence of squamous pearls may be noted even though the ceIIuIarity and anapIasia be extremely marked. Degrees of anapIasia are indicated by the grading of the tumor.

The term, “ transitional celI carcinoma,” is empIoyed to designate a group of epidermoid cancers in which keratinization is absent or es.sentiaIIy absent. Such tumors tend to grow in pIexiform character and exhibit varying diffuseness. They resembIe the common epidermoid cancer of the uterine cervix which many European writers describe as basaI ceI1 epithelioma. The Iatter expression is not used in this connection at the Memorial Hospital because of possibilities of eonfusion with the common basa1 ceI1 cpithelioma of the skin. Transitiona celI epitheIioma may exhibit various degrees of cellularity and anaplasia which are Iikewise indicated by the grading assigned. Needless to say, squamous and transitiona celI cancer may be imperfectIy distinguished and there are borderline tumors where specific designation is impossibIe.

The term, ” IymphoepitheIioma,” introduced by kegaud42 and later by Schmincke4& has caused much difficulty. Laboratories are not in agreement as to what constitutes a

I4t

168 MARTIN, SUGARBAKER-CANCER OF TONSIL.

cancer recently studied, which is to be expected since both of these areas are parts of WaIdeyer’s tonsiIIar ring. MaIignant salivary gland tumors are not uncommon in the palatine tonsiI, but none occurred during the period covered by this report. AIthough the numbers in the separate histoIogic groups are too smaI1 to permit of defmite concIusions from a statistical analysis of their respective cIinica1 behaviors, we have attempted to extract certain pertinent data from the avaiIabIe materia1.

With regard to the primary Iesion, the size on admission averaged 4.3 cm. in IymphoepitheIioma as compared to 2.9 cm. in transitiona cell. carcinoma, aIthough the average duration of symptoms was about one month longer (seven months) in transitiona ceI1 carcinoma. There was no appreciabIe difference in the average ages of the various histoIogic groups. Metastases were present on admission in 96 per cent of the transitiona ceI1 cases as compared to about 87 per cent in IymphoepitheIioma and Iymphosarcoma. There were bilateraI

lymphoepithelioma. Thus Ewing 17 states that he examined tumors diagnosed as lympho-

epitheiioma by Lacassagne and found that they included types diagnosed by himself as

transitional ceII carcinoma and Schneiderian carcinoma. Tumors in our own collection

received from Regaud with the diagnosis of Iymphoepithehoma fail to correspond to the

structure iiiustrated by Schmincke in his paper, yet do seem to fit the description

furnished by his text, for Schmincke employs the term “syncytiaIes Karzinom,” an apt designation both for the structure of certain primary tumors and for those metastatic

deposits in nodes which some pathologists would regard as primary endothehoma of

lymph nodes. Schmincke derived the Iymphoepithelioma from the reticuIum of the tonsii which he accepted as being of epithebal (entodermal) origin. If the reticulum of the

tonsi is of epithelial origin, a supposition by no means generaily admitted by morpholo- gists, it has so far departed from its original epitheIiaI morphoiogy that it is not surprising

that tumors originating from it shouId be separable with difficulty or not at ali from

reticuIum celi Iymphosarcoma. To make the distinction between reticuIum ceil sarcoma and lymphoepithelioma on the basis of the usua1 biopsy received in this hospitai is

almost impossible. Specimens are necessarily smaii, and since they are taken with forceps from soft tumors easily crushed, the intimate morphoIogy is apt to suffer from distortion. In reviewing specimens formerly diagnosed by us as lymphoepithelioma, we have

changed a number to iymphosarcoma and have retained a few, with some hesitancy, as certainly resembbng the types described by Regaud and Schmincke. Thus the number of

tonsiiiar Iymphosarcomas in the Memorial Hospital records has been increased at the

expense of tumors previously diagnosed as iymphoepithehoma. The fieid of diagnosis of these tonsiIlar and nasopharyngeai tumors is one of the most difhcult in tumor pathology. Unfortunately the cIinica1 course of the disease is by no means heIpfuI in separating the

types. With the exception of one case of Kaposi’s hemorrhagic sarcoma which occurred in

the tonsiI Iate in the course of generaiized disease and one instance of probabIe granuIa- tion tissue sarcoma in a young maIe, the pathoIogica1 laboratory at Memorial Hospitai

contains no instance of tonsiltar sarcoma other than iymphosarcoma in one or another of its forms.

BI Ill


metastases in about 13 per cent of the lymphosarcoma and transi- tiona1 cell groups as compared to IO per cent in IymphoepitheIioma and squamous carcinoma. The histoIogic grade of the tumor has considerable influence on the cIinicaI course of the disease and espe- ciaIIy on the tendency to metastasize, as shown in Chart I.

LymphoepitheIioma appears to run a much more malignant course, the duration of life from the beginning of symptoms being tweIve months as compared to seventeen months in transitional cell carcmoma.

DIAGNOSIS

Biopsy. AIthough a clinica diagnosis is not dificult in the average case of tonsiIIar cancer, a biopsy should aIways be made before treatment is instituted. It shouId be realized that treatment for cancer, whether by radiation or surgery, must be aggressive and therefore is always attended by definite risk, discomfort and expense. Without biopsy the cIinica1 diagnosis wiI1 inevitably be erroneous in an occasiona case. If the treatment is to be by radiation, not only shouId a tissue specimen be removed from the ton&I but an aspiration biopsy shouId be made for purposes of record to confirm the diagnosis of metastatic nodes. 2gd In the present series, the diagnosis of the primary lesion was confirmed histoIogicaIIy in a11 cases, and in the majority attempts were made to confirm the character of the enlarged cervica1 nodes. At the present time, aspiration biopsy of cIinicaIIy positive metastatic nodes is routine in aI1 cases of intraoral and pharyngea1 cancer.

Dela_y in Diagnosis. On the basis of the average size (4 cm.) of the primary Iesions in this series, it is obvious that cancer of the ton&I at the present time is not being diagnosed at an earIy stage. The most frequent first symptom of this disease, “sore throat,” is not alarming to the average layman because it is also so characteris- tic of the more common acute and chronic tonsilIitis and pharyngitis. The fiction of the benign character of cervica1 lymphadenopathy in the aduIt (usuaIIy metastatic cancer) is often supported by irrespon- sible medical opinion, and the patient is advised that he has “enlarged gIands” and that he shouId appIy hot compresses or massage the area with a we11 known proprietary ointment containing iodine. It is of interest to note that over fifty years ago BIand SuttonJg caIIed attention to the malignant character of most asymmetricaIly enIarged cervicaI lymph nodes in the aduIt and stated that he


believed practicaIIy aII were secondary to undiscovered primary cancer in the mucous membranes of the oraI cavity or pharynx. In our series the average interva1 from the first symptom to the first visit to a doctor was about two months, after which there was an average delay of about five months before the patient finaIIy applied to our clinic with a tentative diagnosis of cancer. The medical profession is, therefore, cIearIy responsibIe for an average delay of five months in the diagnosis of this disease.

Chronic Tonsillitis. Since the patient with tonsiIIar cancer usuaIIy compIains onIy of a sore throat when he first consults a doctor, the most common mistaken diagnosis is chronic tonsillitis, for which topica applications and gargles are usuaIIy recommended. In our series, tonsillectomy (an accepted method of treatment for chronic tonsiIIitis) had been performed in thirteen (9 per cent) of the cases before admission, and in nine of these there was gross and microscopic evidence of residual cancer in the pharynx. In onIy four of the cases had the excised tonsils been examined histoIogicaIIy. It is impossibIe to say whether these examinations were routine or whether the surgeon or the pathoIogist had requested the examination because cancer was suspected after gross examination, but it is obvious that most of the operations had been performed under a mistaken diagnosis of chronic tonsiIIitis. In any case, it is probably unavoidabIe that early cancer of the tonsi wiI1 occasionally be excised by a busy IaryngoIogist, and for this reason the routine examination of al1 excised tonsiIs (as practised in many hospitaIs) is indicated.

A differentia1 diagnosis between cancer and the chronically infected and scarred tonsil, folIowing repeated attacks of acute tonsiIIitis with peritonsiIIar abscess, may be very dif?icuIt, especiaIIy in a patient who has histoIogicaIIy positive metastatic cervical cancer. ChronicaIIy infected tonsiIs usuaIIy present deep crypts which, on pressure, extrude inspissated pus with slight bleeding. Careful palpation of both tonsiIs is of vaIue, with a comparison of the two for any differences in size and consistency. Specimens for histoIogic examination shouId be taken from IocaIized indurated areas or from those which 6Ieed on sIight trauma. Systematic examinations of the entire upper respiratory and aIimentary tracts should obviously also be made in such cases.

