Is there an advantage? Considerations for researchers studying the effects of the type of Medicare coverage

OBJECTIVE

To describe common methodological problems that arise in comparisons of Medicare Advantage (MA) and Traditional Medicare (TM) and within-MA studies and provide suggestions of how researchers can address these issues.

STUDY SETTING

Published research evaluating Medicare coverage options in the United States.

STUDY DESIGN

We considered key conceptual challenges and promising solutions that have been used thus far and suggest additional directions.

DATA COLLECTION

Not available.

PRINCIPAL FINDINGS

Many existing studies of MA versus TM include significant limitations, such as failing to account for unobserved confounders driving both beneficiary coverage choice and health outcomes once enrolled, not accounting for variation in benefit generosity, provider networks, or plan design across MA plans, and/or having been conducted at a time when MA enrollment was less than a third of all Medicare beneficiaries. We provide a review of methods that can help researchers to overcome these weaknesses and suggest additional methods and data sources that may aid future research.

CONCLUSIONS

The MA program is becoming an essential part of the US healthcare system. By accounting for non-random movement into and out of MA and studying the heterogeneity of beneficiary experience across plan and market characteristics, researchers can provide the high-quality evidence necessary for policymakers to design the program and reform TM in ways that maximize beneficiary outcomes.

COVID-19 unemployment and access to statin medications in the United States

Objective

To quantify the effect of the unemployment created by COVID-19 on access to (sales of) statin drugs in the United States population.

Methods

Approximately half a billion transactions for statin drugs in the United States between January 2018 and September 2020 are analyzed. We studied the potential causal relation between abnormal levels of unemployment during the first wave of COVID-19 in the U.S. and abnormal levels of sales of statin products (both variables defined at the state/week level). Variables are analyzed using the Two-Stage Least Squares (2SLS) method, which exploits comparisons of statin sales between states where, given the occupational distribution of their workforce, unemployment was more structurally vulnerable to mobility restrictions derived from COVID-19 against states where it was less structurally vulnerable.

Results

While we do not find unemployment effects on statin sales on most of the population, our estimates link COVID-fueled unemployment with a sharp sales reduction among Medicaid-insured populations, particularly those in working age. For the period between March and August of 2020, these estimates imply a 31% drop of statin sales among this population.

Discussion

COVID-fueled unemployment may have had a negative and significant effect on access to statin populations among Medicaid-insured populations.

Physician Time Burden Associated with Querying Prescription Drug Monitoring Programs

Objective

Prescription drug monitoring programs (PDMPs) enable prescribers to review patient prescription histories, and their use is mandatory in many states. We estimated the cost of physicians retrieving PDMP patient reports compared with a model where a delegate (i.e., administrative staff) retrieves reports.

Methods

We performed a cost analysis with a one-year time horizon, from the perspective of physicians' employers. We obtained specialty-specific estimates of controlled substance prescribing frequency from the National Ambulatory Medical Care Survey, 2012-2014. We defined three PDMP usage cases based on the frequency of queries: comprehensive (before every Schedule II-IV controlled substance prescription), selective (before new Schedule II-IV prescriptions and every six months for continuing medications), and minimal (before new Schedule II or III prescriptions and annually for continuing medications).

Results

The delegate model was less costly for all specialties in the comprehensive usage case and most specialties in the selective usage case, and it was similar to physician model costs in the minimal usage case. Estimated annual costs of the physician model to a large health care system (1,000 full-time equivalent physicians) were $1.6 million for comprehensive usage, $1.1 million for selective usage, and $645,313 for minimal usage. The delegate model was less costly in the comprehensive (savings of $907,283) and selective usage cases (savings of $156,216).

Conclusions

Relying on delegates vs physicians to retrieve reports is less costly in most cases. Automation and integration of PDMP data into electronic health records may reduce costs further. Physicians, health care systems, and states should collaborate to streamline access to PDMPs.

Opioid Discussion in the Twittersphere

BACKGROUND

The rise in opioid use and overdose has increased the importance of improving data collection methods for the purpose of targeting resources to high-need populations and responding rapidly to emerging trends.

OBJECTIVE

To determine whether Twitter data could be used to identify geographic differences in opioid-related discussion and whether opioid topics were significantly correlated with opioid overdose death rate.

METHODS

We filtered approximately 10 billion tweets for keywords related to opioids between July 2009 and October 2015. The content of the messages was summarized into 50 topics generated using Latent Dirchlet Allocation, a machine learning analytic tool. The correlation between topic distribution and census region, census division, and opioid overdose death rate were quantified.

RESULTS

We evaluated a tweet cohort of 84,023 tweets from 72,211 unique users across the US. Unique opioid-related topics were significantly correlated with different Census Bureau divisions and with opioid overdose death rates at the state and county level. Drug-related crime, language of use, and online drug purchasing emerged as themes in various Census Bureau divisions. Drug-related crime, opioid-related news, and pop culture themes were significantly correlated with county-level opioid overdose death rates, and online drug purchasing was significantly correlated with state-level opioid overdoses.

CONCLUSIONS

Regional differences in opioid-related topics reflect geographic variation in the content of Twitter discussion about opioids. Analysis of Twitter data also produced topics significantly correlated with opioid overdose death rates. Ongoing analysis of Twitter data could provide a means of identifying emerging trends related to opioids.

Characteristics of Hospitals Earning Savings in the First Year of Mandatory Bundled Payment for Hip and Knee Surgery

This study uses Medicare data to compare the characteristics of hospitals that did vs did not realize episodes savings under the Centers for Medicare & Medicaid Services’ Comprehensive Care for Joint Replacement program, which bundled payments for hip and knee surgery and paid bonuses to hospitals that exceeded quality and cost benchmarks.

