Abstract P1-07-26: Clinicopathological predictors of axillary lymph node metastasis in early breast cancer patients

Axillary lymph node (ALN) status is key in the prognosis of early breast cancer. Sentinel lymph node biopsy (SLNB) is the standard treatment in determining ALN status in clinically node negative patients with early breast cancer, but nearly 70% exhibit no ALN metastasis (a regional lymph node (RLN) metastasis). Likewise, selective use of SLNB has engendered modeling ALN metastasis. Using the National Cancer Database (NCDB), this study aimed to evaluate clinicopathological factors to help predict first and subsequent RLN metastasis in patients with early breast cancer.

Methods: We identified 660,258 women from 2004-2013 with early breast cancer stage (cT1-T3, cN0-N1, and M0). A two-part model was estimated using multivariable logistic regression to evaluate 1) predictors, by odds (OR), of having at least one RLN metastasis, and 2) predictors, by rate difference (RD), of having additional RLN metastasis relative to women with only one RLN metastasis. The same set of predictor variables were included in both parts of the model. All analyses were conducted using SAS v. 9.4 with p <.05 indicating statistical significance.

Results: Adjusted ORs and RDs of RLN metastasis for selected variables from the model are shown in Table 1. Increased likelihood of at least one RLN metastasis was significantly associated with the presence of larger tumor size (p <0.0001), a primary tumor in the nipple region (p <0.0001) of the left breast (p <0.0001), lobular or ductal type histology (p <0.001, and p <0.0001, respectively), positive estrogen (ER) and progesterone (PR) receptor statuses (p <0.0001 for both), younger age (p <0.0001), being white (p <0.0001), and greater comorbidity (Charlson/Deyo - CD) (p < 0.0001). Predictors of at least one RLN metastasis were also significantly associated with higher adjusted rates of further metastasis, except age, tumor size, ductal type histology, ER, and PR. However, a primary tumor in the central region (p =0.037), not the nipple region, was most associated with additional metastasis.

Table 1. Adjusted OR/RD (95% CI) for RLN Metastasis Events≥1 event >1 event0.99 (0.99-0.99)Age (continuous)1.00 (-)1.10 (1.05-1.14)White vs. Other1.02 (1.00-1.04)0.98 (0.95-1.00)CD = 0 vs. 10.98 (0.97-0.99)0.89 (0.85-0.93)CD = 0 vs. ≥20.97 (0.95-0.99)1.02 (1.02-1.02)Tumor Size (continuous)1.00 (-) Primary Tumor Site - Nipple vs. (left), Central vs. (right) 1.80 (1.61-2.02)Overlap1.02 (1.01-1.04)1.87 (1.59-2.20)Tail0.98 (-)1.64 (1.46-1.84)LOQ1.00 (-)1.71 (1.53-1.92)UOQ1.02 (1.00-1.03)2.07 (1.84-2.33)LIQ1.07 (1.04-1.09)2.73 (2.43 -3.07)UIQ1.11 (1.09-1.13)1.44 (1.28 -1.63)Central (left), Nipple (right)1.06 (1.02-1.11) Histological Type - Lobular vs. 4.00 (3.72-4.30)Other1.52 (1.46-1.57)0.99 (-)Ductal1.18 (1.17-1.19)1.04 (1.04-1.04)ER (Positive vs. Negative)1.00 (-)1.02 (1.02-1.02)PR (Positive vs. Negative)1.00 (-)1.06 (1.05-1.08)Laterality (Left vs. Right)1.02 (1.01-1.02)≥1 event (vs. 0): OR; >1 event (vs. 1): RD; (-) = insignificant

Conclusion: Utilizing a large dataset, several clinicopathological factors emerged from the NCDB as independent predictors of at least one, or additional RLN metastasis, supporting their weighted inclusion in prediction tools for ALN metastasis. Notably, different primary tumor sites of the breast predicted the two events modeled.

Citation Format: Satish M, Walters R, Silberstein PT. Clinicopathological predictors of axillary lymph node metastasis in early breast cancer patients [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-07-26.