SypMis. From an analysis of the present series, it is apparent that an erroneous diagnosis of syphiIis is made Iess often in cancer of


the tonsiI than in maIignant growths of the tongue. In onIy three of our cases had antifuetic treatment been given-in one case for three years and in two cases for six weeks each. Gumma of the tonsi (which cannot be differentiated cIinicaIIy from cancer) is exception- aIIy rare. In the records of our cIinic going back to 1923, during which time over 450 cases of tonsiIIar cancer were treated, only one instance of gumma of the tonsi is recorded. The diagnosis was estabIished by a biopsy of the tonsi1, a positive Wassermann test and compIete regression of the Iesion in Iess than a month under antiIuetic therapy.

Papilloma. This benign tumor occurs on the tonsi about as frequentIy as in any other area of simiIar size in the ora cavity or pharynx. Such growths have a warty papiIIary appearance, and show a greater tendency than cancer to fungate from the mucous surface. The diagnosis may be suspected from the clinica appearance, but finaI confirmation by histologic examination is required. The treatment is by narrow excision either by scaIpe1 or endothermy. During the period covered by this report, one case of papiIIoma of the tonsil was seen at MemoriaI Hospital.

Rarely, patients with Ieukemia present markedIy enIarged tonsils as we11 as generaIized Iymphadenopathy, but errors in diagnosis cannot occur with proper Iaboratory and physica examinations. OccasionaIIy, patients with acute peritonsiIIar abscess, simpIe granuIoma, actinomycosis and retention cysts are referred to a cancer cIinic for a differentia1 diagnosis. If biopsy is routine in the diagnosis of a11 suspected superficia1 cancer, these conditions wiII present no specia1 probIem.

GENERAL PRINCIPLES IN TREATMENT OF CANCER OF THE TONSIL

There is no structure of the pharynx more accessibIe to surgica1 removal than the tonsi1. It performs no vita1 function and may be completely excised without great operative risk or postoperative shock. For benign diseases of the tonsi1, tonsiIIectomy has Iong been one of the most successfu1 of surgica1 operations, and from the theoretica standpoint aIone early cancer of the tonsi shouId be amenable to surgical excision.

Despite these theoretica advantages, surgery offers IittIe in the treatment of this disease, since the primary Iesions are seIdom smaI1 or suffrcientIy IocaIized in the tonsiIIar fossa to be operabIe when first seen. In our series of 157 cases, the growths averaged about 4 cm. in diameter on admission. There were onIy two (about I per cent) in

I[r4ll


which the primary lesion was I cm. or Iess, and in both of these metastases were aIready cIinicaIIy demonstrabIe on admission, an indication in genera1 of the inoperabiIity of this disease. Simmons48 found onIy one operabIe case among fifty-four patients with cancer of the tonsil. The Iiterature from I 870 to x9 IO contains a few reports on the surgica1 ‘treatment of cancer of the tonsi1, after which time, with the development of scientific radiation therapy, the surgical reports have steadiIy diminished in number as the suitabiIity of radiation therapy in this disease has been demonstrated.

The earIier writers reported an operabiIity of 40 to 60 per cent; but on weighing the evidence one must concIude that their surgica1 judgment in this regard was hardIy sound since their operative mortaIity was about 30 per cent, and they appear to have obtained practically no cures. The simpIest form of operation, tonsiIIectomy through the open mouth, was advised by But1in.j JacobsonzO recommended spIitting of the cheek from the commissure back as in Jaeger’s exposure for cancer of the tongue. In this cheek spIitting operation and in a11 the more extensive procedures subsequentIy deveIoped, most of the authors suggested preliminary tracheotomy as we11 as temporary ligation of the common carotid artery.

In order to excise the tonsiIIar area more wideIy, MikuIicz,37 Langen bet k, 26 von Bergman1 and Kocher 24 advised a vertical section of the mandibIe just anterior to the ascending ramus and a dis- articuration of the ascending ramus and its remova with a portion of the lateral pharyngea1 waI1, IncIuding the tonsi and the growth. SediIIot’s”6 exposure for cancer of the tongue was also recommended -sectioning the Iower Iip and mandibIe at the symphysis, the incision being continued backward in the ffoor of the mouth to the anterior tonsiIIar piIIar, after which the mandibuIar fragments couId be retracted IateraIIy to obtain access to the tonsiIIar area. Another approach to the tonsi was by a IateraI pharyngotomy,7 the opening being made through the side of the neck without incising the cheek. Such an operation has recentIy been popuIarized51 for various maIig- nant growths of the pharynx. Jacobson took exception to ButIin’s statement that simpIe tonsiIIectomy was the most suitable operation for cancer of the tonsi and maintained that the more extensive operations produced the Iongest periods of survival. Despite Jacob- son’s staunch support of the more radicai procedures, however, he was abIe to report only one cure (eIeven years) among an unstated number of his own cases. Eggersl” has recently described and


advocated several of these oId technics with no essential modifica- tions, but he incfudes no cures nor case reports.

In the fina anaIysis, one must conclude from its history that the operative treatment of cancer of the ton41 justifies Despons’ epithet of “surgery of despair.“13 The reasons for the faiIure of surgery in cancer of the tonsi are, first, that the primary Iesions are extensive and inoperabIe when first seen and, second, that ~5 per cent of a11 cases have metastases on admission. Even though both the primary lesion and the cervical metastases were operable (as is rareIy the case}, the prudent surgeon wiI1 concede that a wide excision of the tonsilIar tumor through the mouth cannot be combined safeIy with a block dissection of the neck at the same operation, nor can the bIock dissection be performed safeIy unti1 at Ieast partiaI healing has occurred in the pharynx. Such a deIay wouId usuaIIy resuIt in aIIow- ing the cervicaI metastases to become inoperabIe. Furthermore, it wouId be injudicious to combine bIock dissection of the neck with latera pharyngotomy; and if the pharyngotomy were done first, the scarring of the neck would precIude the performance of an adequate neck dissection subsequentIy. As we have aIready mentioned, simpIe tonsiIIectomy had been performed in fifteen cases of the present series before the patients were admitted to our cIinic. Of these, four patients had no recurrence in the tonsi and, superficiaIIy considered, these successfu1 resuIts might be considered surgica1 cures, except for the fact that aII were treated for cervica1 metastases and the tonsiIIar area was incIuded in the beam of radiation exactIy as if the growth in the tonsi had never been removed. The cure rate in these fifteen cases (27 per cent) is better than the average, which is undoubtedly due to the fact that a11 were obviousIy reIativeIy earIy when first treated.

WhiIe surgery offers IittIe, radiation therapy is particuIarIy suitable in tonsiIIar cancer. This area is accessible to irradiation both through the skin of the neck and through the mouth by a peroral portaI. Cancer of the tons& in generaI, is among the more radiosensitive of pharyngea1 and oral tumors, and the upper portion of the pharynx, in contrast to the hypopharynx, can be heaviIy irradiated without seriousIy impairing any vita1 function or bringing on grave comphcations. The most frequent site of metastasis in cancer of the tonsiI is the subdigastric Iymph node which Iies in the upper deep cervical region onIy a IittIe beIow the Ievel of the tonsi itseIf. This almost direct superposition of the first and most frequently invoIved


Iymph node permits of the irradiation of both the primary lesion and the metastases through the same portal, which avoids the use of portaIs so Iarge as to tax severeIy the patient’s genera1 tolerance. By contrast, in cancer of the nasopharynx the primary lesion and the metastases are widely separated and each must be irradiated through individua1 portals, thereby resuIting in a greater tax upon genera1 tolerance.

The first form of radiation treatment for tonsiIIar cancer was the impIantation of seeds or interstitia1 needIes.21 At the MemoriaI HospitaI Janeway and Quick, 2 l* 41 in 19 I 7, empIoyed bare gIass radon seeds interstitiaIIy, suppIemented by moderate doses of externa1 radiation (radium packs or x-ray). With the deveIopment by Coutardll of the principIe of fractionated x-radiation, this growth was one of the first in which the method was found successfu1. Berven,2 in 193 I, reported success by the use of fractionated doses of radium eIement pack. He has aIso pubhshed the description of an eIaborate device for hoIding radium pIaques in contact with the tonsi1, which to us does not seem practica1 for a number of reasons.

