16th Edition
`
`HARR|SON’S
`PRINCIPLES OF
`
`Internal
`Medicine
`
`Editors
`
`DENNIS L. KASPER, MD
`William Ellery Channing Professor of Medicine,
`Professor of Microbiology and Molecular Genetics,
`Harvard Medical School; Director, Channing
`Laboratory, Department of Medicine, Brigham and
`Women’s Hospital, Boston
`
`ANTHONY S. FAUCI, MD
`Chief, Laboratory of Immunoregulation; Director,
`National Institute of Allergy and Infectious Diseases,
`National Institutes of Health, Bethesda
`
`DAN L. LONGO, MD
`Scientific Director, National Institute on Aging,
`National Institutes of Health,
`Bethesda and Baltimore
`
`EUGENE BRAUNWALD, MD
`Distinguished Hersey Professor of Medicine,
`Harvard Medical School; Chairman, TIMI Study Group,
`Brigham and Women’s Hospital, Boston
`
`STEPHEN L. HAUSER, MD
`Robert A. Fishman Distinguished Professor and Chairman,
`Department of Neurology,
`University of Califomia San Francisco, San Francisco
`
`J. LARRY JAMESON, MD, PHD
`Irving S. Cutter Professor and Chairman,
`Department of Medicine,
`Northwestern University Feinberg School of Medicine;
`Physician-in—Chief, Northwestern
`Memorial Hospital, Chicago
`
`McGraw-Hill
`
`MEDICAL PUBLISHING DIVISION
`
`AMERIGEN 1025
`
`New York
`Milan
`
`Mexico City
`
`Chicago
`New Delhi
`
`San Francisco
`San Juan
`
`Lisbon
`Seoul
`
`London
`
`Madrid
`
`Singapore
`
`Sydney
`
`Toronto
`
`

`
`%
`
`76th Edition
`
`Kasper Braunwald
`
`Fauci?
`
`Hauser
`
`Longo
`
`Jameson
`
`‘
`
`9 ?
`
`E
`
`‘
`
`
`
`
`
`
`
`
`
`
`
`— x L;as.:.-'-._;-;s.~..~1.»-.g...-..=.....,u..:,.~.. _.".._.v.-A.a.-.:..v_;.c;o.‘:2flaa " ‘~‘~ ' "~;=x-£,ws4v «*~-.'-,5“- ' ' ......v
`
`~
`
`.:..c...-_._-:__-«.m..-.-,.—.---
`
`

`
` The McGraw-Hill Companies
`
`Note: Dr. Fauci’s and Dr. Longo’s works as editors and authors were performed outside
`the scope of their employment as U.S. government employees. These works represent their
`personal and professional views and not necessarily those of the U.S. government.
`
`Harrison’s
`PRINCIPLES OF INTERNAL MEDICINE
`Sixteenth Edition
`
`Copyright © 2005. 2001, 1998, 1994, 1991, 1987, 1983, 1980, 1977, 1974, 1970, 1966, 1962, 1958 by The
`McGraw-Hill Companies, Inc. All rights reserved. Printed in the United States of America. Except as per-
`mitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or
`distributed in any form or by any means, or stored in a data base or retrieval system, without the prior written
`permission of the publisher.
