Detection of mammary serum antigen in sera from breast cancer patients using monoclonal antibody 3E1.2.

Murine monoclonal antibody 3E1.2, made against human breast cancer cells, detects a glycoprotein (Mr greater than 300,000) called mammary serum antigen (MSA) which is elevated in the serum of patients with breast cancer. An enzyme immunoassay was developed to detect MSA in human serum and used to detect MSA in subjects with breast cancer and other diseases. Raised levels of MSA (greater than 300 inhibition units) were found in the serum of 1.9% of 2406 blood donors, in 18% of sera from 40 subjects with benign breast disease, and in 16% of sera from 222 subjects with non-breast cancers. However, in patients with a diagnosis of breast cancer, 76% (84 of 110) of Stage I and II, and 86% (142 of 166) of Stage III and IV had levels of greater than 300 inhibition units. Nineteen % of patients, classified clinically disease free, had raised MSA levels. In 34 of 37 (92%) patients followed over 2 to 11 mo the level of MSA correlated with the clinical course of disease. Changes in MSA levels not only corresponded to changes in the clinical course of disease, but also preceded the clinical detection of progressive disease. Immunoblotting has detected a heterogeneous molecule of Mr greater than 300,000 and been used to confirm the elevation of MSA in breast cancer patients. Determination of MSA level may be useful for the detection of breast cancer and for monitoring progress of disease and response to therapy.

Induction of transforming growth factor beta 1 in human breast cancer in vivo following tamoxifen treatment.

We have investigated the ability of tamoxifen to regulate members of the transforming growth factor beta (TGF-beta) family in human breast cancers in vivo. Using immunohistochemical techniques, we find that 3 months of tamoxifen treatment causes a consistent induction of extracellular TGF-beta 1 in breast cancer biopsies, compared with matched pretreatment samples from the same patient. The induced TGF-beta is localized between and around stromal fibroblasts and appears to be derived from these cells. Lower levels of TGF-beta 1,-beta 2, and -beta 3 seen in epithelial cells were not altered by tamoxifen treatment. The increased stromal staining of TGF-beta 1 occurred in estrogen receptor-negative as well as estrogen receptor-positive tumors. These results provide in vivo evidence for a novel, estrogen receptor-independent mechanism of action for tamoxifen, involving the stromal induction of a potent growth inhibitor for epithelial cells.

HER-2 amplification impedes the antiproliferative effects of hormone therapy in estrogen receptor-positive primary breast cancer.

In experimental models, human epidermal growth factor receptor-2 (HER-2) amplification leads to estrogen independence and tamoxifen resistance in estrogen receptor (ER)-positive human breast cancer cells. Some but not all reports suggest an association between HER-2 positivity and hormone independence in breast cancer patients. This study aimed to evaluate the antiproliferative effects of endocrine therapy in HER-2-positive/ER-positive primary human breast cancer. The effect on proliferation (Ki67) of hormone therapy was assessed at 2 weeks and/or 12 weeks in biopsies from 115 primary breast cancers with ER-positive tumors. The patients took part in one of 3 neoadjuvant trials of hormonal therapy with a SERM (tamoxifen or idoxifene) or an aromatase inhibitor (anastrozole or vorozole). HER-2 status was assessed by immunocytochemistry and fluorescence in situ hybridization (FISH). Fifteen patients were defined as HER-2 positive by both immunohistochemistry and FISH, with the remaining 100 patients HER-2 negative. Geometric mean Ki67 levels were substantially higher in HER-2-positive than HER-2-negative tumors (27.7% versus 11.5%, respectively; P = 0.003). In HER-2-negative patients, Ki67 was reduced by 62 and 71% at 2 and 12 weeks, respectively (P < 0.0001 for both), but HER-2-positive patients showed no significant fall. The proportional change in Ki67 was significantly different between HER-2-positive and -negative patients (P = 0.014 at 2 weeks; P = 0.047 at 12 weeks). Mean ER levels were lower in the HER-2-positive patients (P = 0.06) but the change in Ki67 was impeded even in those with high ER. Apoptotic index was reduced by 30% at 2 weeks in the HER-2-negative group. However, there were no statistically significant differences in apoptotic index between the groups. It is concluded that ER-positive/HER-2-positive primary breast carcinomas show an impeded antiproliferative response to endocrine therapy that nonetheless may vary between individual treatments. This together with high baseline proliferation is likely to translate to poor clinical response.

