as to the action of the drugs that are used for its treatment.

It is here that a proper point of view is of the utmost importance as a means of orienting therapeutic investigations. In the first place, there are a few simple facts concerning the biology of syphilis that it would be well for all of us to bear in mind. We should remember that syphilis begins insidiously and that the infecting organism is widely distributed before any clinical manifestations of disease can be detected. In fact, there are many patients in whom infection goes unsuspected for months or even years due to the absence of any outward manifestation of disease, and it is rare that the early lesions of syphilis progress without interruption for any considerable period of time even though little or no treatment is given. Moreover, the character and severity of the disease may vary according to the age, sex, or race of the patient as well as with changes in physiological or pathological states. Finally, it should be remembered that while syphilis is a prime cause of ill health or disability and is the indirect cause of many deaths, it is rarely the immediate cause of death in treated or untreated patients, except in infants and as an ultimate result of involvement of the cardiovascular or central nervous system, even though the infection may persist for the duration of life.

These facts, and others of similar character, possess an important therapeutic significance in that they give some indication of what may or may not be accomplished and of methods. that may be employed with the hope of obtaining a favorable influence on

the course of disease. For example, it would appear that while the infecting organism is an essential factor in the causation of syphilis it is not necessarily the dominant factor in determining the course of this disease; that, from the standpoint of control of the infection the natural resources of the patient are of great importance; that until we possess therapeutic agents that are sufficiently powerful to insure an absolute sterilization there will remain two primary considerations in the treatment of syphilis, namely, the patient and the parasite, and that there are two possible methods of approach, the one direct and the other indirect. It is obvious, therefore, that therapeutic agents or procedures may be evaluated from either of these points of view but that a comprehensive evaluation must include both.

Until recently it was impossible to carry out experimental investigations embracing the possibilities indicated above due to our lack of knowledge of experimental syphilis. Hence, available data on the action of the therapeutic agents now in use are extremely meagre. Innumerable estimations of trypanocidal and spirocheticidal action and of toxicity have been made together with exhaustive studies of absorption, distribution, and elimination. But, nowhere is there to be found a comprehensive report of the biological action of even such substances as arsphenamine and neoarsphenamine. The early reports dealt largely with spirocheticidal action and toxicity and the determination of the so-called therapeutic index, but at the time these substances were first investigated it was not possible to determine even their spirocheticidal action in syphilitic ani

mals with any degree of certainty. The original estimates of therapeutic activity were based upon the disappearance of spirochetes, the healing of primary lesions, and the absence of recurrence within a certain period of time which, as we now know, was insufficient to warrant any very definite conclusions. In addition, some stress was laid upon the speed of reactions which may or may not have been important.

It is only within the last few years that methods have been devised for studying actual curative effects and for studying therapeutic effects other than spirocheticidal action which are of the foremost importance. For the first time, these features of drug action are now receiving attention. If, therefore, we consider the relative merits and the particular indications for the use of one drug in preference to another there is very little to guide

us.

Most of our knowledge concerns parasiticidal action and even this is inaccurate.

If we define the immediate object in the treatment of syphilis as an attempt to eradicate the infection, or failing in this, to exercise an effective control over the infection and to preserve or restore the health of the patient, it is obvious that, in a disease as varied and complex as syphilis, the accomplishment of these ends may call for the use of a variety of therapeutic procedures or the exhibition of many different kinds of action depending upon the conditions that exist at the time treatment is undertaken. During the early stages of acquired syphilis our efforts may be concentrated on the destruction of the parasites, but once a characteristic lesion has been

developed it is useless to attempt to prevent a systemic distribution of organisms; every part of the body is potentially infected, but it is still possible to prevent the localization and growth of spirochetes in inaccessible foci and to prevent the development of lesions which afford more or less protection against the action of spirocheticidal agents. Fortunately, nature aids us in this in a most effective way through the adaptation of different tissues to the growth of spirochetes. Thus, the first focal infections occur in tissues that offer no great hindrance to drug action, and with the exception of the chancre, the early lesions are readily permeated by therapeutic agents.