`
`1234567890 DOWDOW 0987654
`
`ISBN 0-07—139140-l (Set)
`ISBN 0-07-l3914l—X (Vol. 1)
`
`ISBN 0-07—l40235—7 (Combo)
`
`ISBN 0-07-139142—8 (Vol. H)
`
`FOREIGN LANGUAGE EDITIONS
`Arabic (l3e): McGraw-Hill Libri Italia srl
`(est. 1996)
`Chinese Long Form (15e): McGraw-Hill Inter-
`national Enterprises, Inc., Taiwan
`Chinese Short Form (15e): McGraw-Hill Edu-
`cation (Asia), Singapore
`Croatian (l3e): Placebo, Split, Croatia
`French (15e): Medecine—Sciences Flammarion,
`Paris, France
`German (15e): ABW Wissenschaftsverlagsge—
`sellschaft mbH, Berlin, Germany
`Greek (15e): Parissianos, S.A., Athens, Greece
`Italian (15e): The McGraw-Hill Companies, Srl,
`Milan, Italy
`Japanese (15e): MEDSI—Medical Sciences Inter-
`national Ltd, Tokyo, Japan
`
`Korean (15e): McGraw-Hill Korea, lnc., Seoul,
`Korea
`
`Polish (14e): Czelej Publishing Company, Lu-
`bin, Poland (est. 2000)
`Portuguese (15e): McGraw-Hill Interamericana
`do Brazil, Rio de Janeiro, Brazil
`Romanian (l4e): Teora Publishers, Bucharest,
`Romania (est. 2000)
`Serbian (15e): Publishing House Romanov,
`Bosnia & Herzegovina. Republic of Serbska
`Spanish (15e): McGraw—Hil1Interamericana de
`Espana S.A., Madrid, Spain
`Turkish (15e): Nobel Tip Kitabevleri, Ltd., Is-
`tanbul, Turkey
`Vietnamese (15e): McGraw-Hill Education
`(Asia), Singapore
`
`This book was set in Times Roman by Progressive Information Technologies. The editors were Martin
`Wonsiewicz and Mariapaz Ramos Englis. The production director was Robert Laffler. The index was pre-
`pared by Barbara Littlewood. The text designer was Marsha Cohen/Parallelogram Graphics. Art director:
`Libby Pisacreta; cover design by Janice Bielawa. Medical illustrator: Jay McElroy, MAMS.
`
`R. R. Donnelley and Sons, Inc., was the printer and binder.
`
`Cover illustrations courtesy of Raymond J. Gibbons, MD; George V. Kelvin; Robert S. Hillrnan, MD; and
`Marilu Gomo—Tempini. MD.
`
`Library of Congress Cataloging-in-Publication Data
`
`.
`
`.
`
`[et al.].
`
`p. cm.
`
`Harrison’s principles of internal medicine— 16th ed./editors, Dennis L. Kasper .
`Includes bibliographical references and index.
`ISBN 0-07-139140-1 (set)—ISBN 0-07-139141-X (v. l)—-ISBN 0-07-139142-8 (v. 2)—ISBN 0-07-
`140235-7 (combo)
`1. Internal medicine. I. Title: Principles of internal medicine. II. Kasper, Dennis L. III. Harrison,
`Tinsley Randolph, 1900- Principles of internal medicine.
`[DNLM: 1. Internal Medicine. WB 115 H322 2005]
`RC46.H333 2005
`6l6—dc2l
`
`
`
`200404493 1
`
`

`
`81 Huperplastic and Malignant Diseases of the Prostate
`
`547
`
`strictures requiring dilatation developed in 1% of cases, all of whom
`had undergone a TURP. Pooled data show that the frequency of grade
`3 to 4 toxicities is 6.9 and 3.5%, respectively, for patients who received
`>70 Gy. The frequency of erectile dysfunction is related to the quality
`of erections pretreatment, the dose administered, and the time of as-
`sessment. The etiology is related to a disruption of the vascular supply
`and not the nerve fibers.
`
`Neoadjuvant hormone therapy has also been studied in combina-
`tion with radiation therapy to increase local control rates, decrease the
`size of the prostate so that the exposure of normal tissues to full-dose
`radiation is reduced, and decrease the rate of systemic failure. Short-
`terrn hormone exposures can reduce toxicities and improve local con-
`trol rates, but long-term (2 to 3 years) treatment is needed to prolong
`the time to PSA failure and the development of metastatic disease. The
`impact on survival has been less clear.
`Brachytherapy involves the direct implantation of the prostate with
`radioactive sources. It is based on the principle that the deposition of
`radiation energy in tissues decreases exponentially as a function of the
`square of the distance from the source. The goal is to deliver intensive
`irradiation to the prostate, minimizing the exposure of the surrounding
`tissues. Techniques have evolved from intraoperative manual insertion
`methods to the current standard, in which customized templates based
`on CT and ultrasonographic assessment of the tumor are used for seed
`placement based on computer-optimized dosimetry to achieve more
`homogeneous dose distributions. The implants themselves are now
`performed transperineally, without an open procedure, with real-timc
`imaging. The result is a marked reduction in local failure rates with
`fewer complications. In a series of 197 patients followed for a median
`of 3 years, 5-year actuarial PSA relapse—free survival for patients with
`pretherapy PSA levels of 0 to 4, 4 to 10, and >10 pg/mL were 98,
`90 and 89%, respectively. In a separate report of 201 patients who
`underwent posttreatment biopsies, 80% were negative, 17% indeter-
`minate, and 3% were positive. The results did not change with longer
`follow-up. Nevertheless, many physicians feel that implantation is best
`reserved for patients with good or intermediate prognostic features.