Controlled-release oxycodone compared with controlled-release morphine in the treatment of cancer pain: A randomized, double-blind, parallel-group study

Controlled-release oral formulations of oxycodone and morphine are both suitable analgesics for moderate to severe pain. They were compared in cancer-pain patients randomized to double-blind treatment with controlled-release oxycodone (n = 48) or controlled-release morphine (n = 52) every 12 h for up to 12 days. Stable analgesia was achieved by 83% of controlled-release oxycodone and 81% of controlled-release morphine patients in 2 days (median). Following titration to stable analgesia, pain intensity (0=none to 3=severe) decreased from baseline within each group (p </= 0.005), from 1.9 (0.1) to 1.3 (0.1), mean (SE), with controlled-release oxycodone, and from 1.6 (0.1) to 1.0 (0.1) with controlled-release morphine (no significant between-group differences). Typical opioid adverse experiences were reported in both groups. Hallucinations were reported only with controlled-release morphine (n = 2). Visual analog scores (VAS) for 'itchy' and 'scratchin' were lower with controlled-release oxycodone (p </= 0.044), as was peak-to-trough fluctuation in steady-state plasma concentration (p = 0.004). The correlation between plasma concentration and dose was stronger (p = 0.026) for oxycodone (0.7) than morphine (0.3). The relationship between pain intensity (VAS) and plasma concentration was more positive for oxycodone (p = 0.046). There was a positive relationship between morphine-6-glucuronide concentrations and urea nitrogen and creatinine levels (p = 0.001). Controlled-release oxycodone was as effective as controlled-release morphine in relieving chronic cancer-related pain, and as easily titrated to the individual's need for pain control. While adverse experiences were similar, controlled-release oxycodone was associated with less itching and no hallucinations. Controlled-release oxycodone provides a rational alternative to controlled-release morphine for the management of moderate to severe cancer-related pain.

Denileukin diftitox in combination with rituximab for previously untreated follicular B-cell non-Hodgkin's lymphoma

Follicular lymphoma exhibits intratumoral infiltration by non-malignant T lymphocytes inluding CD4+CD25+ regulatory T (T_reg) cells. We combined denileukin diftitox with rituximab in previously untreated, advanced-stage follicular lymphoma patients anticipating that denileukin diftitox would deplete CD25+ T_reg cells while rituximab would deplete malignant B-cells. Patients received rituximab 375 mg/m2 weekly for 4 weeks and denileukin diftitox 18 mcg/kg/day for 5 days every 3 weeks for 4 cycles; neither agent was given as maintenance therapy. Between August 2008 and March 2010, 24 patients were enrolled. One patient died before treatment was given and was not included in the analysis. Eleven of 23 patients (48%; 95% CI: 27–69%) responded; 2 (9%) had complete responses and 9 (39%) had partial responses. The progression-free rate at 2 years was 55% (95%CI: 37–82%). Thirteen patients (57%) experienced grade ≥3 adverse events and 1 patient (4%) died. In correlative studies, soluble CD25 and the number of CD25+ T-cells decreased after treatment, however there was a compensatory increase in IL-15 and IP-10. We conclude that while the addition of denileukin diftitox to rituximab decreased the number of CD25+ T-cells, denileukin diftitox contributed to the toxicity of the combination without an improvement in response rate or time to progression.

Survival outcomes in metastatic renal carcinoma based on histological subtypes: SEER database analysis

381

Background: Renal cell cancer (RCC) represents a heterogeneous group of tumors with distinct histopathologic, genetic, and clinical features. This study was conducted to evaluate the prognostic value of histology in RCCs.

Methods: 3,062 patients with metastatic RCC from year 1998-2007 were identified from Surveillance Epidemiology and End Results (SEER) database. Data regarding their age, sex, race, treatment modality utilized and cancer specific survival was extracted and analyzed.

Results: Clear cell, sarcomatoid, adenocarcinoma NOS, papillary, collecting duct, chromophobe and cyst associated renal carcinoma accounted for 2,166 (70%), 433 (14%), 216 (7%), 160 (52%), 47 (1.5%), 35 (1.14%), and 5 (0.01%) respectively. The mean age of presentation was 63.5 in clear cell RCC and 62.7 years in non-clear cell RCC. Surgery was performed in 57.23% of all clear cell cancers and 50.84% in non-clear cell-RCC group. 27.34% of all clear cell cancer patients received radiation in contrast to 34.53% in the other group. Both clear cell RCC (64.54%) and non-clear cell RCC (69.52%) occurred commonly in males. Incidence of clear cell RCC was 4.67% in Asians, 7.01% in Blacks, 88.32% in Whites. The incidence of non-clear cell RCC was 3.81% in Asians, 14.78% in Blacks, and 81.41% in Whites. Cancer-specific survival was calculated in all histologic subtypes. Median survival for clear cell RCC was 8 months, 3 months for adenocarcinoma NOS, 7 months for cyst associated renal carcinoma, 8 months for papillary carcinoma, 7 months for chromophobe type, 4 months for sarcomatoid and 4 months for collecting duct RCC. Median overall survival was significantly better for clear cell cancer as compared to non-clear cell RCCs (8 months vs. 4 months; p< 0.0001).