Beginning about 1930, we attempted at the MemoriaI HospitaI to steriIize tonsiIIar growths by the use of fractionated x-radiation. Further experience has reveaIed that permanent regression by this method aIone often necessitates pushing the dosage beyond the limit of the patient’s immediate general toIerance or beyond the point at which the IocaI tissues wiI1 recover to an acceptabIe degree. It has since been found that the number of permanent cures can be increased and the percentage of untoward sequeIae reduced by using submaximal doses of fractionated x-radiation, supplemented in most cases by the impIantation of small doses of radon seeds in the residual tumor, either directIy into the primary Iesion or in immediateIy adjacent metastatic nodes.

A review of the recent pubIications on tonsiIIar cancer reveaIs some diversity of opinion as to the preferabIe method of treatment.6,12,18,19,22,23,28.34,43,47 J-'h e method most commonIy advised is fractionated x-radiation and next in order is fractionated treatment by radium element packs. A minority recommend radon needIes or seeds alone or combined with surface appIications of radium. Among these authors, onIy those who depend mainIy upon fractionated radiation by x-ray or radium eIement packs report five-year cures.

Gerteral Hygienic Measures. The adequate radiation treatment of any form of pharyngea1 cancer is a severe tax on the patient’s

UI7l!


generaI strength. In the larger growths there is aImost always a marked local sepsis; and if swallowing has been painful, there is at least some degree of malnutrition. The convaiescence may be short-

FIG. 3. In externa1 radiation of the tons& the skin portal

shouId be centered about over the angIe of the jaw and

the beam pointed horizontaIly toward the opposite angIe.

Accuracy in treatment is favored by meta cyIinders

which exactly determine the size of the portal and remain

in contact with the skin, steadying the patient during treatment.

ened and the severity of the complications greatly reduced if the patient can be hospitalized for a month or six weeks during and immediateIy foIIowing radiation treatment. If hospitalization is not practicable during this Iong a period, it should be arranged for at Ieast two weeks during the height of the reaction. At the beginning of treatment, measures should be taken to counteract the various complications. which may aIready be present, as we shall discuss presently under “ Complications.”

!Wl


Treatment of the Primary Lesion. Treatment shouId begin with fractionated external radiation given through the cheeks and sides of the neck and, provided that the lesion does not extend downward onto the pharyngeal wall, peroraI x-radiation through the open mouth. To irradiate the tons& the external portals shouId be centered about at the angIes of the mandibles (Fig. 3), the exact position varying with the individua1 case as determined after carefuI examination. Metastases are most often Iocated over the carotid bulb (2 to 2.5 cm. beIow and a IittIe behind the angIe of the mandible), in which case the porta should be sIightIy larger (8 rather than 7 cm.) and centered nearIy over the node with the beam tiIted upward and forward toward the tonsi1. Even though the position of such an external portal may seem fairIy obvious, in relation to some fixed anatomic structure, nevertheIess we fmd it of advantage to mark the center of the portal accurateIy by tattooing with India ink.* The external portals shouId be circular in order to be most effrcient,i and the size should be about 7 cm. in diameter unIess there are definite reasons for making it Iarger, such as the widespread distribution of cervica1 nodes or an unusuaIIy large primary Iesion, when the portals may be 8, 9, IO cm. or more in diameter. OrdinariIy, if a porta must be over 8 cm. in diameter, it is of advantage to make it ova1 rather than circuIar, so as to avoid as much as possibIe the irradiation of an unnecessariIy large volume of tissue.

If the primary Iesion is of average size and confmed to the tonsi itseIf, and if the jaws can be separated wideIy enough to admit a cyIinder at least 4 cm. in diameter, peroral x-radiation should aIso be given. Fur this purpose, meta cyIinders of various diameters are used, specially equipped with master cylinders so that the desired size and distance may be seIected for the individual case. One of us (M.) has aIready published a description of such appIicators and their several indications .2ga RecentIy we have devised an eIectricaIIy Iighted periscope which permits checking the accurate centering of

* The skin is cIeansed by rubbing with alcohol, and a small drop of India ink is placed at the desired point. With a steriIe sharp needle, six to eight punctures are made through the ink into the skin, and the excess ink wiped off. The tattoo mark wiI1 remain

permanentIy, and its size depends on the number of punctures. If later considered unsightly, it may be excised through a smaI1 ehipse.

t Square and circular portaIs of equal diameter are about equaIIy efficient if centered over a deep-lying mass, since the edge of the beam at its cIosest point is at the same distance from the edge of the tumor in either case. The area of a square is 25 per cent larger than that of an inscribed circle, and therefore a circular portal is t.o be preferred

since it accomphshes the same result with a smaIIer area.

MARTIN, SUGARBAKER-CANCER OF TONSIL ‘77

such peroral portals by indirect vision. (Fig. 4.) It is fortunate if the patient is edentuIous in the beginning; but if not, we beIieve that the

FIG. 4A, B, C. For descriptive Iegend see opposite page.

extraction even of sound teeth is justified when perora1 x-radiation is indicated and there is no other impediment to its administration.

With portals averaging 7 to 8 cm. in diameter (other factors being: 200 to 250 kv., 50 to 60 T.S.D., filter 0.5 to I .o mm. Cu) the

UNI


X-RAY TUBE HOLDER

LENS LIGHT

FIG. 4D.

FIG. 4. For perora1 x-radiation, a usefu1 attachment is an eIectricaIly Iighted periscope by which the accurate centering of the peroraI cylinder over the lesion may be checked

at the compIetion of the setup. A, master cylinder with a perora1 cylinder attached; B, eIectricaIIy Iighted periscope; c, checking the accuracy of a peroral setup with the periscope; D, diagram of construction of perora1 cylinder and eIectricalIy lighted periscope. (The Iens system and periscope were made by the American Cystoscope

Makers, Inc., New York City, and al1 other attachments by Mr. A. Schreiner, Master Mechanic of Memorial Hospital, New York City.)


daiIy dose shouId be about 330 r given to aIternate sides untiI a dose of about 3,500 r times 2 has been administered. If a peroraI porta can aIso be used, the single dose with a 4 cm. portal should be about 400 r

THE INFLUENCE OF HISTOLOGIC GRADE ON CERVICAL METASTASIS IN CANCER OF THE TONSIL

TOTAL CASES CASES tie NO. OF SQUAMOUS WITHOUT WITH WiTH CASES CAR- M ET. MET- MET

6 GRADE I 16

61 II 181 43 70

20 m 00

1 I 1 100

TRANS. CAR.

3 GRADE If: 3 100 3 III t 8 88

7 fp 1 7 100

7 LYMPHOEP. 1 6 05 20 LYMPHOSAR- c 20 100

CHART I. The frequency of metastases in cancer of the tonsil varies with the histologic

grade of the tumor from 16 per cent in squamous carcinoma grade I to IOO per cent in

squamous carcinoma grade xv, transitiona ceH carcinoma and lymphosarcoma.

given in rotation with the external cheek portaIs, the tota dose being about 2,500 r times 2 through the cheek portaIs and 2,800 r through the perora1 porta1. After the compIetion of such dosage by one or the other of the above mentioned pIans, it should be decided whether the depth of the reaction and the degree of regression of the tumor indicates continuing with the externaI radiation beyond this point or whether further treatment shouId be given by radon seeds impIanted in the metastatic nodes and also in the primary Iesion. By external radiation aIone, using two skin portals 7 to 8 cm. in diameter and with the other factors as above mentioned, the tota dose may be raised justifiabIy to 4,000 r or even 4,500 r times 2. If a peroral portal is used in addition, the tota dose may be as high as 3,000 r or 3,500 I- times 2 pIus 3,500 r through the peroral portaL

A better pIan than either one of these is the administration of externa1 radiation until the tumor has been reduced to a quarter or a

I[4


fifth of its origina size after which suppIementary treatment is given with radon seeds. By such suppIementary radon treatment, many of the undesirabIe effects of Iarge doses of x-radiation may be eIimi- nated. The dose of suppIementary seeds is somewhat empiric. A residual mass of cancer I cm. in diameter within a heaviIy irradiated pharynx shouId obviousIy not require the same additional tissue dose as if no externa1 irradiation had aIready been given; but on the other hand many of these residua1 masses, aIthough onIy a fraction of the origina tumor, are much more radioresistant than the origina tumor appeared to be at the beginning. One must not be too sanguine, therefore, in the hope that such masses will disappear under very smaI1 suppIementary doses. In the average case the strength of the individual seeds should be between 1.0 and 2.0 mc., and they should be geometricaIIy distributed throughout the residua1 mass, spaced about I cm. apart. Since the requirements in individua1 cases are obviousIy so variabIe, it is impossibIe to give more accurate recom- mendations for this type of therapy, but some estimate may be made from a review of the dosage in the present series.