In very early cases of syphilis the spirocheticidal action of the drug is, therefore, a consideration of the foremost importance. For the most part, spirochetes are in exposed positions and can be destroyed most readily by substances that possess the greatest spirocheticidal action. The effect proceeds almost like a test tube reaction. Still, we cannot overlook the fact that a certain number of organisms are not readily accessible even in the earliest cases of syphilis. Hence, other things being equal, the choice of agents may be determined by other qualities of action such as penetrability or the ability of the drug to induce or to facilitate resolution of lesions as a means of exposing protected organisms. These properties are not necessarily commensurate with spirocheticidal action as ordinarily determined. There are many substances that possess a high parasiticidal action that are wholly incapable of developing this action except where

organisms are readily accessible, as in the blood stream, while other substances with much lower parasiticidal action may prove more effective on account of their ability to reach hidden foci of infection. In like manner, there are substances that will induce rapid resolution of lesions which do not possess a high degree of spirocheticidal action, and conversely there are other drugs with high spirocheticidal action which do not cause a correspondingly rapid resolution of lesions. Arsphenamine and neoarsphenamine show distinct differences in these respects.

When we come to more advanced stages of syphilitic infection, congenital as well as acquired, treatment is more complex and the difficulties are greatly increased. Organisms have become established in such inaccessible locations as the cardiovascular system and the central nervous system, while the lesions in other parts of the body have assumed an obstructive character. In addition, the health of the patient has begun to suffer either as a direct result of the infection or of the strain imposed by a continuous effort to combat the infection. In either case, processes are in operation which, if unrelieved, may lead to serious consequences.

When this point is reached, the hope of effecting a complete sterilization is greatly diminished and our efforts should be directed in such a way as to insure an effective control of the infection and the preservation of the health of the patient. Under these circumstances the potential parasiticidal action of a drug is of less consequence than the ability of the drug to reach foci of infection that are difficult of

access and the effect produced on the patient. In part, this means penetrability and, in part, an ability to induce resolution of lesions, for, regardless of how great an apparent spirocheticidal action a drug may have, its value is limited to what it can actually accomplish under the conditions of its use, and not infrequently the ultimate result may depend on the response that can be elicited from the patient.

None of the agents at present available is equally adapted to the accomplishment of these several ends. Those that are best suited to the accomplishment of one purpose are not equally adapted to another. Hence, in cases of advanced syphilis, there are definite indications for a mixed therapy, but it is here that our knowledge. of the action of therapeutic agents breaks down.

Until recently, experimental investigations have made no provision for this contingency. It was assumed that parasiticidal action was the all important thing and few attempts have been made to estimate the therapeutic value of antisyphilitic remedies from any other point of view. Still, when Ehrlich expressed his disappointments over the outcome of some of his earlier investigations, he recognized the fact that deductions drawn from studies of parasiticidal action and toxicity were frequently misleading, and recently Forneau and his coworkers at the Pasteur Institute have been greatly disturbed by this same lack of agreement between experimental and clinical results. As a matter of fact, this has been the great obstacle in the path of chemotherapeutic investigators but, unfortunately, the tendency has been to dis

investigations rather than to remedy the cause of such discrepancies as have been noted by the development and use of methods that can be relied upon to yield results that are in agreement with clinical experience.

It is obvious that if vigorous treatment with parasiticidal agents is begun in the early stages of syphilis and continued systematically without accomplishing the desired result, there is little hope that persistence in the use of the same methods may prove successful when the obstacles to success have been greatly increased. It has been clearly demonstrated that the hope of success diminishes with time and with the progress of disease, and that until more effective parsiticidal agents are available an effort should be made to supplement the action of such drugs by the use of other measures, or that the entire system of treatment should be recast with a view to the use of indirect methods of control. Many clinicians have long since reached this conclusion, and it is here that the treatment of the patient becomes a consideration of the foremost importance.2

At no time can we disregard the patient. Every dose of drug that is administered affects the patient in one way or another, and it is imperative that nothing should be done that will render the patient less capable of combating his disease should treatment be discontinued before the infec

The authors have recently had the opportunity of reading an unpublished paper by Dr. W. A. Pusey who for years has held views on the treatment of syphilis that agree in many respects with those expressed in this paper.