Radioimmunotherapy of breast cancer xenografts with monoclonal antibody ICR12 against c-erbB2 p185: comparison of iodogen and N-succinimidyl 4-methyl-3-(tri-n-butylstannyl)benzoate radioiodination methods.

C-erbB2 p185 is a proto-oncogene product expressed in 25-30% of human invasive breast cancers that is associated with poor prognosis and resistance to endocrine therapy and chemotherapy. It is minimally expressed in normal adult tissues (M. F. Press et al., Oncogene, 5: 953-962, 1990). For this reason, it is an attractive target for radioimmunotherapy and other antibody-directed therapies. ICR12 is a rat IgG2a monoclonal antibody directed against a protein epitope of the external domain of the c-erbB2 p185. We performed experiments to optimize the direct iodination of ICR12 with 131I using the IodoGen method, and we found impairment of immunoreactive fraction with increasing specific activity. N-Succinimidyl 4-methyl-3-(tri-n-butylstannyl)benzoate (MATE) is a tin ester that can be radioiodinated easily and then coupled to the epsilon-amino group of lysine residues. This method has been shown to have improved uptake in tumors compared with antibody labeled by direct iodination (P. K. Garg et al., Nucl. Med. Biol., 20: 379-387, 1993). ICR12 could be labeled up to 16 mCi/mg by this technique without loss of immunoreactive fraction. Whole-body retention of MATE-labeled ICR12 was less than IodoGen (P < 0.0001). Radioimmunotherapy experiments in athymic mice bearing established MDA MB 361 human breast cancer xenografts showed growth inhibition for > 24 days at a dose of 600 microCi/mouse (P < 0.0001) when labeled by the IodoGen technique, and 12 days using the MATE method (P < 0.0001).

Clinical prognostic and predictive factors for primary chemotherapy in operable breast cancer.

PURPOSE: This study aimed to identify clinical factors that are of prognostic significance or that predict for subsequent treatment outcome in patients with large operable breast cancer treated with primary chemotherapy (PCT) at our institution. METHODS: One hundred eighty-five patients received the following regimens: CMF or MMM (76 patients), ECF (75 patients), AC or FEC (34 patients), followed by surgery, with radiotherapy (RT) given to those with breast conservation. A number of common clinical variables were assessed in relation to local recurrence-free survival (LRFS), disease-free survival (DFS), and overall survival (OS). RESULTS: Clinical responders had improved DFS (P = .009) and OS (P = .08) compared with nonresponders. There was no association between clinical or pathologic complete remission (CR) and survival. Pretreatment clinical axillary node positivity was a significant predictor of worsened DFS (P = .0001), OS (P = .0001), and LRFS (P = .03). Patients remaining clinically node-positive postchemotherapy had an inferior outcome compared with those becoming node-negative (DFS, P = .03; OS, P = .03) but pathologic axillary node status was not shown to predict for survival. Twenty-nine patients in clinical CR following PCT who electively did not have surgery and were treated with RT alone had significantly increased local recurrence rate compared with partial responders having surgery and RT (P = .02). There were no differences in DFS or OS between these groups. On multivariate analysis, clinical axillary node status was the only independent predictor of OS and DFS, and LRFS. CONCLUSION: Pretreatment and posttreatment clinical axillary node status is a major predictor of outcome following PCT. Complete clinical response does not define a more favorable subgroup compared with those not obtaining CR.

Is surgery necessary after complete clinical remission following neoadjuvant chemotherapy for early breast cancer?

PURPOSE: This retrospective analysis aimed to identify whether breast cancer patients receiving radiotherapy alone following a complete clinical remission (cCR) to neoadjuvant chemotherapy had a worse outcome than those treated with surgery. PATIENTS AND METHODS: One hundred thirty-six patients who had achieved a cCR to neoadjuvant chemotherapy for early breast cancer were identified from a prospectively maintained database of 453 patients. Of these, 67 patients had undergone surgery as their primary locoregional therapy, and 69 patients had radiotherapy alone. Outcome was assessed in relation to local recurrence-free survival, disease-free survival, and overall survival. RESULTS: Median follow-up was 63 months in the surgery group and 87 months in the no surgery group. Prognostic characteristics were well balanced between the two groups. For surgery and no surgery, respectively, there were no significant differences in disease-free survival or overall survival (5-year, 74% v 76%; 10-year, 60% v 70%, P =.9) between the two groups. There was a nonsignificant trend toward increased locoregional-only recurrence for the no surgery group (21% v 10% at 5 years; P =.09), but no long-term failures of local control. Patients in the no surgery group who also achieved an ultrasound complete remission had a 5-year local recurrence rate of only 8%. CONCLUSION: In patients achieving a cCR to neoadjuvant chemotherapy, radiotherapy alone achieve survival rates as good as with surgery, but with higher local recurrence rates. Ultrasound may identify a low recurrence rate subgroup for assessing no surgery in a prospective trial.