The method of treatment most often successfu1 (22 per cent cure rate) was externa1 radiation aIone with portaIs from 7 to g cm. in diameter, the dose varying between 2,000 r times 2 and 4,200 r times 2, and averaging about 3,600 r times 2. This observation must not be misinterpreted, however, for these were the more radiosensitive lesions which promptIy cIeared under the externa1 radiation so that interstitia1 radiation was not considered necessary. X-radiation and suppIementary seeds were used in sixty-one of the determinate cases, with a cure rate of I I per cent. SuperficiaIIy considered, such treatment might appear much Iess successfu1 than that first mentioned, except for the fact that these growths were actuaIIy among the more radioresistant and buIky Iesions which did not compIeteIy regress under the average dose of externa1 radiation. AI1 .advanced cases with widespread nodes were treated by this method, thereby bringing down the percentage of cures. The average dose of externa1 readiation in these cases was higher (between 3,500 and 4,500 r times 2), and the doses of seeds into the tonsi in the successful cases ranged between I o and 16 mc. The radium element pack was used as the source of externa1 radiation in a few cases, the dose ranging from 80,000 to 120,000 mghrs. times 2 with portaIs 7 cm. in diameter, a R.S.D. (radium skin distance) of IO cm., and a frIter of 2.5 mm. brass (equivalent). In most of these cases there was a suppIementary dose


of radon seeds varying between IO and 20 mc. Radon seeds alone were successful in two instances of smaIIer Iesions treated by fairIy heavy dosage, about I 5 mc. each in two cases and 3 I mc. in one case.

Treatment of Minor Salivary Gland Tumors of the Tonsillar Pillars. Mixed tumors and adenocarcinomas histologicahy resembhng those of the parotid gIand and submaxillary saIivary gIands not uncom- monly arise in the minor saIivary gIands of the oraI and pharyngeal mucous membranes. It is well known that these tumors are most common in the hard and soft paIate and rare in the tonsiIIar pilIars or tonsiIIar fossae. There are severa such instances on record at the MemoriaI HospitaI, although none occurred in the present series. They usuaIIy deveIop as globose or ovoid tumors which tend to remain encapsulated or sharpIy delimited unti1 they have reached a size of severa centimeters. Those which histologicahy resembIe mixed tumors are essentiaIIy benign. The more ceIIuIar adenocarcinomas, although encapsuIated at first, may metastasize wideIy to bones and viscera, often skipping the cervica1 Iymph nodes. When occurring in the tonsiIIar region, these tumors are usuahy severa centimeters in diameter, seIdom uIcerated, with a history of several months or even years with very sIight symptoms. The painIess rounded swelling in the tonsiIIar region often suggests Iow grade peritonsiIIar abscesses anatomicaIIy, and these Iesions are often incised for purposes of drainage.

PracticaIly a11 of these minor saIivary gIand tumors are highIy radioresistant from the anatomica standpoint and can be safely enucIeated, preferabIy by cautery and 6Iunt dissection. There are two such successfu1 cases in the records of our clinic, but none occurred in this series. PreIiminary Iigation of the external carotid artery should be done if the tumors are large and deepIy seated. One case at the MemoriaI HospitaI was treated by massive doses of interstitia1 radon sufficient to produce rather widespread radionecrosis. The patient recovered and has remained we11 for more than ten years. It is obvious that such treatment by caustic doses of radiation is not justified if these tumors can be enucleated.

Treatment of Cervical Metastases. As has been previously men-

tioned, metastatic cervical nodes were cIinicaIIy demonstr&Ie on admission in 76 per cent of the present series. It is obvious, therefore, that no pIan of treatment for cancer of the tonsi is adequate which does not provide for the management of this complication as we11 as the primary Iesion. As has been previously stated, it is fortunate that

fiNI


the most frequentIy invoIved node, and often the onIy one in this disease--the subdigastric-Iies only a IittIe beIow the IeveI of the primary lesion, so that both can be irradiated by the same beam of

TABLE II MlLLlCURTES IN GOLD SEEDS REQUIRED TO DELIVER SPECIFIED DOSES TO MASSES

OF VARIOUS DIAMETERS

Diameter of Mass-Centimeters Skin

Erythema Doses

1.0 1 1.5 1 2.0 1 2.5 1 3.0 1 3.5 / 4.0 / 4.5 1 5.0 1 6.0 1 7.0 / 8.0

--. i- Number of Millicuries

5. _. I .o 6 . . , . . . I .2

7. . , . . . ’ .4

a... ._._._ 1.6 ‘) ,._... _.. 1.8

10 . , . . 2.0

II.. . . . . . . . 2.2

12.,.. . . . . . 2.4

2.5 3.0 3.5 4,o 4.5 5-O 5-5 6.0

4.0 7.5 ‘0 4.8 9.0 I2

5.6 t0 14

6.4 12 16

7.2 14 18 8.0 ‘5 20 8.8 ‘7 27.

* 18 24

20 27 24 32 28 38 32 43 36 49 40 54 44 48 L:‘:.

35 42 49 56 63 70 77

_s4

x-radiation. In our series, in about 40 per cent of those patients who had cIinica1 metastases on admission, the nodes were Iimited to this area. Were it not for this fact, the cure rate in tonsiIIar cancer wouId be even Iower than it is at present.

Under the above mentioned pIan, the externa1 irradiation of the most frequent metastases and of the primary lesion is carried out simuItaneousIy and through the same porta1. A decision as to the necessity for suppIementary seeds in the cervica1 nodes and in the primary Iesion shouId be made on the compIetion of a submaxima dose. In most cases it wili be found that cervical metastases wiI1 not regress compIeteIy under average or even under Iarge doses of external radiation, and suppIementary seeds are indicated for the invoIved Iymph nodes more often than for the primary Iesion. The seeds may be inserted either through puncture wounds in the skin or after cervical exposure of the outer surface of the node. The dosage is somewhat empiric and depends both upon the size of the residua1 node and upon the amount of externa1 radiation previousIy given. A tissue dose of 5 to 6 S.E.D. (as caIcuIated from TabIe II) shouId be sufhcient as a suppIementary dose, but 8 to IO S.E.D. should be used if seeds alone are empIoyed without preIiminary x-radiation. The successful doses in the present series after externa1 radiation ranged between 9 and 22 mc. In the present series eight patients with histoIogicaIIy positive nodes (proved by aspiration biopsy) remained free of disease for a fuI1 five-year period after irradiation therapy. It wouId be misleading to compare the percentage of successful resuIts obtained


by radiation treatment of metastatic cervical nodes in this series with those obtained by surgery, since it is obvious that those treated by radiation incIuded the Iess favorabIe and advanced cases, whiIe those treated by surgery were seIected cases and limited to those patients who deveIoped metastases only after compIete contro1 of the primary lesion by irradiation.

In the treatment of tonsiIIar cancer, if a metastatic node Iies beIow the upper deep cervicaI region, it may in some instances be permissibIe to use a Iarge porta to include the primary Iesion and one or more scattered nodes. In the average case, however, we beheve it best to treat wideIy separated nodes by individual portaIs 3.5 to 5 cm. in diameter, the size of each being just Iarge enough to incIude ampIy the individuaI node. Treatments may be given on aIternate days for a tota dose of 7,000 to 8,000 r to each node through such smaI1 portals, and after about three weeks a suppIementary dose of radon seeds impIanted.

In tonsiIIar cancer, neck dissection is of little practical vaIue in the treatment of cervica1 metastases which are present at the time of admission. It is obvious that neck dissection cannot be carried out safeIy in conjunction with adequate irradiation of the primary Iesion. Neck dissection is occasionaIIy indicated in Iarge metastases occurring some months after heaIing of the primary Iesion. The choice between neck dissection and radiation in such cases depends upon the judgment, experience and abiIity of the individua1 surgeon with the two methods and the equipment avaiIabIe. In the present series, five patients in whom metastases deveIoped severa months after the treatment of the primary Iesion were treated by neck dissections. In three, the excised nodes were found to be histoIogicaIIy positive, and of these one remained we11 for a period of five years.