`The procedure is well tolerated, although most patients experience
`urinary frequency and urgency that can persist for several months.
`Incontinence has been seen in 2 to 4% of cases. Higher complication
`rates are observed in patients who have undergone a prior TURP or
`who have obstructive symptoms at baseline. Proctitis has been reported
`in <2% of patients.
`Watchful waiting, or deferred therapy, is a policy of no therapeutic
`intervention(s) until the tumor progresses. Progression can be based
`on PSA changes, local tumor growth, the development of symptoms,
`or metastatic disease. The practice evolved from studies of predomi-
`nantly elderly men with well-differentiated tumors in whom clinically
`significant progression could not be demonstrated for protracted pe-
`riods, during which a significant proportion died of intercurrent dis-
`ease. In a structured literature review of patients treated by radical
`surgery, a deferred approach, or external beam radiation, the 10-year
`mean survivals were 93% for radical prostatectomy, 84%'for deferred
`treatment, and 74% for external beam radiation. Risk of progression
`was related to grade. Men with grade 1 or 2 umors had a 13% risk of
`death and 19% risk of metastases at 10 years; those with grade 3
`tumors had 63 and 74% risks, respectively.
`Case selection is critical, and the criteria to select those to whom
`watchful waiting can be applied safely are under intense study. In a
`recent prostatectomy series, it was estimated that 10 to 15% of patients
`had “insignificant” cancers. Given the multifocality of the disease, a
`concern is the limited ability to predict pathologic findings on the basis
`of a needle biopsy, even when multiple cores are obtained. Arguing
`against this approach is the result of a randomized trial of radical
`prostatectomy vs. watchful waiting from Sweden. With a median fol-
`low-up of 6.2 years, men treated by radical surgery had a lower risk
`of prostate cancer death relative to watchful waiting patients (4.6 vs.
`8.9%) and a lower risk of metastatic progression, hazard ratio .63.
`
`- emes can be predicted using postoperative nomograms that con-
`" pretreatment factors and the pathologic findings at surgery. PSA
`” is defined as a detectable value of 0.2 or 0.4 ng/mL, although
`exact definition varies among series. The techniques continue to
`_ ve as the ability to localize the tumor within or beyond the pros-
`are refined with different biopsy algorithms and with imaging. The
`<4 it is better case selection and surgical planning, which in turn have
`—1 to more rapid recovery and higher rates of continence and potency.
`- wrs associated with incontinence include older age, shorter urethra
`..n.u surgical technique, preservation of neurovascular bundles, and
`1 lopment of an anastomotic stricture. Surgical experience is also a
`_ 7r. In one series, 6% of patients had mild stress urinary inconti-
`.31 (SUI) (requiring 1 pad/day), 2% moderate SUI (>1 pad/day),
`'- 0.3% severe SUI (requiring an artificial urinary sphincter). At 1
`an . 92% were completely continent. In contrast, the results in a Med-
`= -z . population treated at multiple centers showed that at 3, 12, and
`zronths following surgery, 58, 35, and 42% wore pads in their
`2 ..
`year, and 24, 11, and 15% reported “a lot” of urine leakage.
`.
`.ors associated with recovery of erectile function include younger
`» . quality erections before surgery, and the absence of damage to
`,
`: neurovascular bundles. Erectile function returns in a median of 4
`' 6 months if both bundles are preserved. Potency is reduced by half
`E least one nerve bundle is sacrificed. In cases where cancer control
`
`
`
`;.ires the removal of both bundles, sural nerve grafts are being ex-
`‘ ,-. Overall, with the availability of drugs such as sildenafil, in-
`I Hi ~.thral inserts of alprostadil, and intracavernosal injections of va-
`réilators, many patients recover satisfactory sexual function.