Skin cancer risk in BRCA1/2 mutation carriers

Women with BRCA1/2 mutations have an elevated risk of breast and ovarian cancer. These patients and their clinicians are often concerned about their risk for other cancers, including skin cancer. Research evaluating the association between BRCA1/2 mutations and skin cancer is limited and has produced inconsistent results. Herein, we review the current literature on the risk of melanoma and nonmelanoma skin cancers in BRCA1/2 mutation carriers. No studies have shown a statistically significant risk of melanoma in BRCA1 families. BRCA2 mutations have been linked to melanoma in large breast and ovarian cancer families, though a statistically significant elevated risk was reported in only one study. Five additional studies have shown some association between BRCA2 mutations and melanoma, while four studies did not find any association. With respect to nonmelanoma skin cancers, studies have produced conflicting results. Given the current state of medical knowledge, there is insufficient evidence to warrant increased skin cancer surveillance of patients with a confirmed BRCA1/2 mutation or a family history of a BRCA1/2 mutation, in the absence of standard risk factors. Nonetheless, suspected BRCA1/2 mutation carriers should be counselled about skin cancer risks and may benefit from yearly full skin examinations.

Agreement of dermatopathologists in the evaluation of clinically difficult melanocytic lesions: how golden is the 'gold standard'?

BACKGROUND

The 'gold standard' for the diagnosis of melanocytic lesions is dermatopathology. Although most of the diagnostic criteria are clearly defined, the interpretation of histopathology slides may be subject to interobserver variability.

OBJECTIVES

The aim of this study was to determine the variability among dermatopathologists in the interpretation of clinically difficult melanocytic lesions.

METHODS

This study used the database of MelaFind®, a computer-vision system for the diagnosis of melanoma. All lesions were surgically removed and sent for independent evaluation by four dermatopathologists. Agreement was calculated using kappa statistics.

RESULTS

A total of 1,249 pigmented melanocytic lesions were included. There was a substantial agreement among expert dermatopathologists: two-category kappa was 0.80 (melanoma vs. non-melanoma) and three-category kappa was 0.62 (malignant vs. borderline vs. benign melanocytic lesions). The agreement was significantly greater for patients ≥40 years (three-category kappa = 0.67) than for younger patients (kappa = 0.49). In addition, the agreement was significantly lower for patients with atypical mole syndrome (AMS) (kappa = 0.31) than for patients without AMS (kappa = 0.76).

LIMITATIONS

The data were limited by the inclusion/exclusion criteria of the MelaFind® study. This might represent a selection bias. The agreement was evaluated using kappa statistics. This is a standard method for evaluating agreement among pathologists, but might be considered controversial by some statisticians.

CONCLUSIONS

Expert dermatopathologists have a high level of agreement when diagnosing clinically difficult melanocytic lesions. However, even among expert dermatopathologists, the current 'gold standard' is not perfect. Our results indicate that lesions from younger patients and patients with AMS may be more problematic for the dermatopathologists, suggesting that improved diagnostic criteria are needed for such patients.

Noninvasive genomic detection of melanoma

BACKGROUND

Early detection and treatment of melanoma is important for optimal clinical outcome, leading to biopsy of pigmented lesions deemed suspicious for the disease. The vast majority of such lesions are benign. Thus, a more objective and accurate means for detection of melanoma is needed to identify lesions for excision.

OBJECTIVES

To provide proof-of-principle that epidermal genetic information retrieval (EGIR™; DermTech International, La Jolla, CA, U.S.A.), a method that noninvasively samples cells from stratum corneum by means of adhesive tape stripping, can be used to discern melanomas from naevi.

METHODS

Skin overlying pigmented lesions clinically suspicious for melanoma was harvested using EGIR. RNA isolated from the tapes was amplified and gene expression profiled. All lesions were removed for histopathological evaluation.

RESULTS

Supervised analysis of the microarray data identified 312 genes differentially expressed between melanomas, naevi and normal skin specimens (P<0·001, false discovery rate q<0·05). Surprisingly, many of these genes are known to have a role in melanocyte development and physiology, melanoma, cancer, and cell growth control. Subsequent class prediction modelling of a training dataset, consisting of 37 melanomas and 37 naevi, discovered a 17-gene classifier that discriminates these skin lesions. Upon testing with an independent dataset, this classifier discerned in situ and invasive melanomas from naevi with 100% sensitivity and 88% specificity, with an area under the curve for the receiver operating characteristic of 0·955.

CONCLUSIONS

These results demonstrate that EGIR-harvested specimens can be used to detect melanoma accurately by means of a 17-gene genomic biomarker.

Prospective analysis of predictors of survival in melanoma patients with brain metastases

9074

Background: Melanoma patients (pts) with brain metastases (BM) have limited survival, and BM remains an exclusion criterion in most clinical trials. A recent retrospective analysis at Memorial Sloan Kettering Cancer Center (MSKCC) identified 4 clinical variables that were associated with worse post BM survival (Raizer J et al, Neuro Oncol 2008). In this study, we investigated whether primary tumor features could improve the predictability of post BM survival and examined the reproducibility of the variables identified in MSKCC study. Methods: Melanoma pts with BM prospectively enrolled in an interdisciplinary database at NYU Medical Center from 2002 to 2008 were studied. Six primary tumor characteristics, 21 clinical variables, and treatments were examined. Univariate associations were analyzed using Kaplan Meier survival analysis and the independent effect of identified predictors was assessed by multivariate cox proportional hazards regression analysis. Results: Eighty-nine pts (36 F, 53 M, median age 57) were identified. Median post BM survival was 5.75 months. Median follow-up time based on survivors was 4.2 months. Ulceration and mitotic index ≥3/field were univariately associated with worse post BM survival (p=0.004, p=0.009 respectively). Age >65, ≥3 BM lesions, presence of neurological symptoms, and extracranial metastases were also univariately associated with worse post BM survival (the same 4 variables identified in MSKCC retrospective study). An additional 4 clinical parameters were significant by univariate analysis: frontal lobe location (p=0.01), bilateral lesions (p=0.01), ≥2 neurological symptoms (p=0.005), and weakness/fatigue (p<0.0001). After reproducing the significance of the 4 MSKCC variables in a multivariate model, ulceration of the primary tumor was also an independent predictor of post BM survival (hazard ratio [HR] = 2.75; 95% CI = 1.30, 5.83; p=0.008) whereas mitotic index ≥3/field was not (HR=1.24; 95% CI = 0.57, 2.71; p=0.59). Conclusions: Data suggest that ulceration of the primary melanoma might indicate an adverse biologic behavior that impacts post BM survival. Our data also lend independent support for the predictive model of post BM survival.