Conclusions: Non-clear histology RCCs represent 30% of metastatic RCC and are associated with poor prognosis when compared to clear cell RCC.While development of novel targeted therapies has improved survival in clear cell RCC, it has not made a impact in survival of non-clear RCCs. An indepth understanding of the genetic and molecular changes associated with non-clear cell RCCs is important to improve their prognosis.

No significant financial relationships to disclose.

Racial disparities in stage IV prostate cancer outcomes in the Veterans Affairs hospital system

199

Background: Epidemiological studies have demonstrated that African-American (AA) men with prostate cancer have lower overall survival and cancer-specific survival rates than Caucasian (CA) men with prostate cancer. We aim to assess whether racial disparities exist for prostate cancer within an equal access health care system like the Veterans Affairs (VA) hospitals.

Methods: A retrospective analysis of AA and CA with metastatic prostate adenocarcinoma diagnosed between 1995 to 2007 via the Veterans Affairs Central Cancer Registry was conducted. Age, Race and type of treatment received were studied with respect to overall survival by using log-rank and Kaplan-Meier analysis.

Results: A total of 8,195 patients with advanced prostate cancer were analyzed, majority of them where CA (66.32%), 30.8% were AA and 2.8% belonged to other races. The median survival for AA was 2.71 years, 2.88 years in CA and 4.02 years in other races (P value <0.0001). Subgroup analysis based on treatment modality used showed the following median survival rates in years: No treatment (CA: 1.73, AA: 1.39, other races: 3.79, P<0.816), Hormonal therapy alone (CA: 2.51,AA: 2.45, Others:3.72, P<0.022), Radiation therapy alone (CA: 2.2, AA: 2.7, Others: 2.26, P<0.832), combined hormone and radiation therapy (CA: 2.71, AA: 2.31, Others: 6.64, P<0.029). However when CA and AA survival were compared excluding other races there was no statistically significant difference in survival irrespective of type of therapy received.

Conclusions: In advanced prostate cancer, AA and CA have uniformly poor prognosis. Type of therapy received did not influence the survival of both races. The numbers for other races is too small to make a definitive conclusion regarding their prognosis.

No significant financial relationships to disclose.

Primary signet ring cell carcinoma of colon: Retrospective analysis of VACCR database

58

Background: Signet ring cell carcinoma accounts for less than 1% of all colon cancers. We examined the clinical pathological features and prognosis of signet ring cell carcinoma of colon and compare it with mucinous and non-mucinous adenocarcinoma of colon.

Methods: A total of 206 patients diagnosed with signet ring cell carcinoma from 1995 to 2009 were identified from the VA Central Cancer Registry (VACCR) database. Age, race, histology, grade, lymph node status, stage and type of treatment received data were collected.

Results: Out of 206 patients, 173 (83.9%) were white, 31 (15%) were black, and 2 patients were listed as unknown. Median age of diagnosis was 67 years as compared to 70 years for both mucinous and non-mucinous adenocarcinoma of colon. Pathological T-stages were as follows: T1 = 2.9%, T2=5.3%, T3=33.9%, T4= 25.7%, and unknown 32%. Of the total, 22.3% were located in caecum, 21.8% in ascending colon, 15.5% in sigmoid colon, 7.7% in appendix and hepatic flexure of colon, 11.1% in transverse colon, 2.9% in splenic flexure and 4.4% in descending colon. 33.5% were lymph node positive, 34.6% were lymph node negative, and 31.8% were unknown. Histologically grade 3 (55.4%) was most commonly reported followed by grade 2 (7.3%), grade 1 (2.5%), grade 4 (1.9%)and in 33% grade was unknown. 41.3% patients received only surgery while 34% received surgery with adjuvant chemotherapy, 7.3% received chemotherapy alone and 7.8% patients received either chemotherapy, radiation or hormonal therapy alone, 9% did not receive any therapy. 1 year, 3 year and 5 year survivals for signet ring cell cancer compared to adeno carcinoma was 60% vs 80%, 33% vs 60%, and 24% vs 47% respectively. Median survival of signet ring cell carcinoma compared to mucinous and non mucinous adenocarcinoma was 19 months, 48 months and 62 months respectively.

