PubMed Health⌕ Search

Biomedical subjects

Dianne M Finkelstein

Publications and source records attributed to Dianne M Finkelstein.

At least 19 recordsLinked to original sources
PubMed2007

Regression analysis of failure time data with informative interval censoring.

Interval censoring arises when a subject misses prescheduled visits at which the failure is to be assessed. Most existing approaches for analysing interval-censored failure time data assume that the censoring mechanism is independent of the true failure time. However, there are situations where this assumption may not hold. In this paper, we consider such a situation in which the dependence structure between the censoring variables and the failure time can be modelled through some latent variables and a method for regression analysis of failure time data is proposed. The method makes use of the proportional hazards frailty model and an EM algorithm is presented for estimation. Finite sample properties of the proposed estimators of regression parameters are examined through simulation studies and we illustrate the method with data from an AIDS study.

Acquired Immunodeficiency Syndrome↗
PubMed2006

Survival analysis of longitudinal microarrays.

MOTIVATION: The development of methods for linking gene expressions to various clinical and phenotypic characteristics is an active area of genomic research. Scientists hope that such analysis may, for example, describe relationships between gene function and clinical events such as death or recovery. Methods are available for relating gene expression to measurements that are categorized or continuous, but there is less work in relating expressions to an observed event time such as time to death, response or relapse. When gene expressions are measured over time, there are methods for differentiating temporal patterns. However, methods have not yet been proposed for the survival analysis of longitudinally collected microarrays. RESULTS: We describe an approach for the survival analysis of longitudinal gene expression data. We construct a measure of association between the time to an event and gene expressions collected over time. Statistical significance is addressed using permutations and control of the false discovery rate. Our proposed method is illustrated on a dataset from a multi-center research study of inflammation and response to injury that aims to uncover the biological reasons why patients can have dramatically different outcomes after suffering a traumatic injury (www.gluegrant.org).

Algorithms↗
PubMed2006

Perioperative CA19-9 levels can predict stage and survival in patients with resectable pancreatic adenocarcinoma.

PURPOSE: Different prognostic factors stratify patients with pancreatic adenocarcinoma. The purpose of this study was to determine whether preoperative CA19-9 levels can predict stage of disease or survival and whether a change in preoperative to postoperative CA19-9 or the postoperative CA19-9 predicts overall survival. PATIENTS AND METHODS: Four hundred twenty-four consecutive patients with pancreatic adenocarcinoma underwent resection between January 1, 1985 and January 1, 2004. Of the patients with a bilirubin less than 2 mg/dL, 176 had preoperative CA19-9 values, and 111 had pre- and postoperative CA19-9 values. Survival was measured from the first postoperative CA19-9 level measured (median, 39 days) until death or last follow-up. A multivariate failure time model was fit using clinical, operative, pathologic, and adjuvant treatment characteristics, and a categorization was defined by the values and changes in CA19-9 before and after surgery. RESULTS: Of the 176 patients, 128 (73%) had T3 lesions, and 99 (56%) had N1 disease; 138 patients (78%) underwent pancreaticoduodenectomy. Median preoperative CA19-9 levels were lower in N0 patients compared with patients with positive nodes (nine v 164 U/mL, respectively; nonparametric P = .06) and in T1/T2 patients versus T3 patients (41 v 162 U/mL, respectively; P = .03). Median follow-up time (n = 111) was 1.8 years (range, 1 to 12.9 years), with overall actuarial 1-, 3-, and 5-year survival rates of 70%, 36%, and 30%, respectively. Significant predictors of survival on multivariate analysis included a decrease in CA19-9 (P = .0005), negative lymph nodes (P = .001), lower T stage (P = .0008), and postoperative CA19-9 less than 200 U/mL (P = .0007). CONCLUSION: In patients with pancreatic adenocarcinoma, preoperative CA19-9 correlates with stage of disease. Both a postoperative decrease in CA19-9 and a postoperative CA19-9 value of less than 200 U/mL are strong independent predictors of survival, even after adjusting for stage. CA19-9 levels should be included in a patient's perioperative care and should be considered for prognostic nomograms.