No significant financial relationships to disclose.

Evaluation of the melanocortin-1-receptor gene in melanoma predisposition, progression, and recurrence

9018

Background: The melanocortin-1-receptor (MC1R) gene is highly polymorphic in humans. Multiple studies have shown the association between MC1R allelic variants and increased risk of melanoma. Over seventy MC1R single nucleotide polymorphisms (SNPs) have been identified making MC1R a likely target for the development of genetic markers for melanoma predisposition and progression. Alleles described as R and r result in a protein with reduced function compared to wild type, with r alleles having the greatest effect. We sought to investigate the impact of MC1R genotype on risk of developing melanoma and will test for association with disease progression and recurrence in a cohort of melanoma patients enrolled in a prospective study. Methods: Our cohort includes 291 newly diagnosed melanoma patients seen at the New York University Langone Medical Center. The control cohort included 449 subjects. DNA was isolated from leukocytes and analyzed for MC1R SNP status by PCR and direct sequencing. Associations were tested for R-variant R151C and r-variants V60L and V92M. Fisher's Exact test was used to detect significant differences in allele frequency. Odds ratios and confidence intervals were computed for each SNP. Results: Allele frequencies were significantly different in cases and controls for each of the variants (p<0.01). The greatest effect was seen with V60L (Odds Ratio=6.281 95% Confidence Interval 1.77- 22.20). Conclusions: Variants that result in reduced MC1R function increase an individual's risk of developing melanoma. We will continue to test associations with additional variants and determine if these alleles also contribute to risk of disease progression and recurrence. In addition, these variants will be tested for utility as biomarkers for risk assessment of melanoma in a clinical setting.

No significant financial relationships to disclose.

Developing genetic markers for melanoma risk assessment

9046

Background: Risk assessment for melanoma is currently based on phenotype, family and exposure history. This approach is subject to recall bias and excludes at-risk groups such as those with darker skin pigmentation. Poorly stratified risk pools also result in unnecessary dermatologist visits and biopsies for those at lower risk. Use of genetic markers may improve risk assessment; however few susceptibility markers have been developed to date. There have been a number of reports of association between melanoma and genetic markers though few have been replicated or validated. In addition, these studies frequently utilized specific coding region variants as markers and failed to test the entire gene. We have therefore assembled a case-control cohort in which to search for potential biomarkers for melanoma risk by interrogating genes using recently developed tools for genetic analysis. A pilot study was performed to test the utility of our cohort. Methods: A cohort of 326 individuals diagnosed with melanoma and treated at the New York University Langone Medical Center and 400 controls obtained from the New York Cancer project was assembled. Candidate genes were selected based on involvement in determining melanoma predisposition factors (skin pigmentation and DNA repair capability) and previous studies showing association. Three genes, ERCC1, ERCC4 (DNA repair) and MATP (skin pigmentation) were selected. Tag Single Nucleotide Polymorphisms (tSNPs) were selected using Haploview (Hapmap.org) and DNA genotyped (Sequenom Inc, San Diego, CA). Odds ratios and confidence intervals were computed for each SNP. Results: An association was found between SNP rs11615 at the ERCC1 locus and melanoma (Odds ratio = 1.718, 95% Confidence interval: 1.259 - 2.343 for TT vs TC/CC). Conclusions: A tSNP approach is thus useful in identifying associations in our melanoma case-control cohort. Sequence variation at the ERCC1 locus contributes to melanoma risk and the gene will now be screened for clinically useful susceptibility biomarkers. Additional DNA repair and pigmentation genes will also be interrogated using this approach. Genes found to be associated with melanoma will be screened by high- density SNP analysis to identify the most appropriate biomarker/s for use in risk assessment.

No significant financial relationships to disclose.

Association between HU177 serum level and prognosis in patients with primary melanoma

9044

Background: Cleavage of type IV collagen during extracellular matrix (ECM) remodeling leads to exposure of cryptic regulatory sites within the ECM shown to be involved in tumor angiogenesis. Increased levels of a soluble form of the cryptic epitope HU177 in sera of melanoma patients have been shown to be associated with greater tumor thickness and nodular histological subtype. In this study, we investigate the association between HU177 serum levels and melanoma patients' clinical outcomes. Methods: Sera from 209 patients with primary melanoma prospectively enrolled in the Interdisciplinary Melanoma Cooperative Group at the New York University Langone Medical Center (85 females, 124 males, mean age=58, mean thickness=2.09 mm, Stage I n=140, Stage II n=40, Stage III n=29) were analyzed for HU177 level. HU177 serum levels at the time of diagnosis were then correlated with disease-free survival (DFS) and overall survival (OS). Results: Median follow-up time for survivors was 54.9 months (range 2–81 months). Thirty-eight of the 209 (18%) patients developed recurrences, and 34 of the 209 (16%) patients died during follow-up. HU177 sera levels ranged from 0–139.9 ng/ml (mean=6.2 ng/ml; median=3.7 ng/ml). Because the distribution of HU177 levels was positively skewed, we analyzed the data using the median in addition to the mean. HU177 level > 3.7 ng/ml (the median) was associated with a higher rate of melanoma recurrence (p=0.04) and increasing mortality (p=0.01) in a Kaplan Meier analysis. HU177 remained an independent prognostic factor for DFS and OS when controlling for tumor thickness and histological subtype in multivariate Cox proportional hazards regression models. In the DFS hazard model controlling for tumor thickness and histology, the hazard ratio for HU177 >3.7 ng/ml (the median) was 2.01 (95% CI= 1.002, 4.04; p=0.049). In the OS hazard model controlling for tumor thickness and histology, the hazard ratio for HU177 >3.7 ng/ml (the median) was 2.23 (95% CI=1.06, 4.70; p=0.03). Conclusions: Increased serum level of HU177 identifies a subset of primary melanoma patients with worse prognosis and suggests that anti-angiogenic therapy in the adjuvant setting may be a rational approach.