COMPLICATIONS

Heavy protracted irradiation through Iarge portaIs to any part of the body is aIways folIowed by a chronic radiation sickness. When Iarge fieIds are used over the neck or cheeks, such an effect is one of the most serious compIications, and it may occur in the absence of any intense IocaI tissue reaction such as skin bristering. The patient becomes IistIess, weak, toxic, loses weight and becomes cachectic. These symptoms are not due entireIy to the IocaI discomfort, dysphagia or reactions in the skin or mucous membranes, but probabIy arise in part as the resuIt of toxic absorption from the Iarge voIume of


tissue irradiated as we11 as from the partia1 derangement in the function of the various vascuIar, gIanduIar and nerve structures of the neck. Many cancer patients who receive heavy irradiation therapy undoubtedIy die more from the effects or radiation than from the disease itself. Such systemic effects of radiation can be IargeIy pre- vented by avoiding unnecessariIy large portaIs. In pharyngea1 cancer the portaIs seIdom need be more than 7 or 8 cm. in diameter, and with such portaIs the IocaI intensity may usually be safeIy increased so as to reach a cancer IethaI dose. The use of portaIs 2 to I 4 cm. in diameter or Iarger in the pharynx, whether or not required by the extent of the disease, is aIways indicative of an unfavorabIe prognosis.

The Acute EJect in the Skin and Mucous Membranes. -TGe skin of the neck and the mucous membranes of the pharynx, which are directIy exposed under the beam of radiation, shouId in a11 cases show an acute bIistering or epidermicida reaction within the first month of treatment. This degree of reaction is usuaIIy necessary if roentgen radiation is to produce a IethaI effect on a tonsiIIar tumor. The depth of the cutaneous reaction may seem alarming in appearance, but one may be assured that such a reaction produced by repeated x-ray treatments over a period of fifteen to thirty days aIways heaIs under proper care without any serious sequelae. Under fractionated x-radiation by the technic aheady described, the typica reaction begins after about a week with a miId erythema which becomes deeper but remains dry and sometimes sIightIy pigmented during the second week. About the fifteenth to twentieth day, the superficia1 or epiderma1 Iayers of the skin become sIightIy raised by serum, and the corium, being denuded (at first over onIy scattered areas), becomes a deep red weeping raw surface which bIeeds on sIight trauma. After a few days the entire area of skin exposed to the direct beam may be denuded of epidermis. WhiIe the skin is in a dry state, frequent applications of petroIatum or liquid petroIatum shouId be used with no other dressing. When the surface is denuded, it shouId be kept covered with a singIe layer of fine mesh gauze bandage impregnated with petroIatum, boric acid ointment or the foIIowing hydroscopic ointment:

Parts Hydrous fat, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Zincoxide......................................................

MARTIN, SUGARBAKER-CANCER OF TONSlL 185

The raw surface should be sponged daiIy with steriIe sahne soIution to remove the exudate, and the surrounding normal skin shouId be cIeansed with soap and water and aIcoho1. If the raw surface is protected from dryness, crusting, trauma and secondary infection, it ordinariIy heals compIeteIy in from ten to fourteen days, leaving a soft pIiabIe skin in which moderate teIangiectasis may or may not Iater appear.

The mucosal reaction, a counterpart of that in the skin, should go on to a membranous mucositis of diphtheritic appearance in the tissues exposed to the direct beam. AIthough not painfu1 whiIe the part is at rest, this Iesion is tender on movement and bIeeds sIightIy but easily if the membrane is disturbed. It reaches its maximum in about twenty or thirty days after the beginning of treatment and usuaIly disappears in ten or fifteen days, its duration depending on the size of the tota dose. When radon seeds are used to suppIement the externa1 radiation, this type of reaction becomes more acute in the neighborhood of the impIantation. During its height there is considerabIe serous and mucoid discharge from the surface, with a diminution in the quantity and an increase in the viscosity of the saIiva. If the condition is not reIieved, the accumuIation of these secretions is the cause of a great dea1 of discomfort.

As a matter of routine at the beginning of’treatment, irrigations of the throat shouId be given every two hours with an irrigating can equipped with a rubber tube and gIass nozzle. A warm solution of sodium bicarbonate and water shouId be used. The patient is instructed to take the irrigation at night if he wakens and notes marked discomfort. This singIe procedure affords considerabIe relief, as is proved by the fact that it is voIuntariIy continued by the majority of patients without further urging. The viscidity of the secretions is greatIy reIieved as soon as the mucositis cIears, but dryness of the mucous membranes may persist in a Iessening degree for severa years after treatment, a compIication which wiI1 be discussed under “ Xerostomia.” The reaction of the mucous membrane is partly responsible for the dysphagia which sometimes occurs.

L$fsp bagia. This compIication is always present to some extent when an acute radiation reaction is produced in the pharynx. Even though the disability is onIy moderate, it still may be responsibIe for a marked Ioss of weight and strength in a nonco-operative patient. The patient should be instructed as to the necessity for maintaining his weight and nutrition, and shouId be encouraged to take an


adequate balanced Iiquid diet af high caIoric vaIue. If necessary, a rubber feeding tube (No. 16 French urethra1 catheter) may be inserted through the nasa1 cavity down into the esophagus and Ieft for a period of severaI days during the height of the reaction. We have found that most cancer patients who must subsist on even the most inteIIigentty prepared liquid diets are benefited by suppIe- mentary liver and vitamins (especiaIIy B and c). One teaspoonfu1 of Iiver extract powder may be given two to three times daily or two to three heaping tabIespoonfuIs daily of dried yeast may be dissoIved in the Iiquid feeding. Fresh fruit juices supply vitamin c.

Pain. As we have previously mentioned, pain is one of the earhest symptoms in cancer of the tonsi1. If the disease is uncon- troIIed or if radionecrosis and IocaI sepsis occur, the pain in the local area may eventuaIly radiate to the ear or to the whoIe side of the head. When these compIications are also associated with uncon- troIIed, deepIy ir&Itrating cervica1 metastases, the continua1 intract- abIe pain may be so severe as to be a cause of death from exhaustion. NeuroIysis by aIcoho1 is not suitabIe in this area since not onIy is the fifth cranial nerve invoIved, but aIso the seventh and ninth and often the upper cervica1 sensory roots. LocaIized pain on admission due to sepsis and to pressure by the growth in the tonsi is often partIy relieved by hot throat irrigations and tends also to diminish as the tumor regresses under radiation therapy. Late radionecrosis is aImost always associated with pain radiating to the ear over the chorda tympani. Since in a11 intraora1 cancer the pain is apt to continue for severa weeks before compIete relief is obtained, attempts shouId aIways be made in the beginning to obtain relief by such miIder sedatives as codeine in doses of $5 gr. (30 mg.) and acety1 saIicyIic acid IO gr. (600 mg.) given every six to tweIve hours, increasing the number of daiIy doses at first and finaIIy increasing dose to I gr. (60 mg.) of codeine if necessary after two to three weeks when the patient becomes accustomed to the drug. Morphine shouId be reserved for the advanced stages or for acute transitory conditions since, once estabIished, the morphine habit can be broken onIy with diff!culty and is often the cause of more suffering than the cancer itseIf.

Radionecrosis and Osteomyelitis. There is a widespread erroneous beIief that radionecrosis depends entirely upon the intensity and quality of the radiation. As we 296 have previousIy pointed out, radionecrosis at a depth never occurs except when infection is introduced,

b9ll


no matter how heavy the radiation dose. If a mass of tissue at a depth is compIeteIy devitaIized by irradiation (or by any other physical agent) and no Iocal infection occurs, the devitalized mass wiI1 be absorbed or repIaced by fibrous tissue or wiI1 caIcify and be toIerated IocaIIy as any form of steriIe foreign body. If infection is introduced, the phenomenon generaIIy known as radionecrosis occurs, and the devitaIized mass suppurates and must be completeIy extruded before heaIing can take pIace.

Since infection is aIways present on the surface of the skin or mucous membranes, superficia1 radionecrosis occurs in superficia1 growths with greater frequency than with those at a depth. In the tonsiIIar area, infection is always present in uIcerated tumors and may aIso be present in the depths of the tonsiIIar crypts, so that the conditions favorabIe to sIough are unusuaIIy frequent. Radionecrosis may occur after heavy external radiation aIone, but the incidence is increased by the use of interstitia1 radon both because of the possibiIity of too great concentration of the radiation effect and because infection may be introduced during the insertion of the seeds. A smaI1 voIume of radionecrosis in the upper part of the pharynx or ora cavity is compatibIe with an excellent end result. The greater the voIume of the sIough, the greater the chance of erosion of such vita1 structures as blood vessels.