Adenocarcinoma↗
PubMed2006

Analysis of co-aggregation of cancer based on registry data.

OBJECTIVE: An exploratory analysis of co-aggregation of cancers using registry-based data. METHODS: We utilized sibships from over 18,000 families who had been recruited to the NCI-sponsored multi-institutional Cancer Genetics Network. The analysis assesses co-aggregation at the individual and family level and adjusts for ascertainment. RESULTS: We found statistically significant familial co-aggregation of lung cancer with pancreatic (adjusted p < 0.001), prostate (adjusted p < 0.003), and colorectal cancers (adjusted p = 0.004). In addition, we found significant familial co-aggregation of pancreatic and colorectal cancers (adjusted p = 0.018), and co-aggregation of hematopoietic and (non-ovarian) gynecologic cancers (adjusted p = 0.01). CONCLUSION: This analysis identified familial aggregation of cancers for which a genetic component has yet to be established.

Female↗
PubMed2005

Phase I trial and pharmacokinetics of escalating doses of paclitaxel and concurrent hyperfractionated radiotherapy with or without amifostine in patients with advanced head and neck carcinoma.

BACKGROUND: Amifostine was developed to protect normal tissues from radiation exposure. The current study was undertaken to determine whether amifostine would allow the delivery of greater numbers of weekly paclitaxel treatments with concomitant, hyperfractionated radiotherapy in patients with advanced head and neck carcinoma. METHODS: Patients received radiation therapy twice daily using 1.6-gray (Gy) fractions up to a total of 70.4 Gy over an elapsed time of 6.5 weeks. All patients received paclitaxel 60 mg/m(2) once weekly starting on Day 1. The number of doses of paclitaxel was escalated from three to a maximum of six in groups of three patients. For the patients who received amifostine, a dose of 400 mg/m(2) was given intravenously over 15 minutes on Days 1-5, 8, 29-33, and 36. Patients underwent surgery for persistent tumor after radiotherapy. The plasma pharmacokinetics of paclitaxel were characterized during treatment with the first weekly dose to determine the effect of concurrently administered amifostine. RESULTS: Thirty-six patients were evaluable for this study. In the absence of amifostine, a maximum of four doses of paclitaxel were tolerated in combination with the radiotherapy. With amifostine, up to five doses of paclitaxel could be given. Generally, the treatment resulted in Grade 2 and 3 stomatitis. Overall, 69% of patients had a complete remission, and 29% had a partial remission. Both progression-free survival and overall survival were 66% at 30 months. Amifostine had no effect on the pharmacokinetics of paclitaxel. CONCLUSIONS: The administration of amifostine allowed the authors to give an additional dose of paclitaxel to patients who were undergoing hyperfractionated radiotherapy for head and neck carcinoma. This treatment regimen resulted in a high frequency of complete remissions and an excellent progression-free survival pattern without compromising the plasma kinetics of paclitaxel.

Amifostine↗
PubMed2005

Brain tumor oncolysis with replication-conditional herpes simplex virus type 1 expressing the prodrug-activating genes, CYP2B1 and secreted human intestinal carboxylesterase, in combination with cyclophosphamide and irinotecan.

The treatment of malignant glioma is currently ineffective. Oncolytic viruses are being explored as a means to selectively lyse tumor cells in the brain. We have engineered a mutant herpes simplex virus type 1 with deletions in the viral UL39 and gamma(1)34.5 genes and an insertion of the two prodrug activating genes, CYP2B1 and secreted human intestinal carboxylesterase. Each of these can convert the inactive prodrugs, cyclophosphamide and irinotecan (CPT-11), into their active metabolites, respectively. This new oncolytic virus (MGH2) displays increased antitumor efficacy against human glioma cells both in vitro and in vivo when combined with cyclophosphamide and CPT-11. Importantly, cyclophosphamide, CPT-11, or the combination of cyclophosphamide and CPT-11 does not significantly affect oncolytic virus replication. Therefore, MGH2 provides effective multimodal therapy for gliomas in preclinical models when combined with these chemotherapy agents.