`’ High-risk patients are those with a predicted high probability of
`lfilare with surgery alone based on pretreatment factors. In these sit-
`ions, nomograms and predictive models can only go so far. Exactly
`w°:;=_t probability of success or failure would lead a physician to rec-
`mznend and a patient to seek alternative approaches is controversial.
`Fzr example, it may be appropriate to recommend radical surgery for
`it younger patient with a low probability of cure. To improve the out-
`l azcres of surgery for high-risk patients, neoadjuvant hormonal therapy
`its been explored. The results of several large trials testing 3 or 8
`>E'.‘¥3Ll1S of androgen ablation before surgery showed that serum PSA
`R-rels decreased by 96%, prostate volumes reduced by 34%, and mar-
`positivity rates declined from 41 to 17%. Unfortunately, hormones
`éixrl not produce an improvement in PSA relapse—free survival. Thus,
`zeoadjuvant hormonal therapy is not recommended.
`
`Elation Therapy Radiation therapy is given by external beam, the
`izplantation of radioactive sources into the gland, or a combination
`5:’ both. Contemporary external beam radiation techniques now use
`ézree-dimensional conformal treatment plans to maximize the admin-
`Isaered dose to the tumor and to minimize the exposure of the sur-
`:r:s:nding normal structures. The addition of intensity modulation
`rL\[RT) has allowed further shaping of the isodose curves and the
`éelivery of higher doses to the tumor and a further reduction in normal
`-'.5;<$ue exposure. These advances have allowed the safe administration
`of doses >80 Gy, higher local control rates, and fewer side effects.
`Overall, radiation therapy is associated with a higher frequency of
`bowel complications (mainly diarrhea) than surgery. Measures of can-
`:31‘ control include the proportion of patients who show a decline in
`PSA to <0.5 or 1 ng/mL, the proportion with “nonrising’,’ PSA values,
`or the proportion with a negative biopsy of the prostate 2 years after
`completion of treatment. PSA relapse is defined as three consecutive
`rising PSA values from the nadir value, with the time to failure as the
`=:_idpoint between the nadir and first rising value.
`Radiation dose is important. A PSA nadir of <1.0 ng/mL was
`observed in 90% of patients receiving 75.6 or 81.0 Gy vs. 76 and 56%
`for those receiving 70.2 Gy and 64.8 Gy, respectively. The positive
`isiopsy rates at 2.5 years were 4% for those treated with 81 Gy, vs. 36
`and 27% for those receiving 70.2 or 75.6 Gy. The frequency of rectal
`complications relates directly to the volume of the anterior rectal wall
`receiving full-dose treatment. Grade 3 rectal or urinary toxicities were
`seen in 2.1% of cases at a median dose of 75.6 Gy. Grade 3 urethral
`
`

`
`
`
`e
`Progestationaiantiandrogens
` 9133]
`
`5 Type I
`Pure androgens
`
`Type II
`
`
`
`
`Cholesterol
`conversion
`inhibition
`Q Hydroxylase
`inhibitors
`
`
`
`V Adrenal Glands
`
`
`
`Cholesterol
`conversion
`inhibitors
`
`free of metastases at 5 years. Factors a:
`with progression include Gleason gr:
`to recurrence, and PSA doubling til
`those with Gleason grade 28 tumors. 1
`ability of metastatic progression was
`and 71% at 3, 5, and 7 years, respec
`the time to recurrence was <2 years :
`doubling time was long (> 10 mot:
`proportion with metastatic disease we
`and 53% vs. 47, 69, and 79% if thet
`time was short (< 10 months) duringi
`time intervals. These models continue
`
`fined. A difficulty making these pretfi
`that most patients with a rising PSA
`some fortn of therapy before the dew‘:
`of metastases.
`
`METASTATIC DISEASE: NONCASTRATE M
`
`Peripheral
`organs
`
`
`'
`
`(X Fleductase inhibitors
`
`
`
`FIGURE B1-3
`
`Sites of action of different hormone therapies.
`
`.\'evertheless, it can be anticipated that more patients may be candi-
`dates for a deferred approach as PSA testing is applied more widely
`and earlier.