Conclusions: Signet ring cell carcinoma of colon has poor survival rates than the other histological subtypes. Signet ring cell carcinoma presents at an earlier age, higher tumor grade and advanced stage at diagnosis when compared to mucinous and non-mucinous adenocarcinoma of colon. Due to rarity of this disease further multi-institute studies are required for in-depth understanding and analysis of this disease.

No significant financial relationships to disclose.

Prognostic significance of positive lymph node ratio in resected esophageal cancer

100

Background: Nodal involvement in esophageal cancer is associated with poor survival. We aim to determine whether the ratio of metastatic to examined lymph nodes (the lymph node ratio [LNR]) is a better predictor of survival as compared to the number of positive lymph nodes in resected esophageal cancer.

Methods: 1,149 patients with resected esophageal cancer from 1995 to 2009 were identified from the VA Central Cancer Registry (VACCR) database. The patients were further characterized to 3 lymph node quartiles based on LNR and their median survivals were calculated using the Kaplan-Meier method.

Results: Out of 1149 patients 26.4% patients (303) had squamous cell carcinoma and 73.6% (846) were of adenocarcinoma histology. Median age of diagnosis is 63 years. 353 (31%) are stage 1, 384 (33%) are stage 2, and 412 (36%) are stage 3. Majority of them 71% arise in lower third of esophagus followed by 13% in middle third, 4% in upper third and 12 % had unknown site of origin. The group was subdivided into 3 quartiles with 62.7% in LNR1 (0.0-0.1), 25.6% in LNR2 (0.1-0.5) and 11.7% in LNR3 (0.5-1.0). 13.7% had less than 2 nodes removed, 29.3% had 3-6 nodes and 57% had >7 nodes examined. 28% of them had tumor invading sub mucosa, 23.5% had tumor invading muscularis mucosa, 43.2% had involvement of adventitia and 5.3% had penetrating tumor at the time of diagnosis The 5 year survivals based on number of lymph nodes examined, number of positive lymph nodes and positive lymph node ratio are listed in the table. The median overall survival for resected esophageal cancer based on LNR quartiles was 37 vs 14 vs 11.5 months (p<0.0001).

Conclusions: Number of positive lymph nodes and positive lymph node ratio correlated with survival outcomes but number of lymph nodes retrieved did not predict any survival differences. However LNR was a better predictor of survival when compared to number of positive nodes. Further validation of this observation needs to done in large multicenter studies.

[Table: see text]

No significant financial relationships to disclose.

Primary non-Hodgkin lymphoma of the colon among patients in the Veterans Affairs Health System

554

Background: Primary non-Hodgkin lymphoma (NHL) of the colon is rare. There are no randomized controlled trials describing treatment outcomes for this tumor. We provide the largest descriptive study of this tumor to date.

Methods: Retrospective analysis of 109 patients diagnosed with primary non-Hodgkin lymphoma of the colon from 1995 to 2008 was done via the Veteran's Affairs Central Cancer Registry. By definition, all cases presented with a lymphomatous involvement of the colon as the first manifestation of their disease with no previous diagnosis of NHL of any type or site. Demographic, staging, histology, treatment, and outcome data was recorded. Lymphomas were classified as aggressive versus indolent based on their histology.

Results: There were 36,260 colon cancers diagnosed in 1995-2008 of which 109 (0.3%) were primary non-Hodgkin colon lymphomas. The median age of diagnosis was 67 years. 55 pts had aggressive disease, 27 pts had indolent disease, and 27 pts had inadequate histological data. Diffuse large B cell lymphoma (73%) was the most common aggressive lymphoma whereas it was marginal zone (56%) in the indolent group. The indolent group had 5- year survival rate of 76.9% compared to 48.6% for the aggressive group. Both groups had received different treatment regimens as seen in the Table with variable mean survival outcomes.

Conclusions: Our data suggests addition of postoperative adjuvant chemotherapy appears superior to surgery alone in the treatment of aggressive disease whereas it does not appear to provide any benefit in the treatment of indolent disease. However, patient numbers are too small to draw definite conclusions and warrant future investigation in multinational randomized fashion.

[Table: see text]

No significant financial relationships to disclose.

Racial disparities in colon cancer: Retrospective analysis of VACCR database

528

Background: A racial/ethnic difference in colon cancer survival has been described in literature. However the reasons for this are unclear and access to health care is one of the factors that have been implicated. We plan to examine the clinicopathologic factors of Caucasians (CA) and African Americans (AA) in an equal to access health care system.