Antineoplastic Combined Chemotherapy Protocols↗
PubMed2005

Results of whole-brain radiation as salvage of methotrexate failure for immunocompetent patients with primary CNS lymphoma.

PURPOSE: This study evaluates the efficacy and toxicity of whole-brain radiation therapy (WBRT) as salvage therapy for immunocompetent patients who failed initial high-dose methotrexate for primary CNS lymphoma (PCNSL). PATIENTS AND METHODS: The study cohort included 27 consecutive patients who failed initial high-dose methotrexate and then received salvage WBRT (median dose, 36 Gy). Actuarial survival was measured from the initiation of radiotherapy. RESULTS: Ten patients (37%) achieved a complete radiographic response (CR), and 10 patients (37%) a partial response to WBRT, for a 74% overall radiographic response rate. At the time of maximal response, Karnofsky performance status improved in 12 (44%) of 27 patients and at least stabilized in 67%. Median estimated survival from initiation of WBRT was 10.9 months (range, 0.3 to 63.7 months). The univariate predictor of longer survival was age less than 60 years at the time of WBRT (P = .028). Among patients who survived 4 months, achievement of a CR to WBRT by 4 months (P = .002) predicted longer survival. Late treatment-associated neurotoxicity was diagnosed in four patients (15%) and was significantly associated with total radiation doses greater than 36 Gy (P = .04). No patient treated with daily fractions less than 1.8 Gy developed late neurotoxicity. CONCLUSION: For patients with PCNSL who experience treatment failure with methotrexate, WBRT provides high response rates (74%) and a median survival of 10.9 months. Age less than 60 years and response to WBRT predict post-WBRT survival. Modest rates of late neurotoxicity (15%) were seen and were associated with a total dose greater than 36 Gy.

Antimetabolites, Antineoplastic↗
PubMed2003

Modelling the relationship between continuous covariates and clinical events using isotonic regression.

In a medical study we are often interested in graphically displaying the relationship between continuous variables and clinical events indicating disease progression. Often, it is reasonable to make the minimal assumption that the risk of progression is an arbitrary monotone function of the continuous variable. Sometimes the continuous variable is a disease marker which is recorded longitudinally, and so the goal is to provide a graphical display of the relationship between the hazard for progression and the most recent measurement of the longitudinal marker. For example, we know that for a patient with HIV infection, declining CD4 count is associated with an increased risk of opportunistic infection. The goal of this paper is to extend isotonic regression techniques to failure time data with a continuous covariate, to obtain a non-parametric estimate for the hazard of disease progression as a monotonic function of the continuous variable. We propose two methods for modelling the relationship of the hazard and covariate: the first assumes that the hazard is constant over time, and the second allows the hazard to be an arbitrary function of time. These methods will be applied to graphically display the risk for an AIDS patient of an opportunistic infection as a function of CD4 count.

AIDS-Related Opportunistic Infections↗
PubMed2003

Regression of microalbuminuria in type 1 diabetes.