`
`RISING PSA This state includes patients in whom the sole manifesta-
`tion of disease is a rising PSA after surgery and/or radiation therapy.
`By definition, no evidence of disease is found on scan. For these pa-
`tients the central issue is whether the rise in PSA is the result of per-
`sistent disease in the primary site, a systemic recurrence, or both. In
`theory, disease that persists or has recurred in the primary site may be
`curable by additional local treatment. For patients who had undergone
`surgery, the question is whether external beam radiation therapy to the
`prostate bed can eliminate the disease and lead to an undetectable PSA.
`For radiation therapy—treated patients, the question is whether a pros-
`tatectomy would achieve cure.
`The decision to recommend radiation therapy is often made on
`clinical grounds, as imaging studies such as CT and bone scan are
`typically uninformative. Some recommend a Prostascint scan: imaging
`with a radiolabeled antibody to prostate-specific membrane antigen
`(PSMA), which is highly expressed on prostate epithelial cells. Anti-
`body localization to the prostatic fossa suggests local recurrence; 10-
`calization to extrapelvic sites predicts failure of radiation therapy. Oth-
`ers recommend that a biopsy of the urethrovesical anastamosis be
`obtained before considering radiation. Factors that predict for response
`to salvage radiation are a positive surgical margin, a lower Gleason
`grade, a long interval from surgery to PSA failure, a slow PSA dou-
`bling time, and a low (<0.5 to 1.0 ng/mL) PSA value at the time of
`treatment. Radiation is generally not recommended if the PSA was
`persistently elevated after surgery (indicating that disease-free status
`was not achieved).
`
`For patients with a rising PSA after radiation therapy, a salvage
`prostatectomy can be considered if the disease was “curable” at the
`
`onset. persistent disease has been documented by a biopsy of the pros-
`tate, and no metastatic disease is seen on imaging studies. Unfortu-
`nately, case selection is poorly defined in most series, and morbidities
`are significant. As currently performed, virtually all patients are im-
`potent, and ~45% have either total urinary incontinence or stress in-
`continence. Major bleeding, bladder neck contractures, and rectal in-
`jury are not uncommon.
`In the majority of cases, the rise in PSA indicates systemic disease.
`In these cases, the need for treatment should consider the probability
`of developing clinically detectable disease on scan and in what time
`frame. That immediate therapy is not required was shown in a series
`where patients did not receive systemic therapy until metastatic disease
`was documented. Overall, the median time to metastatic progression
`was 8 years and 63% of the patients with rising PSA values remained
`
`disease noncastrate refers to patients‘
`mors visible on an imaging study afil,
`trate levels of testosterone. The patia
`newly diagnosed or have recurrent difi,
`treatment for localized disease. Stanth
`ment is to block androgen action or decrease androgen p
`‘
`medical or surgical means. Over 90% of male hormones
`the testes; < 10% are synthesized in the adrenal gland. «~
`chiectomy is the “gold standard” approach but is least a
`patients. Medical therapies can be divided into those that
`tosterone levels, e.g., gonadotropin-releasing hormone ( u
`nists and antagonists, estrogens and progestational agents- -
`tiandrogens that bind to the androgen receptor but do not .,
`81-3). Ketoconazole inhibits adrenal androgen synthesis
`after first-line castration is no longer effective. In this se-
`renal glands may contribute up to 40% of the active an
`prostate.
`
`
`
`<
`
`GnRH analogues (leuprolide acetate and goserelin ~- -st
`produce a rise in luteinizing hormone and follicle-stimul u
`(FSH), followed by a downregulation of receptors in ~-
`gland, which effects a chemical castration. They were ap r
`basis of randomized comparisons showing an improved ~‘
`(specifically, reduced cardiovascular toxicities) relative to ‘
`bestrol (DES), with equivalent potency. The initial rise in ~
`may result in a clinical flare of the disease. As such, u-vv;
`contraindicated in men with significant obstructive sym
`related pain, or spinal cord compromise. Estrogens such
`lower testosterone levels but have fallen out of favor m»
`
`of vascular complications such as fluid retention, phi"
`and stroke.