[Table: see text]

Detection of BRAF kinase mutations in melanoma, ovarian, and prostate carcinomas: Evidence for tumor heterogeneity in clinical samples

11031

Background: Several studies have provided evidence that solid tumors are polyclonal malignancies, an observation which may contribute to difficulties in achieving durable treatment responses. In some patients, molecularly targeted therapies may be compromised due to heterogeneity among tumor subclones. In this study we compared conventional DNA sequencing with a fluorescent-based mutant-specific PCR (MS-PCR) assay to detect the BRAF hotspot mutation V600E in a large panel of patient tumors, including paired primary and metastatic tumors from individual patients. Methods: BRAF MS-PCR and conventional sequencing were performed on DNA from 304 tumors (112 melanoma, 110 ovarian, 82 prostate) to determine the presence of the BRAFV600E hot-spot mutation. Among the melanomas were 18 matched primary and metastatic specimens, and 40 metastatic specimens from 19 patients, each of whom had 2 or more metastases. Results: DNA sequencing detected mutations in 5/110 (4.5%) ovarian tumors, 1/82 (1.2%) prostate tumors, and 36/112 (32%) melanomas. In contrast, the MS-PCR assay detected mutations in 12/110 (11%) ovarian tumors, 15/82 (18%) prostate tumors and 85/112 (76%) melanomas. The presence of contaminating normal tissue was scored for each melanoma sample, but excess normal tissue did not influence the results using either methodology. In all cases mutations detected by sequencing were also detected by MSPCR. Among 18 patients with matched primary and metastatic melanoma, 8/18 (44%) had discordant results including 2 patients with mutant primary tumors and wild-type metastases; among the 19 patients with multiple metastases 5/19 (26%) had discordant (both wild-type and mutant) tumors. Conclusions: Using a highly sensitive BRAF mutation detection method, we observed substantial evidence for heterogeneity within clinical tumor specimens. This was especially true in melanoma samples, where multiple specimens from individual patients differed with respect to the presence of the mutant BRAF allele. These results suggest that failures of molecularly targeted therapies, such as those directed against mutant BRAF, may be due in part to a lack of clonality among the tumors under treatment.

No significant financial relationships to disclose.

The unique molecular signatures of nodular and superficial spreading melanoma

9047

Background: Primary nodular melanoma (NM) patients have a relatively poor prognosis compared to superficial spreading melanoma (SSM) patients. The disparity is generally attributed solely to NM's advanced thickness at presentation. In this study we attempted to define molecular signatures of NM and SSM that may explain their clinical differences. Methods: We performed an in silico gene expression analysis of 2 public data sets consisting of 36NM and 54 SSM primary melanoma tissues (CCR 2007;13 and JNCI 2006;98). We then utilized DNA microarray to generate gene expression profiles of a panel of 22 melanoma cell lines (2SSM, 4 NM, 12 met, 4 melanocytes). Differentially expressed genes and over-represented pathways in NM and SSM were identified based on a pooled analysis of the 3 data sets. We then used SNP array to define genomic alterations unique to NM and SSM but not altered in normal melanocytes. Finally, we correlated SNP array with gene expression. Results: Genes significantly overexpressed (p<0.05) in NM showed over-representation of pathways related to MAPK signaling (p=0.05) and cytoskeleton organization (p=0.02), while SSM showed over-representation of cell communication (p=0.05) and primary metabolic processes (p=0.002). Notable correlations between gene expression and copy number alteration in NM include increased copy number/overexpression of SOX5 (transcription factor related to embryonic development and cell fate) and the downregulation/deletion of ST14 (suppression of tumorigenicity 14). SSM demonstrated concordance of increased copy number/overexpression of EZR (cell adhesion protein implicated in human cancer) as well as PALLD (a protein related to motility, adhesion, and extracellular matrix interactions). Notable SSM genes showing correlation between downregulation/deletion include BNIP3 (a pro-apoptotic protein) and MTAP (often co- deleted with tumor suppressor p16). Conclusions: Simultaneous integration of gene expression with SNP array revealed molecular signatures characteristic of NM and SSM. These results suggest that NM and SSM are distinct biologic entities and that molecularly targeted adjuvant therapy may be more effective if tailored to the molecular signatures of melanoma subtypes. Validation is necessary to draw further conclusions.

No significant financial relationships to disclose.

Beware of the isolated intermediate quality measure: A lesson in choosing an ovarian cancer surgeon

6546

Background: Measuring surgical quality has recently focused on isolated processes (such as P4P measures) and intermediate outcomes (such as infection rates). Some have argued that such isolated measures are inherently flawed since they may have only minor influence on more important primary outcomes such as survival. Hence, referral decisions based on such isolated processes or intermediate outcomes may lead to less than optimal care. We asked whether surgeon type influenced isolated process measures (longer operative time, thought to be a function of increased node sampling), intermediate outcomes (deep wound infection, often found to be a function of operative time) and the primary outcome of survival. Methods: We obtained SEER-Medicare claims for the years 1991–2001 for ovarian cancer. We identified the first definitive cancer surgery and the surgical provider type for that procedure: General Surgeon (GS) N=545 and GYN Oncologist (GO) N=1,554. We determined operative time from the anesthesia claims, number of nodes sampled from the SEER record, deep wound infection from the ICD9CM secondary diagnoses and survival from Medicare vital status files. Models were adjusted for 30 patient characteristics and comorbidities. To account for potential reclassification bias due to variations in the pattern of node dissection across GO and GS surgical types, which can potentially influence stage identification, we report survival results both with and without stage adjustments. Results: Definitive surgery by GOs was associated with more extensive node dissection and longer operative time. GO patients also displayed a trend for increased deep wound infection as compared to cases operated on by GS. However, 5-year survival was significantly longer in the GO group, with or without stage adjustment. Conclusions: GO cases took longer, tended to have higher infection rates, yet had better overall 5-year survival. Isolated intermediate outcome measures, such as deep wound infection, may not be good quality indicators.