BACKGROUND: In the present study, we aimed to determine the frequency of a significant reduction in urinary albumin excretion and factors affecting such reduction in patients with type 1 diabetes and microalbuminuria. METHODS: The study included 386 patients with persistent microalbuminuria, indicated by repeated measurements of urinary albumin excretion (estimated on the basis of albumin-to-creatinine ratios) in the range of 30 to 299 microg per minute during an initial two-year evaluation period. Subsequent measurements during the next six years were grouped into two-year periods, averaged, and analyzed for regression of microalbuminuria, which was defined as a 50 percent reduction in urinary albumin excretion from one two-year period to the next. RESULTS: Regression of microalbuminuria was frequent, with a six-year cumulative incidence of 58 percent (95 percent confidence interval, 52 to 64 percent). The use of angiotensin-converting-enzyme inhibitors was not associated with the regression of microalbuminuria. However, microalbuminuria of short duration, salutary levels of glycosylated hemoglobin (less than 8 percent), low systolic blood pressure (less than 115 mm Hg), and low levels of both cholesterol and triglycerides (less than 198 mg per deciliter [5.12 mmol per liter] and 145 mg per deciliter [1.64 mmol per liter], respectively) were independently associated with the regression of microalbuminuria. Patients with salutary levels of all modifiable factors had a hazard ratio for regression of 3.0 (95 percent confidence interval, 1.5 to 6.0), as compared with patients with no salutary levels of any modifiable factor. CONCLUSIONS: Frequent regression of microalbuminuria in patients with type 1 diabetes indicates that elevated urinary albumin excretion does not imply inexorably progressive nephropathy. Identification of the multiple determinants of the regression of microalbuminuria has implications for current theories about the mechanisms of early diabetic nephropathy.

Adolescent↗
PubMed2003

Prognosis and management of extramammary Paget's disease and the association with secondary malignancies.

BACKGROUND: Extramammary Paget's disease (EMPD) is a rare clinical entity and can be associated with other malignancies. We analyzed our experience for prognosis and impact of therapy on outcomes. STUDY DESIGN: We conducted a retrospective review of all patients (n = 33) with EMPD treated at a tertiary care center from 1971 to 1998. Pathologic features of EMPD, concurrent secondary malignancies, and the effect of operations on recurrence were analyzed. Overall survival was compared with that of the general population. RESULTS: Male-to-female ratio was 4:29, and median age was 70 years. Median followup was 68 months, and no patient died from EMPD. The lesion was predominantly found on the vulva (76%). Patch-like, nonconfluent growth was present in 45% of patients, and no patient had pathologic lymph nodes. The most common signs and symptoms were irritation or pruritus (73%) and rash (61%). The presence of patches, invasive tumor growth, or a second malignancy were significantly associated with a higher recurrence rate. The type of operation, either local excision or hemivulvectomy, was not related to the time to recurrence. Complete gross resection was achieved in 94% of cases. Fifty-six percent of patients had microscopically positive margin and this correlated with a significantly higher recurrence rate (p = 0.002). The tumor recurred clinically in 14 of 33 patients (42%) after a median of 152 months (range 5 to 209 months). In those patients, between one and six reexcisions were performed. In 14 of 33 patients with EMPD (42%), 16 concurrent secondary malignancies were found. Overall survival rates for EMPD patients were similar to those of the general population. CONCLUSIONS: EMPD is an infrequently diagnosed disease that is preferably managed with complete local excision and reexcisions if needed. A thorough search for frequently occurring secondary malignancies might be beneficial to provide the best outcomes for these patients.

Adenocarcinoma↗
PubMed2002

Predicting time to prostate cancer recurrence based on joint models for non-linear longitudinal biomarkers and event time outcomes.