`
`In contrast, nonsteroidal antiandrogens such as flu --x
`tamide, or nilutamide block the binding of androgens to
`Given alone, testosterone levels remain the same or '
`agents were approved initially to block the flare ass '
`initial rise in testosterone that results following GnRH -
`it
`They have also been studied as part of a combined an a;
`(CAB) or maximal androgen blockade (MAB) and as t
`The concept of CAB was developed to inhibit testic -xi
`androgens at the outset, and it preoccupied the field f I
`It is achieved clinically by combining an antiandrogen
`agonist or surgical orchiectomy. Cumulative results w t
`comparisons involving thousands of patients showed no
`combining an antiandrogen with surgical orchiectomy.
`analyses of trials combining an antiandrogen with a
`have shown a modest (<10%) survival advantage. .\ - ‘.
`all combined androgen blockade trials concluded t,
`benefit to the approach. In practice, most patients
`,
`analogue therapy receive an antiandrogen for the first —
`treatment.
`
`The anti—prostate cancer effects of agents that 10 ’
`
`'-
`
`
`
`

`
`
`
`terone levels are similar, and the clinical course is predictable: an
`initial response, a period of stability in which the cells are dormant
`and not proliferating, followed by regrowth after a variable period of
`time as a hormone-independent tumor. Androgen ablation is not cu-
`rative. Cells that survive castration are present when the disease is first
`diagnosed. Considered by disease manifestation, PSA levels return to
`normal in 60 to 70% of cases and measurable disease regression occurs
`in 50%; while improvements in bone scan occur in 25% of cases, the
`majority remain stable. Survival is inversely proportional to disease
`extent. Agents that lower testosterone are associated with an androgen-
`deprivation syndrome that includes hot flushes, weakness, fatigue, im-
`potence, loss of muscle mass, changes in personality, anemia, depres-
`sion, and a reduction in bone density. The bone changes can be
`prevented by treatment with bisphosphonates along with vitamin D
`and calcium supplementation.
`A question often asked is whether antiandrogens, which are asso-
`.-iated with fewer hot flashes, less of an effect on libido, less muscle
`wasting, fewer personality changes, and less bone loss, can be used
`1l0I'l€ without compromising outcomes. Gynecomastia remains a sig-
`nificant problem but can be alleviated in part with the addition of
`tamoxifen. Most reported randomized trials suggest that the cancer-
`specific outcomes are inferior. Even a comparison of bicalutamide,
`150 mg (three times the recommended dose of 50 mg), versus surgical
`castration showed a shorter time to progression and inferior survival
`for patients with established metastatic disease. Nevertheless, some
`men may accept the trade-off of a potentially inferior cancer outcome
`far an improved quality of life.
`Another question is whether hormones should be given early, in
`the adjuvant setting or at the time recurrence is first documented, or
`late, when metastatic disease or symptoms are manifest. Trials in sup-
`port of early therapy have often been underpowered relative to the “net
`lxnefit” reported or have been criticized on methodologic grounds. In
`one, although a survival benefit was shown for patients treated with
`radiation therapy and 3 years of androgen ablation relative to radiation
`alone, the trial was criticized for the poor outcomes for the control
`group. Another showing a survival benefit for patients with positive
`fies randomized to medical or surgical castration compared to ob-
`servation (p = .02) was criticized because the confidence intervals
`sound the 5- and 8-year survival distributions overlapped between the
`rwo groups. A large randomized study comparing early to late hor-
`LKJIIC treatment (orchiectomy or GnRH analogue) in patients with 10-
`::-Jly advanced or asymptomatic metastatic disease showed that pa-
`:;ents treated early were less likely to progress from M0 to M1 disease,
`jevelop pain, and die of prostate cancer. This trial was criticized be-
`sause therapy was delayed “too long” in the late-treatment group.
`When patients treated by radical surgery, radiation therapy, or watchful
`waiting were randomly assigned to receive bicalutamide, 150 mg, or
`glacebo, hormone treatment produced a significant reduction in the
`._:roportion of patients who developed osseous metastases at 2 years
`for bicalutamide; 13.8% for placebo). This result has not gained
`acceptance in part because too many “good-risk” patients were treated
`md because no effect on survival was demonstrated. These criticisms
`
`are valid; however, the net influence on survival from early hormone
`intervention is similar to that observed in patients with breast cancer
`where adjuvant hormonal therapy is routinely given.