Methods: 21,992 patients from 1995 to 2009 were identified from the VA Central Cancer Registry database. Age, race, stage, histology, lymph node status, and type of treatment received data were collected.

Results: Out of 21,992 patients, 17,924 were CA and 4,068 were AA. 98.07% of CA and 97.83% of all AA were males. Pathological T stages were as follows: T1 = 15.58% vs. 16.13%, T2 = 18.81% vs. 17.56%, T3 = 55.48% vs. 56.03%, T4 = 10.13% vs. 10.28% for CA and AA respectively. Stage-specific incidence, histological grades, and lymph nodal involvement rates were similar in both groups. 77.16% of CA and 77.03% of all AA received chemotherapy. 82.07% of CA and 78.78% of all AA underwent surgery. Median overall survival in CA was 30.73 months and in AA it was 27 months (p < 0.0001). Stage-specific survival (in months) was significantly better for CA in early-stage disease (p < 0.0001) but is similar for both races in metastatic disease (see Table). Median survival in male and female sex irrespective of race was 29.867 months and 35.667 months respectively.

Conclusions: Overall survival is better in CA when compared to AA in an equal to access health care system. This survival difference was present only in early-stage disease while in metastatic disease the survival was uniformly poor. Incidence as per stage, lymph node status,and grade were not significantly different among AA and CA. Tumor biology and post-treatment surveillance are potential factors and this needs to be investigated further.

[Table: see text]

No significant financial relationships to disclose.

Survival outcomes based on race in gastric carcinoma: A SEER database analysis

5

Background: Gastric cancer is one of the leading causes of cancer related deaths worldwide with the incidence declining in the United States. However the prognosis remains poor with variable survivals being reported among different races. We analyzed the effect of race on patterns of disease presentation and survival rates using the Surveillance, Epidemiology, and End Results (SEER) database.

Methods: A total of 9,851 patients were diagnosed with gastric cancer from year 2004 to 2007 were identified from SEER database. Age, race, TNM staging, grade, treatment modalities utilized and cancer specific survival was collected.

Results: Out of 9,851 patients, 64.63% were white, 12.17% were black, and 13.04% were Asian-Pacific islanders. Median age was 73 years for whites, 70 years in blacks, and 71 years in Asians. Sex distribution amongst races was more or less similar with 58.83% of whites, 59.47% of blacks, and 54.24% of Asians being men. 23.62% of whites had T1 lesions, 28.10% had T2, 19.58% were T3 and 28.70% had T4 lesions. 26.76% of blacks presented with T1 lesions, 26.63% with T2, 16.08% with T3 and 30.53% with T4. 18.69% of Asians had T1 lesions, 26.84% with T2, 23.44% had T3 and 31.03% had T4 lesions. 37.80% of whites, 36.70% of blacks, and 44.44% of Asians had lymph node involvement. Tumor grade was similar among all races. Surgery was performed in 31.49% of whites, 33.13% of blacks, and 40.48% of Asians. 14.68% of whites, 14.10% of blacks, and 19.43 % of all Asians underwent radiation therapy. Median overall survival in localized disease was 44 months, 43 months and 98 months (p < 0.0001) while in regional disease it was 16 months, 15 months and 23 months in whites, blacks and Asians respectively (p < 0.0001). Median survival in distant disease was 4 months in both whites and blacks; it was 5 months in Asians (p < 0.0001).

Conclusions: Cancer-specific survival in gastric carcinoma is significantly better in localized, regional and metastatic disease in Asians when compared to whites and blacks independent of T stage, grade, nodal involvement and treatment modalities utilized. The reason for this observation is unclear, exposure and genetic factors are potential causes and this needs to be investigated.

No significant financial relationships to disclose.