No significant financial relationships to disclose.

[Table: see text]

BAY 43–9006 (sorafenib-BAY) alters proliferation pathways and mutant specific-PCR (MS-PCR) improves detection of BRAF mutations in metastatic melanoma (MM)

8542

Background: Approximately 60% of melanomas carry BRAF mutations (mut). BAY is a multi-kinase inhibitor that inhibits the BRAF pathway. A phase II trial of BAY in MM is being conducted to: 1) determine if treatment (tx) with BAY can alter proliferation as measured by cyclin D1, Ki-67 and ERK, 2) assess for differential anti-tumor responses and 3) to assess a newly developed fluorescent-based PCR assay to detect mutant BRAF. Methods: Eligibility criteria: Biopsy (bx)-accessible, untreated MM. Measurable disease (RECIST). Stratification of tumor BRAF status determined by routine PCR sequencing for codon 600 mut prior to tx and fluorescent MS-PCR sequencing for confirmation. MS-PCR specifically amplified the M-BRAF allele without amplifying the wild-type (WT)allele. Tx: BAY 400 mg po BID D1–28 q4w. Repeat bx on Day 28. Bx assessed for Ki-67, cyclin-D1 and ERK. Serum collagen cryptic epitopes were measured serially. Re-imaging was done every 2 cycles and pts treated until POD. Results: 29 pts (9-M1a, 8-M1b, 12-M1c) enrolled. 26 wild type (WT) and 3 mutant (M) BRAF by routine PCR. Due to the low yield of mut on routine PCR, MS-PCR was done. 6 mutants were detected in the first 16 pts with MS-PCR compared to 2 with routine sequencing. Median age: 68 (range 22–91). 8 pts with LDH ≥ 1.5 × nl. Tox: Gr I-diarrhea(7), alopecia(4), rash(6), mucositis(4), nausea(4), pain(4), hand-foot(2); Gr II- HTN(4), fatigue(2),mucositis(1), rash(3), pain(3), hand-foot(2); Gr III-hand-foot(1), rash(1), fatigue(1) and intestinal perforation(1). Responses: 12 NE (2 WD, 8 early POD, 1 tox, 1 too early); M BRAF- 1 PR (lymph nodes and large SQ arm masses) and 2 PD; WT BRAF- 1 PR, 7 POD after 2 cycles, 6 SD. Matched paired biopsies demonstrated down regulation of tumor ki-67, erk and cyclin-D1. Collagen cryptic epitopes correlated with tumor responses. Conclusions: MM patients with tumors that were molecularly characterized for BRAF mutational status were entered on this single agent BAY trial and responses were seen in both M and WT arms. Down-regulation of Ki-67, cyclin D1 and ERK was demonstrated and MS-PCR improved the sensitivity to detect BRAF mut. Supported by NCI N01-CM17103 and TRI.

No significant financial relationships to disclose.

A phase II trial of BAY 43–9006 in metastatic melanoma with molecularly characterized B-Raf status

8046

Background: BAY 43–9006 is a multikinase inhibitor. The RAS/BRAF/MEK/MAPK pathway (pw) is a major pw for transmitting extracellular growth factor signals to the nucleus. Activating mutations in any of the pw kinases would be expected to result in uncontrolled cell proliferation. Approximately 60% of melanomas carry B-Raf mutations. The primary trial goals are: 1) determine if treatment (tx) with BAY 43–9006 can disrupt the RAS/BRAF/MEK/MAPK pw by depleting intra-tumor stores of B-Raf and/or CDK4; and 2) determine if tx with BAY 43–9006 results in differential anti-tumor responses in pts whose tumors have been molecularly characterized. Methods: Eligibility criteria: Biopsy (bx)-accessible, untreated metastatic melanoma. Measurable disease (RECIST). Adequate hematologic and serologic parameters. Signed ICF. All pts had tumor bx for B-Raf status determined by PCR sequencing prior to tx. Exon 3 of N-Ras and exon 15 of B-Raf were sequenced. Pts stratified by B-Raf. Tx: BAY 43–9006 was administered at 400 mg po BID. D1–28 q4w. Repeat bx was done on Day 28. Re-imaging was done every 2 cycles and pts treated until POD. Stats: Stratification by B-Raf mutation status. 2 stage study design with accrual of 13 pts/arm in first stage and total of 29 pts/arm if ≥ 1 objective response (OR) in initial cohort. Results: 8 pts (2-M1a, 1-M1b, 5-M1c) enrolled. 7 wild type (WT) and 1 mutant (M) B-Raf. Median age: 54 (range 22–91). No pts with LDH ≥ 1.5 × nl. 2 withdrew prior to tx. Responses: M B-Raf (1 evaluable)-PR (lymph nodes and large SQ arm masses); WT B-Raf-1 POD during cycle 1 (NE), 1PD, 3 too early. Toxicity: Grade I diarrhea, Grade II hypertension, fatigue and oral mucositis. Correlative immunoassays of tumor B-Raf, CDK4, phospho-MAPK and cyclin-D1 and correlation of DNA extracted from peripheral blood cells assayed by fluorescent, mutant-specific PCR and tumor B-RAF are being evaluated. Conclusions: This Phase II trial will molecularly characterize tumors for B-Raf status prior to tx with BAY 43–9006 and assess the effects on Ras-Raf signaling and correlate clinical responses. This trial will address potential differences in sensitivity to BAY 43–9006 based on B-Raf. While premature, 1 pt with M B-Raf has had an OR with a significant decrease in tumor volume. Supported by NCI N01-CM17103 and TRI.