Biological markers that are both sensitive and specific for tumour regrowth or metastasis are increasingly becoming available and routinely monitored during the regular follow-up of patients treated for cancer. Obtained by a simple blood test, these markers provide an inexpensive non-invasive means for the early detection of recurrence (or progression). Currently, the longitudinal behaviour of the marker is viewed as an indicator of early disease progression, and is applied by a physician in making clinical decisions. One marker that has been studied for use in both population screening for early disease and for detection of recurrence in prostate cancer patients is PSA. The elevation of PSA levels is known to precede clinically detectable recurrence by 2 to 5 years, and current clinical practice often relies partially on multiple recent rises in PSA to trigger a change in treatment. However, the longitudinal trajectory for individual markers is often non-linear; in many cases there is a decline immediately following radiation therapy or surgery, a plateau during remission, followed by an exponential rise following the recurrence of the cancer. The aim of this article is to determine the multiple aspects of the longitudinal PSA biomarker trajectory that can be most sensitive for predicting time to clinical recurrence. Joint Bayesian models for the longitudinal measures and event times are utilized based on non-linear hierarchical models, implied by unknown change-points, for the longitudinal trajectories, and a Cox proportional hazard model for progression times, with functionals of the longitudinal parameters as covariates in the Cox model. Using Markov chain Monte Carlo sampling schemes, the joint model is fit to longitudinal PSA measures from 676 patients treated at Massachusetts General Hospital between the years 1988 and 1995 with follow-up to 1999. Based on these data, predictive schemes for detecting cancer recurrence in new patients based on their longitudinal trajectory are derived.

Bayes Theorem↗
PubMed2002

Quantitatively determined survivin expression levels are of prognostic value in human gliomas.

PURPOSE: Survivin is a novel antiapoptotic gene that has been recently cloned and characterized. Its expression has been found to be of prognostic significance in several tumor types. This is the first study on the prognostic significance of survivin expression in human gliomas. MATERIALS AND METHODS: We used quantitative Western blot analysis with densitometry to determine survivin protein expression levels in 92 glioma cases for which frozen tissue was available for analysis. Survivin positivity and expression levels were correlated with histopathologic features of the tumors, apoptosis (as measured by cleaved, or activated, caspase 3 levels), and clinical outcome. RESULTS: Survivin expression has clear prognostic value in human gliomas. Patients with detectable survivin expression had significantly shorter overall survival times (P <.0001) compared with those without detectable expression when all glioma patients were considered. Although glioblastoma multiforme (GBM) patients had significantly higher rates of survivin positivity and higher levels of survivin expression (P <.0001) than their non-GBM counterparts, the prognostic value of survivin expression seemed to be independent of histology alone. Survivin-positive GBM patients had significantly shorter overall survival times compared with survivin-negative GBM patients (P <.0001). Likewise, survivin-positive non-GBM patients had shorter survival times compared with survivin-negative non-GBM patients (P =.029). Furthermore, increasing levels of survivin expression significantly correlated with reduced survival times when all glioma patients were considered, and markedly so for GBM patients (P <.0001). Increasing survivin levels significantly correlated with reduced expression of cleaved caspase 3, indicating its association with antiapoptotic activity. CONCLUSION: Survivin positivity and protein expression levels, as determined quantitatively, are of significant prognostic value in human gliomas and seem to be associated with reduced apoptotic capacity of these tumors.

Adult↗
PubMed2002

Efficacy of trimethoprim-sulfamethoxazole for the prevention of bacterial infections in a randomized prophylaxis trial of patients with advanced HIV infection.

We compared the occurrences of several types of infections in HIV-infected patients participating in a randomized clinical trial of three treatment strategies given for the primary prevention of Pneumocystis carinii pneumonia (PCP) and toxoplasmosis. In a phase III open label trial, 842 patients with HIV infection and fewer than 200 CD4+ cells/mm(3) received zidovudine (standard dose) plus one of three randomly assigned prophylactic agents: trimethoprim-sulfamethoxazole (TMP-SMZ), or dapsone (DAP), or aerosolized pentamidine (AP). Patients developing intolerance to treatment were crossed over to another predefined prophylactic therapy. Patients were monitored for infections every other week for 8 weeks and then monthly until the study was completed. Primary statistical models were proportional hazards models adapted to recurrent end points. In an intent-to-treat analysis, compared with AP and DAP, TMP-SMZ significantly reduced the risk of any bacterial infection (combining all distinct types) (p = 0.02 and p = 0.01, respectively). When considering distinct types separately, compared with AP, TMP-SMZ significantly reduced the risk of infectious diarrhea (p = 0.04); compared with DAP, AP and TMP-SMZ significantly reduced the risk of sinusitis/otitis media (p = 0.03 and p = 0.04, respectively); compared with AP and DAP, TMP-SMZ significantly reduced the risk of a second occurrence of pneumonia (p = 0.04 and 0.02, respectively). For any bacterial infection, infection rates per 100 patient-years of follow-up were 31, 39, and 38 for TMP-SMZ, DAP, and AP, respectively. In patients with advanced HIV infection not taking highly active antiretroviral therapy, the treatment strategy that initiates prophylaxis with TMP-SMZ is superior to those initiating with AP or DAP for preventing any bacterial infection, with most of the advantage manifested through infectious diarrhea, sinusitis/otitis media, and pneumonia.