High complete remission rates with primary neoadjuvant infusional chemotherapy for large early breast cancer.

PURPOSE: To investigate the efficacy of continuous infusion fluorouracil (5FU) with every-3-week epirubicin and cisplatin (ECF) as primary chemotherapy instead of immediate mastectomy for patients with large, potentially operable, breast cancer. PATIENTS AND METHODS: Fifty patients with large operable breast cancer, median tumor diameter 6 cm (range, 3 to 12), were treated with 5FU 200 mg/m2/d via a Hickman line using an ambulatory pump for 6 months with epirubicin 50 mg/m2 intravenously (IV) and cisplatin 60 mg/m2 IV every 3 weeks for eight courses. Subsequent surgery and/or radiotherapy was determined by clinical response. RESULTS: Forty-nine patients achieved an overall response (98%; 95% confidence interval [CI], 94% to 100%), including 33 complete clinical remissions (CRs) (66%; 95% CI, 53% to 79%). Only three patients (6%) still required mastectomy. Tumor cellularity was markedly reduced on repeat needle biopsy following 3 weeks of treatment in 81% of patients versus only 36% in similar patients after conventional chemotherapy (P < .002). Severe (World Health Organization [WHO] grade 3 to 4) toxicity was rare, with nausea/vomiting being the most common, occurring in 20% of patients. CONCLUSION: Primary infusional ECF appears to be more active on clinical and histopathologic grounds than conventional chemotherapy for large operable breast cancer and is well tolerated. This approach now merits randomized comparison to determine if high CR rates may translate into improved survival.

Randomized trial of chemoendocrine therapy started before or after surgery for treatment of primary breast cancer.

PURPOSE: To evaluate in a randomized clinical trial systemic chemoendocrine therapy used as primary (neo-adjuvant) treatment before surgery in women with primary operable breast cancer. PATIENTS AND METHODS: Patients aged less than 70 years with clinically palpable, primary operable breast cancer diagnostically confirmed by fine-needle aspiration cytology (FNAC) and suitable for treatment with surgery, radiotherapy, cytotoxic chemotherapy, and tamoxifen were considered eligible. Patients randomized to neoadjuvant treatment received four cycles of chemo-therapy for 3 months before surgery followed by another four cycles after surgery, and were compared with patients randomized to adjuvant therapy who received eight cycles of chemotherapy over 6 months after surgery. RESULTS: Of 212 patients who were randomized to receive either adjuvant (n = 107) or neoadjuvant (n = 105) chemoendocrine therapy, 200 are now assessable for response. The two groups are comparable for age, menopausal status, disease stage, and surgical requirements. The overall clinical response rate was 85%, with a complete histologic response rate of 10%. There was a significant reduction in the requirement for mastectomy in patients who received neoadjuvant treatment (13%) as compared with those who received adjuvant therapy (28%) (P < .005). Symptomatic and hematologic acute toxicity was low and similar for adjuvant and neoadjuvant therapy. The median follow-up period for patients in this trial is 28 months, during which time four patients have relapsed locally and 20, including one of the local relapses, have developed metastatic disease, 19 of whom have died. The follow-up period is too brief to evaluate relapse rate or survival duration. CONCLUSION: This trial confirms previous reports of a high rate of response to neoadjuvant therapy, but is the first to include small primary cancers and to show, in the context of a randomized trial, a reduction in the requirement for mastectomy. Until disease-free and overall survival data are available from the larger National Surgical Adjuvant Breast and Bowel Project (NSABP)-18 trial, such neoadjuvant treatment cannot be recommended outside of a clinical trial.

Hormone replacement therapy in women with previous breast cancer.

As breast cancer is known to be a tumour sensitive to the effects of endogenous oestrogens, clinicians are reluctant to prescribe hormone replacement therapy (HRT) to women with a history of previous breast cancer for fear of stimulating disease recurrence, and it is currently contraindicated in this group of women. However, an increasing proportion of breast cancer patients are requesting the use of HRT to relieve the symptoms of oestrogen deficiency, which are also a common side-effect of adjuvant therapy for breast cancer. Observational data on the use of HRT in breast cancer survivors has not demonstrated an increase in disease recurrence, but uncertainty will continue in the absence of data from prospective, randomized trials. This review aims to demonstrate why it is ethical and scientifically important to undertake such studies.