No significant financial relationships to disclose.

Clinical relevance of neutral endopeptidase overexpression in melanoma

8028

Background: Neutral Endopeptidase (NEP/CD10) is a cell surface peptidase expressed in a variety of normal and neoplastic tissue. Both loss of and overexpression of NEP have been reported in different types of tumors. Defining the role of NEP in the clinical progression of melanoma is particularly relevant given the availability of NEP inhibitors. Our aim was to determine if NEP expression correlated with clinicopathological parameters and survival. We also investigated the upregulation of NEP transcription as a possible mechanism of NEP overexpression. Methods: Eighty-four patients who were prospectively enrolled in the Interdisciplinary Melanoma Cooperative Group database at the New York University School of Medicine were studied. Ninety-three specimens from the 84 patients were examined for NEP protein expression using an immunohistochemical assay, including 33 primary melanoma (thickness <1mm, N=15; 1–4mm, N=15; >4mm, N=3) and 60 metastatic melanoma specimens. NEP overexpression was defined as either moderate or diffuse. NEP transcript expression was assessed using Affymetrix U133Plus2.0 GeneChips in 37 metastatic melanoma cases. NEP transcript upregulation was defined as >2 fold increase compared to NEP transcript expression in normal lymph node tissue. Results: A statistically significant difference was observed between NEP overexpression in primary [1/33 (3%)] versus metastatic disease [18/60 (30%)] (Fisher’s exact test, p=0.002). In addition, there was a statistically observed association between NEP transcript upregulation and NEP protein overexpression (p=0.002). Among patients with metastatic disease, median survival time was shorter for patients who overexpressed NEP compared to those who had normal NEP expression (10.2 yrs. versus 12.7 yrs., respectively); however this difference was not statistically significant (p=0.74 by log-rank test). Conclusions: Our data support a role for increased NEP expression in the progression of melanoma to metastatic disease. Functional studies are needed to better understand the role of NEP upregulation in melanoma pathogenesis and progression and to determine if there is a role for NEP inhibitors in the treatment of melanoma.

No significant financial relationships to disclose.

Does ovarian cancer survival differ by who gives the chemotherapy?

6110

Background: Medical Oncologists (MOs) specialize in administering chemotherapy while Gynecologic Oncologists (GOs) specialize in the surgical management of gynecologic cancers as well as providing chemotherapy. We asked whether survival differed according to the type of specialist providing the chemotherapy after ovarian cancer surgery. Methods: Using merged SEER-Medicare clinical and administrative data from the years 1991 through 2001 we identified all patients who were diagnosed with ovarian cancer after the age of 65 and who received chemotherapy from a GO after surgery performed by any type of surgeon. Using optimal matching and a propensity score based on 35 prognostic characteristics, we identified similar patients who were operated on, and staged, by nearly identical types of surgeons but who received chemotherapy from an MO. Results: We identified 344 patients who were given chemotherapy for ovarian cancer by GOs after having undergone surgery by either GOs (76%), Gynecologists (16%), or general surgeons or other surgical specialties (8%). Patients given chemotherapy by MOs were matched to these patients. Both groups had nearly identical surgical specialists, age, year of treatment, stage, tumor grade, race and 32 other prognostic factors. There was no difference in overall survival between the GO or MO groups (P = 0.45, paired Prentice-Wilcoxon test). The 5 year survival was 35% (± 2.6%) for the GO group and 34% (± 2.6%) for the MO group. However, MOs administered chemotherapy over more weeks than did the GOs (patient mean = 16.5 Vs 12.1 weeks, P < 0.0023, Wilcoxon rank sum test) and MO patients had more weeks that included at least one of the following chemotherapy associated adverse events (neutropenia, anemia, thrombocytopenia, diarrhea, dehydration or mucositis) than did the GO group (patient mean = 16.2 Vs 8.9 weeks, P < 0.0001, Wilcoxon rank sum test). Conclusions: Despite differences in training, specialty emphasis, and practice style concerning the use of chemotherapy (MO patients had more weeks of chemotherapy and more weeks with chemotherapy associated adverse events than GO patients), we found no difference in survival between patients who received chemotherapy administered by gynecologic oncologists and medical oncologists.

No significant financial relationships to disclose.

Altered patterns of RB expression define groups of soft tissue sarcoma patients with distinct biological and clinical behavior

BACKGROUND

Function of the retinoblastoma tumor suppressor protein (pRB) may be compromised at a genetic level by gene loss or mutation or at a post-translational level by hyperphosphorylation. In this study, we examined adult soft tissue sarcomas (ASTS) to determine if alterations of pRB were associated with distinct patterns of pRB expression and clinical outcome.

DESIGN

We investigated 86 ASTS patients using monoclonal antibodies that distinguish between hyperphosphorylated and underphosphorylated pRB products. We also used microsatellite analysis to investigate the genetic status of the RB locus. We correlated pRB alterations with proliferative activity, and with clinicopathological outcomes.

RESULTS

Altered patterns of pRB expression are common in ASTS occurring in 84% of cases, and it is significantly associated with proliferative activity (p<0.001). Patients whose tumors either lack expression of pRB, or express hyperphosphorylated forms of pRB, have poor survivals compared to patients whose tumors exhibit a normal, underphosphorylated pattern of pRB expression (p=0.03). In addition, 63% of cases lacking expression of pRB showed loss-of-heterozygosity at the locus.

CONCLUSIONS

Inactivation of pRB is common in adult STS, which may be due to either gene loss or post-translational modification, namely hyper-phosphorylation. Both mechanisms are associated with tumor cell proliferation and poor survival.