`Another way to reduce the side effects of androgen ablation is to
`ajmirrister hormones on an intermittent basis. This was proposed as a
`way to prevent the emergence of castration—resistant cells by “forcing”
`The cells that survive androgen ablation into a normal differentiation
`gxathway by repleting testosterone. Theoretically, surviving cells that
`are allowed to proliferate in the presence of androgen will retain sen-
`ritivity to androgen ablation. The duration of treatment varies from 2
`m 6 months beyond the point of maximal response. Once therapy is
`stopped, endogenous testosterone levels increase, and the symptoms
`associated with androgen ablation abate. PSA levels also begin to rise,
`at some level, androgen ablation is restarted. Using this approach,
`multiple cycles of regression and proliferation have been documented
`2:: individual patients. It is unknown whether the intermittent approach
`
`81 Htjperplastic and Malignant Diseases of the Prostate
`
`549
`
`increases, decreases, or does not change the overall duration of sensitivity
`to androgen ablation. A trial to address this question is ongoing.
`
`METASTATIE DISEASE: CASTRATE Castration-resistant disease can be man-
`
`ifest in many ways. For some it is a rise in PSA with no change in
`radiographs and no new symptoms. In others, it is a rising PSA and
`progression in bone, with or without symptoms of disease. Still others
`will show soft tissue disease with or without osseous metastases, and
`others have a pattern of visceral spread. The prognosis, highly variable,
`can also be predicted using nomograms designed for this cohort. The
`important distinction is that despite the failure of first-line hormone
`treatment, the majority of these tumors remain sensitive to second-
`and third-line honnonal treatments. Castration resistance does not in-
`
`dicate hormonal resistance. The rising PSA is an indication of contin-
`ued signaling through the androgen receptor axis.
`The manifestations of disease in this patient group hinder the de-
`velopment of drugs and treatment standards because the traditional
`measures of outcome such as tumor regression do not apply. No PSA-
`based outcomes are true surrogates for a survival benefit, and assessing
`changes in osseous disease using bone scans is notoriously inaccurate.
`It is essential to define therapeutic objectives before initiating treat-
`ment, as standards of care have changed on the basis of randomized
`comparisons that provide clinical benefits without prolonging life.
`These endpoints include the relief of symptoms and delaying metas-
`tases or the time to the development of new symptoms of disease.
`The management of these patients requires first that the castrate
`status be documented. Patients receiving an antiandrogen alone who
`have elevated levels of serum testosterone should be treated first with
`
`a GnRH analogue or orchiectomy and observed for response. Patients
`on an anti-androgen in combination with a GnRH analogue should
`have the antiandrogen discontinued, as ~30% will respond to the with-
`drawal of the antiandrogen. Any response occurs within weeks of stop-
`ping flutamide, but may take 8 to 12 weeks with nilutamide and bi-
`calutamide (they have a long terminal half~life). At
`the time of
`progression, a different antiandrogen can be given as these agents are
`not cross-resistant. Other honnones that may be active include estro-
`gens, progestins, ketoconazole, and glucocorticoids. Those who re-
`spond to estrogens or progestins should also be evaluated for a with-
`drawal response at
`the time of progression. Cytotoxic agents are
`considered when hormone responses stop.
`No chemotherapy regimen has been proven to prolong life in these
`patients. However, responses to chemotherapy that improve symptom
`control are not uncommon. Drugs directed at the tumor cell cytoskel-
`eton such estramustine (Emcyt) and a taxane such as paclitaxel or
`docetaxel can induce responses in 250% using measurable disease
`regression as the endpoint. Seventy percent will show a >50% decline
`in PSA from baseline. Studies evaluating survival effects are nearly
`done.