Evaluation of the effect of intravenous calcium and magnesium (CaMg) on chronic and acute neurotoxicity associated with oxaliplatin: Results from a placebo-controlled phase III trial

4025

Background: Cumulative sNT is the dose-limiting toxicity of oxaliplatin which commonly leads to early discontinuation of oxaliplatin-based therapy in the palliative and adjuvant setting. We recently demonstrated the protective effect of CaMg against oxaliplatin-induced sNT as assessed by NCI-CTC (Nikcevich ASCO 2008). It is unclear, though, if CaMg reduced acute and/or chronic, cumulative sNT. Methods: 104 pts with colon cancer undergoing adjuvant therapy with FOLFOX were randomized to IV CaMg (1g calcium gluconate plus 1g magnesium sulfate pre- and post-oxaliplatin) or placebo (PL) in a double-blinded manner. NCI-CTC, an oxaliplatin-specific sNT scale and patient-reported outcome (PRO) questionnaires were used to differentially assess the effect of CaMg on acute (effect on sNT on days 1–4 after oxaliplatin) and chronic sNT (area between curves over whole course of therapy). Results: A total of 102 pts (50 CaMg, 52 PL; 96 mFOLFOX6, 6 FOLFOX4) were available for analysis. Apart from a strong reduction in muscle cramps with CaMg (p=0.002), no difference was found between CaMg and PL in PRO with regard to items reflecting acute sNT (e.g. sensitivity to cold, swallowing of cold liquids, throat discomfort) on days 1–4 after oxaliplatin of any treatment cycle. In contrast, CaMg was able to significantly reduce cumulative sNT in form of numbness in fingers (p=0.02), impaired ability to button shirts (p=0.05), tingling in fingers (p=0.06), and muscle cramps over the course of therapy (p=0.01). Conclusions: In our phase III, placebo-controlled trial, CaMg was able to significantly reduce chronic, cumulative sNT related to oxaliplatin as evaluated by specific PRO questionnaires. No effect was noted on phenomena associated with acute sNT. CaMg can be recommended as neuroprotectant against oxaliplatin-related chronic sNT, oxaliplatin's dose-limiting toxicity.

[Table: see text]

Steroids decrease granulocyte membrane fluidity, while phorbol ester increases membrane fluidity. Studies using electron paramagnetic resonance

High concentrations of corticosteroids inhibit granulocyte responses and disrupt agonist receptor function. Dose-response and time-course considerations make it unlikely that these effects are mediated via the glucocorticoid receptor, a concept further supported by the ability of sex steroids to work similar effects. We postulated that steroids nonspecifically altered granulocyte membrane fluidity, which we measured directly by electron paramagnetic resonance. As predicted, methylprednisolone caused a dose-dependent increase in order parameter (decrease in fluidity) calculated on the basis of EPR spectra, using 5-doxylstearic acid (5-DS) as a probe of resting PMN membranes. This trend was highly significant (P less than 0.001; P at 0.5 mg/ml less than 0.01). Qualitatively similar results (but with different dose-response features) were obtained with conjugated estrogen. Granulocyte agonists (such as PMA) showed an opposite effect, which was not oxidatively mediated and which was steroid-inhibitable. 16-DS showed less prominent effects, suggesting that the membrane leaflets were more strongly affected than was the deep region of the membrane. Ibuprofen, which has similar effects to those of methylprednisolone on PMN aggregation and receptor function, caused a fluidizing rather than a stiffening of the membrane; this surprising result may indicate that there is a critical range of membrane fluidity for normal function, outside of which--in either direction--agonist receptor dysfunction occurs. We conclude that the immediate effects of very high doses of steroids are probably not mediated by corticoid receptors; instead, they may be due to changes in membrane fluidity.

Inhibition of granulocyte function by steroids is not limited to corticoids. Studies with sex steroids

Noting that corticosteroid doses required for protection in shock models exceeded those required to saturate glucocorticoid receptors in mammalian cells, we postulated a nonspecific physicochemical effect of steroids upon the cell membrane, and therefore tested three noncorticoid steroids for their effects on granulocyte function. All three (conjugated equine estrogen, a synthetic progestogen, and a synthetic androgen) behaved in manner analogous to corticoids at similar concentrations, inhibiting granulocyte aggregation, chemotaxis, and chemiluminescence, as well as binding to the granulocytes of the synthetic oligopeptide agonist f-Met-Leu-Phe. Estrogen was further shown to reduce granulocyte membrane fluidity, assessed by electron paramagnetic resonance. We propose that the unique effects of extremely high-dose corticosteroids are not mediated via the glucocorticoid receptor, but result rather from physicochemical effects of the drugs upon the membranes of effector cells.

Hemophilic arthropathy complicated by polyarticular septic arthritis

A case of multiple pyarthroses due to Staphylococcus aureus occurring in a severe classical hemophiliac is presented. Successful management depended on drainage of the infected joints and a prolonged course of antibiotics. Several criteria are suggested for recognition of hemophiliacs who might benefit from joint aspiration as a diagnostic tool for the presence of septic arthritis.