Cathepsin D levels in primary breast cancers: relationship with epidermal growth factor receptor, oestrogen receptor and axillary nodal status.

Cathepsin D and the epidermal growth factor receptor (EGFr) have both been proposed as poor prognostic markers in breast cancer. We have compared the tumour cytosolic cathepsin D level with EGFr and oestrogen receptor (ER) levels and the axillary node status of 131 patients with operable breast cancer, to see if EGFr and cathepsin D are co-regulated. Cathepsin D level was measured using a two-site immunoradiometric assay kit. No correlation was found between the level of cathepsin D and EGFr, ER or nodal status. Since the raised level of cytosolic cathepsin D was not related to EGFr, it may be that measuring the level of both of these markers in the same sample will give additional prognostic information.

Modulation of Bcl-2 and Ki-67 expression in oestrogen receptor-positive human breast cancer by tamoxifen.

The expression of the bcl-2 proto-oncogene, which is associated with prolonged cell survival and prevention of programmed cell death, was investigated in human primary breast carcinomas prior to and following endocrine therapy with the anti-oestrogen, tamoxifen. Using the BCL-2-100 antibody, a 26-kD protein was detected by western immunoblot in the cytosols of oestrogen receptor (ER)+ve human breast cancers. In a cross-sectional study, the immunohistochemical expression of Bcl-2 was observed in 32% of invasive breast cancers, but in 65% of tumours treated with tamoxifen (P = 0.009). There was a significant association of Bcl-2 with ER status, with 64% of untreated and 88% of tamoxifen-treated Bcl-2-positive tumours being ER+ve. A significantly lower Ki-67 score was found in tamoxifen-treated tumours which were Bcl-2-positive compared with Bcl-2-negative (9.3 versus 24.6%, P = 0.01). In a separate series of sequential Trucut biopsies from 18 patients, the frequency of Bcl-2 expression was increased in ER+ve tumours from 3/12 to 8/11 following tamoxifen (P = 0.04). This was also associated with a significant reduction in mean Ki-67 score from 32 to 12% (P = 0.0004). The observations from this study clearly indicate that Bcl-2 in human breast cancer is associated with ER status, and that expression is enhanced in ER+ve tumours following tamoxifen, in association with reduced cell proliferation.

Primary medical (neo-adjuvant) chemotherapy for operable breast cancer.

84 patients with large operable breast cancer have been treated with primary medical chemotherapy rather than mastectomy in three sequential studies. 86% had tumours greater than 4 cm in diameter; median diameter was 6 cm (range 1-12). Median age was 46 years (range 23-66). In the first two studies 64 patients were treated with either CMF [cyclophosphamide 100 mg orally days 1-14, methotrexate 50 mg intravenously (i.v.) days 1 and 8, and 5-fluorouracil 1 g i.v. days 1 and 8, repeating at 28-day intervals for six courses] or MMM (mitozantrone 8 mg/m2 i.v. once every 3 weeks, methotrexate 50 mg i.v. once every 3 weeks, mitomycin C 8 mg/m2 once every 6 weeks, for 8 courses). 69% achieved an overall response including 17% complete remissions. 27% have had local relapse but only 3% uncontrolled local relapse. Only 14% have required mastectomy. In the third study which is ongoing, 19 patients have been treated with infusional FEC (5-fluorouracil 200 mg/m2 i.v. 24 hourly by continuous infusion via a Hickman line for 6 months, epirubicin 50 mg/m2 i.v. bolus once every 3 weeks for 6 months, cisplatin 60 mg/m2 i.v. once every 3 weeks for 6 months with appropriate intravenous hydration). Overall response rate so far is 84% with 58% complete remissions. There have been no local relapses and no patient has required mastectomy. This study demonstrates that primary medical chemotherapy can be used to avoid mastectomy in the great majority of patients presenting with large operable primary breast cancer. Infusional FEC may be more active than conventional chemotherapy in terms of overall response and complete remission rate, and infusional FEC chemotherapy now needs to be compared with conventional chemotherapy. The concept of primary medical therapy should also be compared with conventional mastectomy followed by adjuvant chemotherapy.

The detection of micrometastases in the peripheral blood and bone marrow of patients with breast cancer using immunohistochemistry and reverse transcriptase polymerase chain reaction for keratin 19.