`
`Management of pain is a critical part of therapy. Optimal palliation
`requires assessing whether the symptoms and metastases are focal or
`diffuse and whether disease threatens the spinal cord, the cauda equina,
`or the base of the skull. Neurologic symptoms require emergent eval-
`uation because loss of function may be permanent if not addressed in
`a timely manner. Single sites of pain or areas of neurologic involve-
`ment are best treated with external beam radiation. As the disease is
`
`often diffuse, palliation at one site often leads to the emergence of
`symptoms at another. An important principle of management was es-
`tablished in two randomized trials of mitoxantrone and prednisone vs.
`prednisone alone. In both studies, mitoxantrone-treated patients had a
`greater reduction in pain, used fewer narcotics, were more mobile, and
`had less fatigue. No survival benefit was shown.
`Given the bone-dominant nature of prostate cancer spread, bone-
`directed therapies may be usesful in patients with diffuse disease. Two
`bone-seeking radioisotopes, 39Sr
`(metastron) and ‘53Sm-EDTMP
`(quadramet), are approved for palliation of pain although they have
`no effect on PSA or on survival. Fewer patients treated with an isotope
`
`

`
`2142
`
`Part XIV Endocrinology and Metabolism
`
`some 21q22.3. The gene encodes a transcription factor thought to be
`involved in lymphocyte function. The type I syndrome usually pres-
`ents during childhood, whereas the type H syndrome is usually man-
`ifested in adulthood.
`
`Clinical suspicion of adrenal insufficiency should be high in pa-
`tients with AIDS (Chap. 173). CMV regularly involves the adrenal
`glands (so-called CMV necrotizing adrenalitis), and involvement with
`Mycobacterium avium—intracellulare, Cryptococcus, and Kaposi’s sar-
`coma has been reported. Adrenal insufficiency in AIDS patients may
`not be manifest, but tests of adrenal reserve frequently give abnormal
`results. When interpreting tests of adrenocortical function, it is im-
`portant to remember that medications such as rifampin, phenytoin,
`ketoconazole, megestrol, and opiates may cause or potentiate adrenal
`insufficiency. Adrenal hemorrhage and infarction occur in patients on
`anticoagulants and in those with circulating anticoagulants and hyper-
`coagulable states, such as the antiphospholipid syndrome.
`There are several rare genetic causes of adrenal insufficiency that
`present primarily in infancy and childhood (see below).
`
`insufficiency caused by
`Clinical Signs and Symptoms Adrenocortical
`gradual adrenal destruction is characterized by an insidious onset of
`fatigability, weakness, anorexia, nausea and vomiting, weight loss,
`cutaneous and mucosal pigmentation, hypotension, and occasionally
`hypoglycemia (Table 321-7). Depending on the duration and degree
`of adrenal hypofunction, the manifestations vary from mild chronic
`fatigue to fulminating shock associated with acute destruction of the
`glands, as described by Waterhouse and Friderichsen.
`Asthenia is the cardinal symptom. Early it may be sporadic, usually
`most evident at times of stress; as adrenal function becomes more im-
`paired, the patient is continuously fatigued, and bed rest is necessary.
`Hyperpigmentation may be striking or absent. It commonly appears
`as a diffuse brown, tan, or bronze darkening of parts such as the elbows
`or creases of the hand and of areas that normally are pigmented such
`as the areolae about the nipples. Bluish-black patches may appear on
`the mucous membranes. Some patients develop dark freckles, and ir-
`regular areas of vitiligo may paradoxically be present. As an early sign,
`tanning following sun exposure may be persistent.
`Arterial hypotension with postural accentuation is frequent, and
`blood pressure may be in the range of 80/50 or less.
`Abnormalities of gastrointestinal function are often the presenting
`complaint. Symptoms vary from mild anorexia with weight loss to
`fulminating nausea, vomiting, diarrhea, and ill—defined abdominal
`pain, which may be so severe as to be confused with an acute abdomen.
`Patients may have personality changes, usually consisting of excessive
`irritability and restlessness. Enhancement of the sensory modalities of
`taste, olfaction, and hearing is reversible with therapy. Axillary and
`pubic hair may be decreased in women due to loss of adrenal andro-
`gens.
`
`In the early phase of gradual adrenal destruction,
`Laboratory Findings
`there may be no demonstrable abnormalities in the routine laboratory
`
`
`
`
`TABLE 321-7 Frequency of Symptoms and Signs in Adrenal Insufficiency
`Sign or Symp