When the soft tissues have been eroded by necrosis, heaviIy irradiated bone of the mandibIe may be exposed and osteomyeIitis results, a compIication which occurred in seven of our patients, one of whom required a partiaI resection of one ramus of the mandibIe. Of these seven patients, three remained aIive and we11 for five years.

In treating this complication, the slough should be gentIy manipuIated every day, the Ioosened portion removed and the more adherent cIipped away with scissors. No aggressive attempt shouId be made to cIear out a11 necrotic tissue. In these cases, carefuI attention to ora hygiene by irrigations and gargIes is particuIarIy important.

PracticaIIy a11 necrotic lesions of the ora cavity are produced by ana.erobic saprophytic organisms for which oxygenic applications are specific. The average preparations, such as sodium perborate and hydrogen peroxide, are too transient in their action, but the use of zinc peroxide (marketed under the name “Z. P. 0.“) as recentIy introduced by MeIeney35 wiI1 often clear the offensive odor and improve the appearance of the uIcer within a few hours. Sunderland

Usd


and BinkIey50 have recently reported upon the use of this substance for necrotic intraora1 cancer in our cIinic. In brief, the necrotic cavity is packed once or twice daily with gauze saturated by a suspension of zinc peroxide in water.

In the present series, radionecrosis or sIough appeared as a com- pIication in twenty-nine cases (20 per cent), and r4 per cent of these patients recovered and remained free of disease for five years, a figure onIy a IittIe beIow the average cure rate (I 8 per cent). This compIication does not appear to have been absoIuteIy determined by the size of the radiation dose, either in x-radiation or in seeds. It occurred in one instance after a dose of onIy 3 mc. in seeds and was absent in another instance after 50 mc. in seeds suppIementary to 80,000 mghrs. times 2 to the neck with the eIement pack. One of the most serious concomitants of radionecrosis is hemorrhage.

Hemorrhage. The bIood suppIy of the ton&I arises from severa sources: the tonsiIIar and the ascending paIatine branches of the externa1 maxiiIary artery, the dorsaIis Iinguae branch of the IinguaI artery, the descending palatine branch of the interna maxiIIary artery and the ascending pharyngea1 branch of the externa1 carotid. Most of these vessels anastomose freeIy in the tonsiIIar fossa. The Iargest and most important from the standpoint of hemorrhage is the tonsiIIar branch of the externa1 maxiIIary (facia1) artery, which passes upward between the interna pterygoid and the styIogIossus and pierces the superior constrictor to enter the tonsi on its Iateral surface.

Hemorrhage is most IikeIy to occur when there is deep sIough which approaches the IateraI surface of the tonsi1, and immediate bIeeding is ordinariIy readiIy controIIed by digital tamponage unti1 arrangements can be made (within a few hours) to Iigate the externa1 carotid artery and its lingua1 and facia1 branches. The first hemorrhage is seIdom fata1, but when the necrosis has become so deep and extensive as to cause even moderate bIeeding, the erosion of severa adjacent vesseIs, such as the interna carotid or the juguIar vein, may soon occur so that the eventua1 prognosis is grave. In the present series, hemorrhage occurred in sixteen cases, in seven of which the Ioss of bIood was considered su%cientIy severe to require Iigation of the externa1 carotid and its branches. In aI1 seven of the Iatter cases, subsequent hemorrhages occurred from other vesseIs foIIowing these Iigations, and a11 cases eventuaIIy terminated fataIIy. Obviously the deaths in these cases were not due to the effects of the Iigation of the

MARTIN, SUGARBAKER-CANCER OF TONSIL 1sc)

external carotid, for it is a we11 known fact that Iigation of one or even of both of these arteries is not a serious procedure.

Edema of the Hypopharynx. This compbcation shouId seIdom occur after proper radiation treatment of tonsiIIar cancer except when metastases are widespread through the neck so as to necessitate the use of portaIs 12 to 15 cm. in diameter. As we have previousIy mentioned, the use of such large portals, whether necessary or not, is aIways indicative of a poor prognosis. The occurrence of edema in the hypopharynx folIowing irradiation treatment of tonsiIIar cancer with nodes limited to the upper deep cervical region shouId be a reproach to the radiologist who has given the treatment. In the present series, edema of the hypopharynx occurred in some of the earlier cases when we did not appreciate the disadvantages of Iarge portaIs. In the more recent cases, it has occurred only when Iarge or muItipIe portaIs were necessary because of widespread metastases extending to the Iower portion of the neck.

Xerostomia. The paIatine tonsil lies under the anterior edge of the parotid salivary gland in the horizonta1 corona1 pIane. A porta even as smaI1 as 7 cm. in diameter wiI1, therefore, incIude the greater part of the parotid sabvary gIand as we11 as a portion of the sub- maxiIIary sabvary gIand. When cancer IethaI doses are administered to this area, the secretory function of these gIands is markedIy diminished. The mucous membrane of the upper portion of the pharynx contains numerous minor saIivary and mucous gIands which, with the saIiva, serve to moisten and Iubricate the mucous membranes of the ora cavity and pharynx. For these reasons, adequate radiation therapy of tonsiIIar cancer is aIways foIlowed by some degree of xerostomia. The saIiva decreases in quantity and its viscidity is increased. In such cases, inspection wiII revea1 that the soft paIate and pharyngea1 waIIs are rather dry and gIistening. AIthough this compIication cannot be entireIy avoided, its degree can be markedIy reduced by the use of mouth washes and by frequent sips of liquid petroIatum which the patient may carry in his pocket in a smaI1 bottIe. Chewing gum or hoIding a Iozenge in the mouth seems to stimulate the flow of saIiva and to distribute it more wideIy in some instances. After a year or two, the secretions tend to become more norma a’lthough compIete function is never entireIy restored.

Recurrence. FoIIowing treatment by radiation, even though there is apparent compIete regression and heaIing of the primary Iesion, IocaI recurrence may take pIace within a month or even after


TABLE III

FACTORS INFLUENClNG THE PROGNOSIS IN 148 CASES (DETERMINATE GROUP) OF CANCER

OF THE TONSIL

1931 to 1935, Inclusive -

-_

__

.-

-

-

-

-

_-

.-

Number of Five-year

&res

Total Number of Cases

Per Cent or Five-year

Cures

13 25 IS 17

16 26

------

o

40 15 20

9 22

_____

38 38

8

33 IO

-

25

14 I1 o

44 fo 43

0 14 20

14 -

0

16

25 25

_---

r4 19

Age in years: Below 40.. . 41 to50...... $1 to 60.. Over 60. Not stated.

2

5

6 II

I

......

......

Sex: Males............. ,..,.. ,..,..,,.. ,..,...,..,. Females.............................................

20

6

Size of I&on: 1 cm. or less ..................................... 1.Ito3cm.. ................................... 2.1 to3 cm ........................................... 3.1 to 4 cm. .......................................... 0 ver4cm ........................................... Not stated. ...........................................

Metastases (histologicaIIy proved) : None at any time. ..................................... None on admission .................................... Present on admission. .................... ............ Developed after admission. ............................. Present some time .....................................

HistopathoIogy: (primary) Squamous cell, grade I.. Squamous cell, grade II, Squamous cell, grade III_ Squamous cell. grade IV. Squamous ceI1, ungraded.. Transitiona ~11 grade 11. Transitional ceil grade 111. Transitiona cell grade IV. LymphoepitheIioma, . . Lymphosarcoma.. . . Unclassified. . .

i 2

0

4 I

3 0 I

4 2

Status of disease on admission: Residual. recurrent. and metastatic after treatment eisc-

where to primary: After radiation. _ . . . After surgery. . After combination. After endothermy.. . .

Primary cases, no treatment elsewhere. . . . . . . . . . . .

IO

12

4 4

0

2

I

I .- 1 30 118

4 22


as Iong as a year. In the present series such recurrence appeared in twenty-four cases, the majority (fourteen cases) between the fourth and sixth month, aIthough five occurred after one year. In this group there were no five-year survivaIs, aIthough curative attempts were made in a11 cases by suppIementary treatment-in a few cases by further x-radiation, but in the majority by additiona implantation of radon seeds in the areas of recurrence. The chief reason for the bad prognosis in recurrent cancer of the tonsil undoubtedIy Iies in the fact that the IocaI tissues of the pharynx and neck have already been so heavily irradiated that the tolerance to irradiation has been exhausted, and that further attempts are invariabIy fohowed by- fata compIications or persistent cancer. FoIlowing heavy irradiation there is no such thing as complete recovery of the tissues from the standpoint of toIerance to subsequent radiation.