The aim of this study was to determine whether reverse transcriptase polymerase chain reaction (RT-PCR) for keratin 19 (K19) provides additional information when combined with immunohistochemistry when used to detect micrometastases in blood and bone marrow in patients with primary breast cancer. We studied 78 patients with breast cancer who had no evidence of distant metastases. We collected blood and bone marrow, separated the mononuclear fraction and carried out RT-PCR and immunohistochemistry for K19. RT-PCR was done by two 40-cycle rounds using nested primers. In initial experiments, RT-PCR was shown to be capable of detecting one tumour cell in one million normal bone marrow cells, which was at least 10 times more sensitive than immunohistochemistry, while retaining specificity. Five per cent of the peripheral blood and 22% of the bone marrow samples contained K19 positive cells by immunohistochemistry staining. Using RT-PCR, these proportions increased to 25% and 35%, respectively. This represents a significantly greater detection frequency (P < 0.001 and P = 0.03, respectively). RT-PCR for K19 is a more sensitive method for detecting micrometastases in patients with primary breast cancer when compared with immunohistochemistry.

The prevalence of personality disorders in 210 women with eating disorders.

BACKGROUND: The purpose of this study was to assess the prevalence, reliability, and predictive value of comorbid personality disorders in a large sample of 210 women seeking treatment for anorexia nervosa (N = 31), bulimia nervosa (N = 91), or mixed disorder (N = 88). METHOD: All subjects were interviewed using the Structured Interview for DSM-III Personality Disorders as part of a longitudinal outcome study of eating disorders currently underway at Massachusetts General Hospital. RESULTS: Of the 210 subjects, 27% had at least one personality disorder; the most commonly observed was borderline personality disorder in 18 subjects (9%). The highest prevalence of personality disorders was found in the anorexia nervosa/bulimia nervosa group at 39%, followed by 22% in the anorexics and 21% in the bulimic sample. We found statistically significant differences regarding the distribution of personality disorders across eating disorder groups. The dramatic personality disorder cluster was differentially distributed across groups; this finding was accounted for by higher rates of borderline personality disorder in the bulimia nervosa and anorexia nervosa/bulimia nervosa groups than in the anorexia nervosa group. The anxious personality disorder cluster was differentially distributed across groups with higher rates in the anorexia nervosa and anorexia nervosa/bulimia nervosa samples. Those subjects with a comorbid personality disorder had a significantly slower recovery rate than those without a comorbid personality disorder. CONCLUSION: The prevalence of personality disorders is not high in treatment-seeking women with eating disorders compared with previously studied samples. The greatest frequency of comorbid personality disorders is in the anorexia nervosa/bulimia nervosa group; this subset also had longer duration of eating disorder illness and much greater comorbid Axis I psychopathology compared with the rest of the sample. Future studies should address whether personality disorders have predictive value in the long-term course and outcome of eating disorders.

The course and outcome of bulimia nervosa.

Although knowledge of the long-term course and outcome of bulimia nervosa is critical to the effective management of the bulimic patient, research in this area is still in its infancy. Comparisons between existing outcome studies are complicated by differences in research design and methods, a lingering confusion over the definition of terms such as recovery and relapse, and the possibility of multiple forms of bulimia nervosa with varying outcomes. Nonetheless, these studies agree that bulimia nervosa is a chronic disorder, characterized by high rates of relapse and persistent subclinical symptomatology. Future investigations into the course and outcome of bulimia nervosa should focus on the derivation of a universally acceptable terminology and the identification of clinical and psychosocial variables predicting recovery.

Bone marrow transplantation for myelodysplastic syndrome--who? when? and which?

Although allogeneic transplantation has resulted in long-term disease-free survival in some patients with myelodysplastic syndromes (MDS), the morbidity and mortality of this approach remains high. Additionally, many patients are not candidates for such an approach because of their age or comorbid factors. Autologous transplantation and the use of reduced intensity conditioning prior to allogeneic stem cell transplantation has provided less toxic alternatives as well as increased the numbers of patients eligible for some form of transplantation. While bone marrow transplantation clearly has a role in the treatment of MDS, the decision to proceed to transplantation is not always easy and the optimal approach has not been clearly defined. Improvement in patient selection and novel approaches to transplantation will hopefully allow for more effective, less toxic results.

Determining factors which predict response to primary medical therapy in breast cancer using a single fine needle aspirate with immunocytochemical staining and flow cytometry.