AIDS-Related Opportunistic Infections↗
PubMed2002

Model-based predictions of BRCA1/2 mutation status in breast carcinoma patients treated at an academic medical center.

BACKGROUND: Women with an existing breast carcinoma diagnosis who are found to carry a BRCA1/2 mutation have a substantial risk of developing both a contralateral breast carcinoma and ovarian carcinoma. In a newly diagnosed breast carcinoma patient, this genetic information may influence the management of her disease. To assess the volume of patients who may need genetic services at the time of diagnosis, the authors determined the proportion of women with newly diagnosed breast carcinoma at the study institution who would be eligible for genetic testing. METHODS: Fifty consecutive women with new breast carcinoma who were attending a multidisciplinary clinic were interviewed. Detailed, three-generation pedigrees were collected for each patient by a genetic counselor. Three commonly used probability models were used to calculate each woman's predicted risk of harboring a germline BRCA1/2 mutation. RESULTS: Eleven of 50 patients (22% [95% confidence interval, 12-36%]) were calculated to have a > or = 10% probability of carrying a BRCA1/2 mutation by at least one mathematic model and should have been offered genetic counseling that included the discussion of genetic testing. There were considerable discrepancies between probability calculations among the three mathematic models. One of the 11 patients who was eligible for genetic testing pursued genetic counseling within 12 months of diagnosis. CONCLUSIONS: At a large academic medical center, a substantial proportion of unselected women attending a multidisciplinary clinic were found to have a > or = 10% risk of carrying a BRCA1/2 mutation. The actual number of patients eligible to receive BRCA1/2 genetic testing outweighs the number of patients seen for genetic counseling at the study institution. Finally, limited correlation was found between current predictive models.

Academic Medical Centers↗
PubMed2002

Study of either ifosfamide or teniposide compared to a standard chemotherapy for extensive disease small cell lung cancer: an Eastern Cooperative Oncology Group randomized study (E1588).

This randomized study of previously untreated patients with extensive disease small cell lung cancer was designed (a) to compare the survival of patients treated with either effective standard chemotherapy or an investigational anti-cancer drug as initial therapy and (b) to evaluate response rates and toxic effects of such therapies. One hundred and thirty-five patients were randomly assigned to receive as initial therapy, either the standard CAV regimen--cyclophosphamide (1000 mg/m(2)), doxorubicin (50 mg/m(2)) and vincristine (1.4 mg/m(2)) every 3 weeks--or the phase II drugs ifosfamide (1.5 gm/m(2)/days 1-5) with mesna (300 mg/m(2)) dose at 0, 4 and 8 h after IV daily ifosfamide every 3 weeks or teniposide (60 mg/m(2)/days 1-5) every 3 weeks. Nonresponders received salvage chemotherapy-etoposide (120 mg/m(2) on days 1, 2 and 3) and cisplatin (60 mg/m(2) on day 1), repeated every 3 weeks. Among the 46 patients on CAV, there were two complete and 24 partial responses (56%). Among the 43 patients on ifosfamide, there were three complete and 18 partial responses (49%), while among the 46 patients on teniposide, there were two complete and 18 partial responses (43%). Eighty-three of the patients proceeded onto salvage regimen, of which 81 were analyzable for response and toxicity. Among the 81 patients who continued on salvage therapy and were evaluable for response, the overall best response rate was 61% for CAV+salvage, 54% for ifosfamide+salvage, and 53% for teniposide+salvage. These rates were not significantly different (P=0.962). Of the 135 analyzable patients, 130 (96%) have died. The estimated median survival time was 42 weeks for CAV patients, 43 weeks for ifosfamide, and 38 weeks for teniposide. Seven patients survived longer than 2 years (four on CAV, one on ifosfamide and two on teniposide). There were 29 life-threatening complications to the induction regimen (22 (48%) on CAV, four (9%) on ifosfamide and three (7%) on teniposide) and seven lethal complications (two on CAV, four on ifosfamide and one on teniposide). The treatments were significantly different with respect to the overall degree of toxicity (P < 0.0001) with CAV being more toxic. The data of this study, like the previous ECOG study suggests that the administration of a new agent followed by effective salvage chemotherapy in the treatment of extensive disease small cell lung cancer may have no adverse effect on survival.