Specialty care by child neurologists: a workforce analysis

OBJECTIVE

To provide a current profile of the practice of child neurology, report the attitudes of child neurologists toward practice, and analyze the supply of child neurologists.

METHODS

In March 2002, a questionnaire was sent to all active members of the Child Neurology Society (n = 1,051) and to nonmember physicians under age 70 who listed child neurology as a primary or secondary specialty on the American Medical Association Masterfile (n = 433). The response rate was 65%. Eligibility criteria were then applied to arrive at the sample of main specialty in child neurology working at least 20 hours per week in patient care. The final population was 604. Differences in practice characteristics were tested by practice type, and the number of full-time patient care child neurologists was projected by extrapolating to nonrespondents.

RESULTS

There are 904 full-time patient care child neurologists in the United States and 1.27 per 100,000 children. Career satisfaction is 90%, yet no growth in the supply is projected over the next 20 years. Wait times for an appointment average 53 and 44 days for a new and return visit, with longer wait times in university settings. Average annual income is 151,000 dollars.

CONCLUSION

The practice characteristics of child neurologists suggest that the specialty will be challenged to meet patient demands.

Nursing shortages and international nurse migration

BACKGROUND

The United Kingdom and the United States are among several developed countries currently experiencing nursing shortages. While the USA has not yet implemented policies to encourage nurse immigration, nursing shortages will likely result in the growth of foreign nurse immigration to the USA. Understanding the factors that drive the migration of nurses is critical as the USA exerts more pull on the foreign nurse workforce.

AIM

To predict the international migration of nurses to the UK using widely available data on country characteristics.

METHOD

The Nursing and Midwifery Council serves as the source of data on foreign nurse registrations in the UK between 1998 and 2002. We develop and test a regression model that predicts the number of foreign nurse registrants in the UK based on source country characteristics. We collect country-level data from sources such as the World Bank and the World Health Organization.

RESULTS

The shortage of nurses in the UK has been accompanied by massive and disproportionate growth in the number of foreign nurses from poor countries. Low-income, English-speaking countries that engage in high levels of bilateral trade experience greater losses of nurses to the UK.

CONCLUSION

Poor countries seeking economic growth through international trade expose themselves to the emigration of skilled labour. This tendency is currently exacerbated by nursing shortages in developed countries. Countries at risk for nurse emigration should adjust health sector planning to account for expected losses in personnel. Moreover, policy makers in host countries should address the impact of recruitment on source country health service delivery.

Association of melanoma and neurocutaneous melanocytosis with large congenital melanocytic naevi--results from the NYU-LCMN registry

BACKGROUND

Large congenital melanocytic naevi (LCMN), which develop in utero and are present in approximately one in 20,000 newborns, are associated with markedly increased risks of cutaneous melanoma, leptomeningeal melanoma and neurocutaneous melanocytosis (NCM).

OBJECTIVES

This study examined clinical characteristics associated with melanoma and NCM among patients with LCMN, and estimated the risk of developing melanoma and NCM in these patients.

METHODS

Two hundred and five LCMN patients enrolled in the New York University registry were studied. One hundred and seventy of these patients were followed prospectively. The remaining 35 patients had either melanoma at the time of entry into the registry (n = 6), or had insufficient follow-up information (n = 29). The outcome measures were the occurrence of melanoma and NCM. The associations between these outcomes and the clinical covariates (anatomical location of the LCMN, size of the LCMN, number of satellite lesions, family history of melanoma, patient sex and treatment) were assessed.

RESULTS

Four of 170 (2.3%) prospectively followed patients developed melanomas, representing a standardized morbidity ratio of 324. Among the entire cohort (n = 205), there were associations between increasing numbers of satellite naevi and the occurrence of melanoma (P = 0.04), and the presence of NCM (P = 0.06). Compared with patients who did not develop these diseases, median LCMN diameters were larger among patients who developed melanoma (49 vs. 39 cm) and NCM (55 vs. 46 cm).

CONCLUSIONS

In LCMN patients, increasing numbers of satellite lesions and larger LCMN diameters are associated with melanoma and NCM.

Medication Safety in Older Adults: Home-Based Practice Patterns

OBJECTIVES

To identify the current state of medication-taking practices of community-dwelling older adults on high-risk medications.

DESIGN

Telephone survey of older adults filling prescriptions for warfarin, digoxin, or phenytoin from May 2, 2002, through June 30, 2003.

SETTING

The Pennsylvania Pharmacy Assistance Contract for the Elderly (PACE) Program, a state-run program providing prescription drug coverage for poor older adults.

PARTICIPANTS

PACE members.

MEASUREMENTS

Patients self-reported sources of information on current medications as well as home-based practices for organizing medication regimens.

RESULTS

Four thousand nine hundred fifty-five PACE members were interviewed. Thirty-two percent of the sample reported that they had not received any specific instructions about their medications, 35% reported that they received instructions from their primary care provider, and 46% indicated they received them from a pharmacist. Fifty-four percent of participants indicated that they used a pillbox for organizing their medications. Older adults prescribed warfarin were more likely to report receiving instructions than those prescribed digoxin or phenytoin.

CONCLUSION

A substantial proportion of older adults on high-risk medications do not recall receiving instructions for the use of their medications and do not take advantage of existing systems for organizing medication regimens. Improved patient education and delivery of medication organization systems are immediate opportunities to potentially reduce the risk of medication errors in older people.

Evaluation of germline CDKN2A, ARF, CDK4, PTEN, and BRAF alterations in atypical mole syndrome

Atypical mole syndrome is a sporadic or an inherited condition with an increased risk of melanoma. Germline mutations in the CDKN2A, ARF, CDK4 and somatic mutations in the PTEN and BRAF genes have been associated with melanoma. In this study, we evaluated genes associated with familial and sporadic melanoma for mutations in 28 probands with the atypical mole syndrome. No sequence alterations in the coding regions or in the splice junctions of CDKN2A, ARF, CDK4, PTEN or BRAF were identified. These data suggest that genes evaluated in this study are unlikely to be candidate genes for atypical mole syndrome and support the notion that unknown susceptibility gene/s for this disease exist.