The increasing use of neoadjuvant chemotherapy and endocrine therapy in the management of breast cancer has lead us to evaluate and optimise the standard technique of cytocentrifugation of a single fine needle aspirate (FNA) taken from a breast tumour in-vivo, to determine a range of both immunocytochemical and flow cytometric factors which are predictive of response to primary medical therapy. Some of these factors are also of prognostic significance in early stage disease. An analysis of the cellularity and immunocytochemical staining characteristics of FNAs obtained from a series of 206 patients with palpable breast cancers indicate that in a sample of 46 cases it is possible to measure oestrogen receptor, progesterone receptor and c-erbB-2 providing over 400 cells per slide are obtained, with material obtained in a single FNA prepared by cytocentrifugation, using standard immunocytochemical methods. The staining results obtained were comparable to those obtained using frozen or paraffin embedded tissue sections taken from the same tumour. In addition an estimate of the proliferation indices could be made by flow cytometric analysis of the residual cell suspension fluid with measurement of DNA index and S-phase fraction in 131/164 (80%) and 110/164 (67%) of cases respectively. Providing all FNAs obtained for cytocentrifugation were taken at first presentation rather than immediately following a standard FNA, then it was possible to obtain adequately cellular (> 400 cells/slide) samples in 96 out of 126 (75%) of the last cohort of breast aspirates. These effects may be independent of T stage but not histological type as patients with lobular tumours only produced cellular aspirates in 1/7 (14%) of cases.(ABSTRACT TRUNCATED AT 250 WORDS)

The impact of parental affective disorder on depression in offspring: a longitudinal follow-up in a nonreferred sample.

OBJECTIVE: This study explored the effects of parental affective disorder on offspring in a nonreferred health maintenance organization 4 years after initial examination. METHOD: The sample, average age 18.5 years, included 91% of the 153 youngsters initially studied. The main instruments were structured diagnostic interviews scored according to criterion systems for both parents and children; assessment of the youngsters was blind to the previous assessment. RESULTS: Rates of major depressive disorder were higher in the children of parents with affective disorder (26%) compared with those whose parents had no disorder (10%). CONCLUSION: Depression and other parental affective disorders, as they occur in the community in parents who often are neither recognized nor treated, are associated with serious affective disorder in offspring. Clinical and preventive approaches for these offspring are needed and should be targeted to all families in which there is serious parental affective disorder, not just those who present for psychiatric treatment.

Psychiatric comorbidity in treatment-seeking anorexics and bulimics.

Current and lifetime psychiatric diagnoses were compared in 229 female patients seeking treatment for current episodes of anorexia nervosa (N = 41), bulimia nervosa (N = 98) and mixed anorexia nervosa and Schizophrenia-Lifetime Version, which was modified to include a section for DSM-III-R eating disorders, the Longitudinal Interval Follow-up Evaluation, and the Structured Interview for DSM-III Personality Disorders. Seventy-three percent of the anorexia nervosa subjects, 60% of the bulimia nervosa subjects, and 82% of the mixed anorexia nervosa and bulimia nervosa subjects had a current comorbid Axis I diagnosis. Major depression was the most commonly diagnosed comorbid disorder. Low rates of alcohol and substances abuse disorder were diagnosed, and personality disorder occurred in a minority of the sample. The subjects with mixed disorder manifested a higher lifetime prevalence of kleptomania than either the anorexics or the bulimics. High levels of comorbidity were noted across the eating disorder samples. Mixed disorder subjects manifested the most comorbid psychopathology and especially warrant further study.

Patterns and predictors of recovery in anorexia nervosa and bulimia nervosa.

OBJECTIVE: The purpose of this study was to assess the course and outcome of anorexia nervosa and bulimia nervosa at 1 year in a large cohort of women with eating disorders. METHOD: A prospective, naturalistic, longitudinal design was used to map the course of 225 women with anorexia nervosa, bulimia nervosa, and mixed anorexia and bulimia nervosa. Structured interviews were conducted quarterly. Follow-up data are presented in terms of patterns of recovery, clinical features predictive of time to recovery, and the role of comorbid disorders as fixed predictors. RESULTS: The recovery rate of bulimics was significantly better than that of anorexic or mixed subjects, yet nearly half the anorexic and mixed subjects no longer met full DSM-III-R criteria for at least 8 consecutive weeks during the first year of follow-up. Percent ideal body weight and type of eating disorder were significantly associated with outcome. CONCLUSIONS: Our findings suggest that the diagnosis of anorexia nervosa has severe implications.