Adult↗
PubMed2002

Phase II study of paclitaxel and carboplatin for malignant melanoma.

Conventional chemotherapy in the treatment of malignant melanoma has been disappointingly ineffective. Although the most active single agent is dacarbazine, numerous trials support the activity of platinum analogs against melanoma. Several recent trials also demonstrate the single-agent activity of paclitaxel, and support the administration of these agents together in the treatment of a number of solid tumors. We conducted a phase II study treating patients with metastatic melanoma with paclitaxel as a 3-hour infusion of 175 mg/m2, and carboplatin dosed to yield an area under the curve of 7.5 administered during 30 minutes. Patients were previously untreated or treated once with a regimen not including a platinum or taxane agent and had a performance status of 0 or 1. Seventeen patients (8 males; 9 females; average age 51) were enrolled. Thirty-three percent of patients had disease involving the skin, 40% lungs, 33% lymph nodes, 26% liver, and 13% other visceral organs. Anaphylactic reactions developed in two patients associated with paclitaxel infusion, and the patients were subsequently taken off the study. Of the 15 remaining patients, 3 (20%) had partial responses, 7 (47%) had stable disease, and 5 (33%) showed evidence of progressive disease. Eleven patients experienced grade III or IV hematologic toxicity. All treatment-related toxicities were reversible. There were no treatment-related deaths. The combination of paclitaxel and carboplatin has moderate activity against malignant melanoma, with expected reversible hematologic toxicities. Although not prospectively compared with single agents, the combination of paclitaxel and carboplatin may be a treatment option for some patients. Comparisons with other treatment regimens and the search for additional active compounds against melanoma are needed.

Antineoplastic Combined Chemotherapy Protocols↗
PubMed2002

Testing for dependence between failure time and visit compliance with interval-censored data.

Interval-censored failure-time data arise when subjects miss prescheduled visits at which the failure is to be assessed. The resulting intervals in which the failure is known to have occurred are overlapping. Most approaches to the analysis of these data assume that the visit-compliance process is ignorable with respect to likelihood analysis of the failure-time distribution. While this assumption offers considerable simplification, it is not always plausible. Here we test for dependence between the failure- and visit-compliance processes, applicable to studies in which data collection continues after the occurrence of the failure. We do not make any of the assumptions made by previous authors about the joint distribution of the visit-compliance process, a covariate process, and the failure time. Instead, we consider conditional models of the true failure history given the current visit compliance at each visit time, allowing for correlation across visit times. Because failure status is not known at some visit times due to missed visits, only models of the observed failure history given current visit compliance are estimable. We describe how the parameters from these models can be used to test for a negative association and how bounds on unestimable parameters provided by the observed data are needed additionally to infer a positive association. We illustrate the method with data from an AIDS study and we investigate the power of the test through a simulation study.

AIDS-Related Opportunistic Infections↗