A cIinica1 diagnosis of a recurrent or a persistent growth in the tonsiIIar area is readiIy confirmed by biopsy; but the diagnosis is much more diffrcuIt in suspected recurrent or persistent cervical metastatic cancer, since foIIowing heavy irradiation or surgery there may be indurated masses of scar tissue which cannot always be differentiated cIinicaIIy from residual cancer except by the subsequent cIinica1 course. In the present series, in at Ieast twenty-four cases after apparent compIete regression of the cervicaI lymph nodes there was recurrence or extension of the growth in the neck. Of these, two remained we11 for five years after subsequent treatment.

PROGNOSIS

Before the advent of radiation therapy, cancer of the palatine tonsi was considered practicaIIy a hopeIess disease, and even with modern methods this opinion is stiI1 held by many writers. Ducuing, 14 in 1935, stated that except in very rare cases, treatment of maIignant growths of the tonsi gives onIy paIIiative resuIts of very short duration. WhiIe this viewpoint seems extreme, the prognosis is not as favorabIe as in most other anatomic forms of growth in the oral cavity or pharynx. The reasons for this are fairly obvious. Since the tactiIe sense in the tonsiIIar area is not acute, the primary Iesion usuaIIy reaches a Iarge size before it is detected by the patient. In the present series the average diameter was about 4 cm. on admission. In many other areas of the oral cavity and pharynx the Iesions can be discovered much earIier. The advanced stage of the disease in the average case on admission is further indicated by the fact that three-


fourths of the present cases had metastases on admission. The histoIogic type of the growth is among the most maIignant of the oraI cavity and pharynx; and even if the primary lesion and cervical metastases are eventually cured, the patient is still under the hazard of generaIized dissemination. A review of the oIder literature reveaIs an aImost uniformIy hopeless attitude toward this disease when treated by surgery; for exampIe, E3roders,3 in 1927, stated that in his opinion no definite five-year cure had ever been obtained by surgery. Under radiation methods, the cure rate in the earIier cases is fairly satisfactory as compared to most forms of upper respiratory and aIimentary tract cancer. The factors which influence the prognosis are given in TabIe III, the most significant of which wiI1 be discussed

separateIy. Age. The chance of cure appears to be best between the ages of

forty and fifty. The reIativeIy poor prognosis in those under forty (if not a coincidence) is probably due to the more maIignant his- toIogic type of growth in younger patients, and in those over the age of fifty to the inabiIity of oIder patients to withstand the comphcations of treatment.

Sex. As in most forms of oral and pharyngeaI cancer treated by radiation, the prognosis in femaIes (26 per cent) is definiteIy better than in maIes (I 6 per cent). It is our opinion that such findings are due to the fact that cancer in the female is more radiosensitive than its anatomic and histoIogic counterpart in the maIe.

Size of the Lesion. As might be expected, the cure rate decreases as the size of the lesion increases. It is noteworthy that in the tweIve cases of our series in which the lesion was 2 cm. or less in diameter, four (33 per cent) survived five years, a cure rate twice the average for the whoIe group.

Metastases. It is apparent that its capacity to metastatsize is the most significant factor in the maIignancy of tonsiliar cancer. When no metastases were present on admission or when none occurred at any time, the cure rate is aImost 40 per cent. When metastases were present on admission, the cure rate is onIy 8 per cent.

Histopatbology. From an anaIysis of our series it appears that the prognosis in cancer of the tonsil is better in transitional ceI1 carcinoma and Iymphosarcoma than in squamous carcinoma. Under radiation therapy this difference might be accounted for, at Ieast in part, by the higher radiosensitivity of these groups. However, IymphoepitheIioma which is as radiosensitive as the two above men-

1354


Coned, gives the poorest prognosis of a11 (onIy about one-third of the two best groups). Such figures may be due in part to coincidence.

Status of the Disease on Admission. As might be expected, the cure rate in patients who had received no treatment before admission was better (19 per cent) than in those which had been treated elsewhere (14 per cent). Previous unsuccessful treatment by radiation or surgery does not precIude success, however, as is shown by an eventua1 saIvage of four (14 per cent) of thirty recurrent cases.

END RESULTS

The net five-year cure rate in the present series caIcuIated on 148 determinate cases is 18 per cent (Table rv). We have previously

TABLE IV

MEMORIAL HOSPITAL

FIVE-YEAR END-RESULTS IN CANCER OF’ THE TONSIL

1931 to 1935, IncIusive

This series consists of alI patients with histological1.y proved cases of cancer of the tonsil, both early and advanced, admitted during the specified period. OnIy those

patients are excIuded who, for any reason, were unabIe to return for treatment, paIliation and observation in the out-patient department, and those who were lost track of within

the first month after no more than one or two visits (clinic shoppers).

TotaI number of cases. . . ___~_~_ ~__._____ __..__

indeterminate group : Dead as a resuIt of other causes and without recurrence

Lost track of without recurrence.. . . . . . . Total number of indeterminate resuIts.. . . .

Determinate group: TotaI number minus those of indeterminate group. .

FaiIures: Dead as a result of cancer. L.ost track of with disease (probabIy dead).

Living with recurrence. . . . . . . . . . . . TotaI number of faiIures in treatment. . _ . .

SuccessfuI resuIts: Free from disease after five years or more. .

Five-year end-resuits:

Successful results divided by determinate group (20: r_+ttl) .

r 157

7 2

9

148 I_

117 5 0

122

26

I 8 c/;:

pubhshed the method of caIcuIation of end results which is now used at the Memorial HospitaI. Brieffy stated, the cIinica1 materia1 is

iI361


unseIected; that is, a11 patients who appIy at the cIinic is a11 stages of the disease are accepted for treatment or paIIiation and are included for statistica analysis. The onIy cases excluded are those of patients who, on their own decision, do not return for treatment and are Iost track of within the first month. At Ieast a go per cent foIIow-up is insisted upon. In caIcuIating the net end resuIts, those cases are subtracted as being indeterminate when the patients are Iost track of or die of other causes not connected with cancer or its treatment after at Ieast a year’s freedom from disease.

Duffy15 has reported on two series of tonsiIIar cancer which comprise the cases on the Head and Neck Service at MemoriaI HospitaI from Igzo to 1930, with cure rates about the same as in the present report. From a comparison of the percentage of metastases on admission (60 per cent) which he reports in these earIier cases, it appears that prior to 1930 there was some seIection of patients, since in the present series the percentage of metastases on admission, 75 per cent, indicates a more advanced average stage of the disease.

There are few reports in the Iiterature which are caIcuIated upon unseIected groups of histoIogicaIIy proved cases with a five-year survivaI period as the basis of cure. UnfortunateIy, even in the best reports the authors seIdom state definiteIy that the cases are un- seIected and that none was excIuded because of an advanced stage of the disease. Patterson,40 in 1934, reported nine five-year survivaIs in forty-one proved cases of cancer of the tonsil treated by endothermy. Coutard,ll in 1932, reported 18 per cent five-year cures in thirty-three cases. Mathey-Cornet33 reported I 5 per cent five-year cures in seventy-seven cases. None of these authors definitely states that the cases were unseIected. Schreiner reports two five-year cures in fifty-one cases, or 4 per cent.

SUMMARY AND CONCLUSIONS

The series herein reported consists of 157 consecutive unseIected cases of all patients with cancer of the tonsil, in a11 stages of the disease, admitted to MemoriaI Hospital during the years 1931 to 1935, incIusive. The etioIogy, cIinicaI course, method of treatment, end resuIts and prognosis are discussed in detai1. The net five-year cure rate is 18 per cent.

REFERENCES

J. v. BERGMAN, E. System of PracticaI Surgery, Vol. I. TransIated by W. T. BuII. New York, 1904. Lea Brothers & Co.

U37R

MARTIN, SUGARBAKER-CANCER OF TONSIL I95

2. BERVEN, E. G. E. MaIignant tumors of the tonsil; ctinicaI study with speciaI refer-

ence to radioIogicaI treatment. Acta radial. Supp. XI, I, 1931. BERVEN, E. G. E. DeveIopment of technique and results in treatment of tumors of

oraI and nasaI cavities, Am. J. Roentgenol., 28: 332, 1932.