Unfortunately, we know too little concerning either of these questions, but there are a few facts that may be referred to briefly in this connection. In the first place, it may be stated as a general principle that all therapeutic agents that cause an abrupt cessation of the reaction on the part of the patient, as indicated by a rapid resolution of lesions, but fail to destroy the infecting organisms, operate to the disadvantage of the patient and predispose to the subsequent occurrence of more severe manifestations of disease. All of the highly active parasiticidal agents appear to possess this disadvantage to a greater or less degree, and as a rule, this feature of action is proportional to the difference between the direct spirocheticidal action of the drug and its ability to induce resolution of lesions. In part, this effect is attributable merely to removal of the stimulus to reaction resulting from actual destruction or from temporary suppression of the activity of the spirochetes. In other instances, however, there is a definite and more or less persistent reduction in the effectiveness of the reaction, due to the action of the drug on the patient.

Moreover, in instances of this kind, the earlier the treatment is undertaken the more serious are the consequences of a failure on account of interference with immunological reactions. For example, if treatment is instituted at

a time prior to the development of an immunity that is sufficient to bring the infection under control, the progress of immunological reactions is interrupted and resistance promptly returns to an essentially normal level or may become subnormal for a time. If, however, the course of events is not interfered with until later, the resistance that has been acquired is more stable and tends to persist for some time even after the withdrawal of the stimulus to reaction. Moreover, in animals, it is possible to regulate treatment in such a way as to leave the animal highly refractory to subsequent inoculation in spite of the fact that no infection can be demonstrated in such animals.3 Furthermore, as will be brought out later, similar changes in susceptibility occur naturally or can be induced experimentally by the use of a variety of procedures.

Mercury and the iodides are of interest from this point of view. As strange as it may seem, we have no definite knowledge of the manner in which these drugs produce their specific effects. It has been assumed that mercury acts as a spirocheticidal agent and its action has been largely studied from this point of view. But, the demonstrable spirocheticidal action of the preparations that have been used most extensively is so slight that it is safe to assume that the entire

'In this connection, the reader is referred to a recent paper by Chesney and Kemp on Experimental observations on the "cure" of syphilis in the rabbit with arsphenamine, Jour. Exper. Med., 1924, xxxix, 553.

4 This does not refer to the action of iodides on gummatous lesions for which an explanation has been offered by Jobling and Peterson.

effect is not due to any primary parasiticidal action. The indications are that the beneficial effects are in some way referable to an action on the patient.

In the case of the iodides a similar condition exists. As far as we know the iodides have no direct effect on the spirochetes. Still, they have proven to be of some advantage in the treatment of syphilis, and experimentally it is possible to shorten the course of disease and to diminish its severity without the use of any other therapeutic measure. In this case, we are forced to assume an indirect action or a reinforcement of natural processes of resistance.

This field of therapeutic investigation is almost unexplored. We know comparatively little concerning the mechanism of defense and it is impossible to say what may be accomplished. by an effort to treat syphilis from this point of view. We do know, however, that a profound influence can be exerted upon the course of disease in experimental animals by means other than the use of spirocheticidal agents and that a high degree of resistance can be developed and maintained for a considerable period of time independent of any existing syphilitic infection.

Such factors as age, sex, or breed are sufficient in themselves to exercise a very potent influence on the course of syphilitic infections. Moreover, there are remarkable differences in the character of the disease that may be produced by the use of a given procedure at different seasons of the year or in different years, and these variations occur in spite of every effort to main

Unpublished experiments of Louise. Pearce and C. M. Van Allen.