A general model of the impact of absenteeism on employers and employees

Most studies on the indirect costs of an illness and the cost effectiveness of a medical intervention or employer-sponsored wellness program assume that the value of reducing the number of days employees miss from work due to illness is the wage rate. This paper presents a general model to examine the magnitude and incidence of costs associated with absenteeism under alternative assumptions regarding the size of the firm, the production function, the nature of the firm's product, and the competitiveness of the labor market. We conclude that the cost of lost work time can be substantially higher than the wage when perfect substitutes are not available to replace absent workers and there is team production or a penalty associated with not meeting an output target. In the long run, workers are likely to bear much of the incidence of the costs associated with absenteeism, and therefore be the likely beneficiaries of any reduction in absenteeism.

HDM2 protein overexpression, but not gene amplification, is related to tumorigenesis of cutaneous melanoma

We investigated the role of alterations of HDM2, the human homologue of murine mdm2, in the tumorigenesis and progression of cutaneous melanoma. A well-characterized cohort of 172 cases representing different points in the spectrum of melanocyte transformation (16 dysplastic nevi, 11 melanomas in situ, 107 invasive primaries, and 38 metastatic lesions), as well as 11 human melanoma cell lines were examined by immunohistochemistry and Western blotting for HDM2 protein expression, and by either Southern blotting (SB) or fluorescence in situ hybridization for HDM2 gene amplification. HDM2 overexpression, defined as >20% tumor cells showing nuclear immunoreactivity, was observed in 1 of 16 (6%) dysplastic nevi, 3 of 11 (27%) melanomas in situ, and 81 of 145 (56%) invasive primary and metastatic melanomas. Comparable frequencies of HDM2 overexpression were observed among invasive primary cases with differing tumor thicknesses as well as among the metastatic cases: 21 of 40 (53%) at < or =1.5 mm; 31 of 50 (62%) at 1.6-3.9 mm; 10 of 17 (58%) at >4 mm; and 19 of 38 (50%) metastases. HDM2 amplification was observed in 1 of 88 (1%) primary cases using fluorescence in situ hybridization, and in 0 of 12 (0%) metastatic cases that overexpressed HDM2 using SB. Melanoma cell lines expressed HDM2 protein, but there was no evidence of amplification by SB. Our data suggest that HDM2 protein overexpression is common in invasive and metastatic melanoma. Observing HDM2 overexpression in noninvasive melanoma suggests that expression of this oncogene may play an early role in melanocyte transformation. HDM2 amplification occurs infrequently, and other mechanisms that up-regulate HDM2 expression are under investigation.

The transcriptional repressor of p16/Ink4a, Id1, is up-regulated in early melanomas

The helix-loop-helix transcription factor Id1 coordinates cell growth and differentiation pathways within mammalian cells and has been implicated in regulating G(1)-S phase cell cycle transitions. Recently Id1 has been shown to repress Ets- and E-protein-mediated transactivation of p16/Ink4a. Because the p16/Ink4a protein has been demonstrated to be inactivated in subsets of familial and sporadic melanomas, we sought to determine whether Id1 regulation of p16/Ink4a expression might be involved in the development of this human tumor. Here we evaluate 21 melanocytic lesions at various stages of malignant progression from common melanocytic nevi to metastatic melanomas and examine these lesions for Id1 and p16/Ink4a expression. We demonstrate that Id1 expression correlates with loss of p16/Ink4a expression in melanoma in situ; however, more advanced stages of melanoma do not express Id1 except within perivascular regions, despite overall decreased p16/Ink4a expression in these lesions. Microdissected lesions were evaluated for p16/Ink4a sequence, and invasive melanomas that did not express Id1 were found to have sustained inactivating p16/ink4a mutations. These data suggest a role for Id1 in regulating p16/Ink4a expression in early melanomas and demonstrate that later genetic changes may provide for irreversible loss of p16 expression in advanced stages of this tumor.

High Ki-67 proliferative index predicts disease specific survival in patients with high-risk soft tissue sarcomas

BACKGROUND

Soft tissue sarcomas (STSs) are heterogeneous neoplasms that have variable clinical outcome. Several clinical parameters and few molecular markers, including Ki-67 proliferative index, have been shown to correlate with patient prognosis. To the authors' knowledge, no definitive report exists to identify one molecular marker that can be analyzed easily in a clinical setting and that predicts survival in a cohort of patients with high-risk STS of identical clinical characteristics but variable outcome.

METHODS

The influence of clinical prognostic factors was eliminated by selecting two patient groups with identical high-risk characteristics: large (> 10 cm), high-grade, deep, completely resected primary extremity STS (n = 47). Patients in the first group remained disease free (no evidence of disease [NED]) after primary tumor treatment (n = 19), whereas patients in the second group subsequently died of disease (DOD; n = 28). Triplicate 0.6-mm core biopsies from defined morphologic areas of paraffin embedded primary tumors were assembled on a tissue microarray and analyzed by immunohistochemistry with the MIB-1 antibody, and Ki-67 proliferative indices were correlated with patient outcome.

RESULTS

High Ki-67 proliferative index, defined as greater than 30% tumor cells showing nuclear immunoreactivity, was significantly more frequent in the DOD group than in the NED group and was associated with tumor-related mortality (P = 0.02). This marker identifies an especially aggressive malignant phenotype within a cohort of high-risk tumors that is based on well established clinical and pathologic parameters alone and is easy to use in a clinical setting.

CONCLUSIONS

On the basis of these data and previous reports, high Ki-67 proliferative index is suggested as a significant factor for predicting the prognosis of patients with high-risk STS and should be evaluated prospectively based on clinical trials.