3. BRODBIZS, A. Carcinoma of the mouth; types and degrees of maIignancy. Am. J.

Roentgenol., 17: 90, 1927. 4. BURNAM, C. F. Diagnosis and treatment of malignant tonsi conditions. Surg.,

Gynec. &+ Obst., 55: 633, 1932. 5. BUTIAN, H. T. The Operative Surgery of Mahgnant Disease. PhiIadeIphia, 1887.

P. BIakiston, Son & Co. 6. CANUYT, G. Le traitement du cancer de I’amygdale. Ann. de l’mai. oreille et larynx,

48: 724, 1929. 7. CHEEVER, D. W. Cancer of the tonsi1; removal of the tumor by external incision.

Boston. M. & S. J., 99: 133, 1878. 8. CHRISTOFORIDIS, M. Ueber TonsiIIartumoren und deren BehandIung mit Rant-

genstrahIen. Ztscbr. J. Hal+. Nasen- u. Obrenb., 22: 554, 1929. 9. CITELLI, S. Sur I’existence de IeucopIacie typique sur IFS amygdaIes et Ie cornet

inferieur, Abst. Presse med., 37: 1465, 1929.

IO. COSTOLOW, W. E. Malignancy of the tonsil. Am. J. Surg., 23: 528, 1934.

I I. COUTARD, H. Roentgen therapy of epitheIiomas of the tonsihar region, hypopharynx and Iarynx from rgzo to 1926. Am. J. Roentgenol., 28: 3 I 3, 1932.

12. CUTLER, M. Radium dosage and technique in carcinoma of the tonsi1, pharynx and Larynx. Am. J. Roentgenol., 33: 690, 1935.

13. DESPONS, J. Cancer de I’amygdaIe. Rev. de Laryngol., 49: 305, 339, 1928.

14. DUCLIING, J. and DUCUING, L. Les tumeurs mahgnes de I’amygdale paIatine. Ann.

d’oto-laryng., I : 20, 1935. 15. DUFFY, J. J. Carcinoma of the tonsil. New York State J. M., 34: 865, 1934.

16. EC~GERS, C. PracticaI management of mahgnancies of the tons& Am. J. SW-~.,

30: 254, 1935. I 7. EWING, J. LymphoepitheIioma. Am. J. Path., 5 : 99, 1929.

18. FIGI, F. A. The treatment of maIignant tumors of the mouth and throat. Am. J.

Roentgenol., 23: 648, 1930.

‘9. HARMER, W. 0. The relative vaIue of radiotherapy in the treatment of cancer of

the upper air passages. Lancet, 2: I 057, I 93 I. 20. JACOBSON, N. Contributions to the surgery of malignant disease of the prostate

gland and of the tons&. Ann. Surg., 33: 269, 190 r.

21. JANEWAY, H. H. Treatment of cancer by buried emanation. Am. J. Roentgenol.,

7: 92, 1920. 22. JCTUL, J. Protracted fractionated roentgen treatment of mahgnant tumors ad

modum Coutard. Acta radial., 17: 209, 1936.

23. KISFALUDY, P. Uber die kombinierte Behandlung des TonsiIIenkrebses. Strablen-

tberapie, 55: 429, 1936 24. KOCHER. Ueber RadicaIheitung des Krebses. Deutscrbe Ztscbr. f. Chir., 13: 134,

1880. 25. KRONLEIN. Uber Pharynxcarcinom und Pharynxexst.irpation. &tins Reztr. zur klin.

Cbir., 18: 61, 1897. 26. v. LANGENBECK, B. Uber die Exstirpation des Pharynx. Verbandl. der deutscb. Gesel-

lscb. j_ Cbir., 8: I 15, 1879.

27. LEVIE, B. CIassification and treatment of maIignant tumors of the tonsirs. Acta

OtO-h?Jf?Lg., 25 : 403, 1937. 28. MALLET, L. Uber die t.ranskutane Curie-therapie der TonsilIenkrebse. Strablen-

therapie, 53: 54, 1935-


29.

30. 3’.

32.

33.

34.

35.

36.

37,

38.

39,

40.

41. 42.

43,

44. 45.

46.

47.

48.

49. $0.

51.

52. 53.

a. MARTIN, H. E. PeroraI x-radiation in the treatment of intraorai cancer. Radiol.,

28: 527, 1937. 6. MARTIN, H. E. and SUGARBAKER, E. L. Cancer of the floor of the mouth. Surg,,

Gynec. * Obsr. In press. c. MARTIN, H. E. and BLADY, J. V. Cancer of the nasopharynx. Arch. Otolaryngol.

In press.

d. MARTIN, H. E. and ELLIS, E. B. Aspiration biopsy. Surg., Gynec. @Y Obst., 59:

578, 1934. MATHEWS, J. MaIignant tumors of the tonsiI. LarJrngoscope, 22: 737, 1912. MATTICK, W. L. EpitheIioma of the tonsil; study of 162 cases. New York State J.

M., 39: 1412, 1939. MCCOY, J. Surgical treatment of cancer of the tonsil, with a report of cases. Laryngo-

scope, 29: 422, 1919. MATHEY-CORNAT, R. Les methodes actuelIes d’irradiation des tumeurs mabgnes de

I’amygdaIe et kurs resultats. Paris med., I : 237, 1938.

MAUREL, G. and WEILL, R. Indications et technique du traitement des cancers de I’amygdaIe. J. de med. de Paris, 56: 217, 1936.

MELENEY, F. L. Treatment of traumatic wounds with zinc peroxide. New York State

J. Med., 39: 2188, 1939. MELLER, A. Zur Statistik der Schleimhautcarcinome des Mundes und Rachens.

Deutsche Ztschr. j. C&r., 84: 105, 19o6.

h~naxxz, J. Die seitliche Pharyngotomie behufs Exstirpation maligner GeschwtiIste

der Tonsillargegend. Deutsche med. Wcbns+r., 12: 157, 1886. NATHANSON, E. A. Cancer of the tonsiIs. Arch. Clin. Cancer Res., 3: 181, 1927-1928.

NEW, G. B. and CHILDREY, J. H. Tumors of tonsi and pharynx. Arch. Otolaryngol.,

14: 699, 713, 1931. PATTERSON, N. Use of the diathermy knife in mabgnant disease of the mouth, nose

and pharynx. Lancet, 2: 633, 1934. QUICK, D. Treatment of maIignant neopIasms of the tonsils. J. Radial., 3: 173, rgzz.

REGAUD, C. and CREMIEN, R. Sur Ia suppression definitive du tissue thymique par Ia

rontgentherapie. Compt. rend. Sot. de Viol., 72: 523, 1912. RICHARDS, G. E. The radioIogica1 treatment of cancer: methods and results 1928-

1935. III. MaIignant Iesions of the ton&I and its piIIars. Cunad. M. A. J., 35:

385, ‘936. SCHALL, L. A. Carcinoma of the tonsil. New England J. M., 21 I : 997, 1934.

SCHMINCKE, A. Uber Iymphoepithebale GeschwiiIste. Beitr. z. path. Anat. u. z.

allg. Path., 68: 161, 1921. SEDILLOT. Nouveau procede en extirpation de IR Iangue. Cax. med. de Paris, I 2: 126,

1844. SIMEONI, C. SuI trattamento fisico dei tumori maligni deIIe tonsilIe paIatine. Arc/~.

ital. di otol., 43: 612, 1932. SIMMONS, C. Cancer of the bucca1 mucosa. Ann. Surg., 92: 68 I, 1930~

SUTTON, J. B. Tumors Innocent and Malignant. London, 1893. CasseII & Co. SUNDERLAND, D. A. and BIP~~KI,EY, J. S. The use of zinc peroxide in necrotic tumors

and radionecrosis. Radiol. In Press.

TROTTER, W. Operation for maIignant tumor of the pharynx. Brit. I&Z. J., 16: 485, 1929.

USILTON, P. J. Persona1 communication to the authors. WURTZ, W. J. M. MaIignant disease of the nose and throat with specia1 reference to

cancer of the nasal fossae. nasopharynx and tonsil. New York M. J., I I 3 : 434, I 92 I.

u391

Date post:	01-Dec-2016
Category:	Documents
Upload:	hayes-martin
View:	230 times
Download:	6 times

Cancer of the tonsil

Documents