The Impact of Stroma Admixture on Molecular Subtypes and Prognostic Gene Signatures in Serous Ovarian Cancer.

BACKGROUND: Recent efforts to improve outcomes for high-grade serous ovarian cancer, a leading cause of cancer death in women, have focused on identifying molecular subtypes and prognostic gene signatures, but existing subtypes have poor cross-study robustness. We tested the contribution of cell admixture in published ovarian cancer molecular subtypes and prognostic gene signatures. METHODS: Gene signatures of tumor and stroma were developed using paired microdissected tissue from two independent studies. Stromal genes were investigated in two molecular subtype classifications and 61 published gene signatures. Prognostic performance of gene signatures of stromal admixture was evaluated in 2,527 ovarian tumors (16 studies). Computational simulations of increasing stromal cell proportion were performed by mixing gene-expression profiles of paired microdissected ovarian tumor and stroma. RESULTS: Recently described ovarian cancer molecular subtypes are strongly associated with the cell admixture. Tumors were classified as different molecular subtypes in simulations where the percentage of stromal cells increased. Stromal gene expression in bulk tumors was associated with overall survival (hazard ratio, 1.17; 95% confidence interval, 1.11-1.23), and in one data set, increased stroma was associated with anatomic sampling location. Five published prognostic gene signatures were no longer prognostic in a multivariate model that adjusted for stromal content. CONCLUSIONS: Cell admixture affects the interpretation and reproduction of ovarian cancer molecular subtypes and gene signatures derived from bulk tissue. Elucidating the role of stroma in the tumor microenvironment and in prognosis is important. IMPACT: Single-cell analyses may be required to refine the molecular subtypes of high-grade serous ovarian cancer.

Heart and Lung Transplants from HCV-Infected Donors to Uninfected Recipients.

BACKGROUND: Hearts and lungs from donors with hepatitis C viremia are typically not transplanted. The advent of direct-acting antiviral agents to treat hepatitis C virus (HCV) infection has raised the possibility of substantially increasing the donor organ pool by enabling the transplantation of hearts and lungs from HCV-infected donors into recipients who do not have HCV infection. METHODS: We conducted a trial involving transplantation of hearts and lungs from donors who had hepatitis C viremia, irrespective of HCV genotype, to adults without HCV infection. Sofosbuvir-velpatasvir, a pangenotypic direct-acting antiviral regimen, was preemptively administered to the organ recipients for 4 weeks, beginning within a few hours after transplantation, to block viral replication. The primary outcome was a composite of a sustained virologic response at 12 weeks after completion of antiviral therapy for HCV infection and graft survival at 6 months after transplantation. RESULTS: A total of 44 patients were enrolled: 36 received lung transplants and 8 received heart transplants. The median viral load in the HCV-infected donors was 890,000 IU per milliliter (interquartile range, 276,000 to 4.63 million). The HCV genotypes were genotype 1 (in 61% of the donors), genotype 2 (in 17%), genotype 3 (in 17%), and indeterminate (in 5%). A total of 42 of 44 recipients (95%) had a detectable hepatitis C viral load immediately after transplantation, with a median of 1800 IU per milliliter (interquartile range, 800 to 6180). Of the first 35 patients enrolled who had completed 6 months of follow-up, all 35 patients (100%; exact 95% confidence interval, 90 to 100) were alive and had excellent graft function and an undetectable hepatitis C viral load at 6 months after transplantation; the viral load became undetectable by approximately 2 weeks after transplantation, and it subsequently remained undetectable in all patients. No treatment-related serious adverse events were identified. More cases of acute cellular rejection for which treatment was indicated occurred in the HCV-infected lung-transplant recipients than in a cohort of patients who received lung transplants from donors who did not have HCV infection. This difference was not significant after adjustment for possible confounders. CONCLUSIONS: In patients without HCV infection who received a heart or lung transplant from donors with hepatitis C viremia, treatment with an antiviral regimen for 4 weeks, initiated within a few hours after transplantation, prevented the establishment of HCV infection. (Funded by the Mendez National Institute of Transplantation Foundation and others; DONATE HCV ClinicalTrials.gov number, NCT03086044.).

Bias-corrected estimates for logistic regression models for complex surveys with application to the United States' Nationwide Inpatient Sample.

For complex surveys with a binary outcome, logistic regression is widely used to model the outcome as a function of covariates. Complex survey sampling designs are typically stratified cluster samples, but consistent and asymptotically unbiased estimates of the logistic regression parameters can be obtained using weighted estimating equations (WEEs) under the naive assumption that subjects within a cluster are independent. Despite the relatively large samples typical of many complex surveys, with rare outcomes, many interaction terms, or analysis of subgroups, the logistic regression parameters estimates from WEE can be markedly biased, just as with independent samples. In this paper, we propose bias-corrected WEEs for complex survey data. The proposed method is motivated by a study of postoperative complications in laparoscopic cystectomy, using data from the 2009 United States' Nationwide Inpatient Sample complex survey of hospitals.

A latent variable transformation model approach for exploring dysphagia.

Multiple outcomes are often collected in applications where the quantity of interest cannot be measured directly or is difficult or expensive to measure. In a head and neck cancer study conducted at Dana-Farber Cancer Institute, the investigators wanted to determine the effect of clinical and treatment factors on unobservable dysphagia through collected multiple outcomes of mixed types. Latent variable models are commonly adopted in this setting. These models stipulate that multiple collected outcomes are conditionally independent given the latent factor. Mixed types of outcomes (e.g., continuous vs. ordinal) and censored outcomes present statistical challenges, however, as a natural analog of the multivariate normal distribution does not exist for mixed data. Recently, Lin et al. proposed a semiparametric latent variable transformation model for mixed outcome data; however, it may not readily accommodate event time outcomes where censoring is present. In this paper, we extend the work of Lin et al. by proposing both semiparametric and parametric latent variable models that allow for the estimation of the latent factor in the presence of measurable outcomes of mixed types, including censored outcomes. Both approaches allow for a direct estimate of the treatment (or other covariate) effect on the unobserved latent variable, greatly enhancing the interpretability of the models. The semiparametric approach has the added advantage of allowing the relationship between the measurable outcomes and latent variables to be unspecified, rendering more robust inference. The parametric and semiparametric models can also be used together, providing a comprehensive modeling strategy for complicated latent variable problems.

Evaluation of American Joint Committee on Cancer, International Union Against Cancer, and Brigham and Women's Hospital tumor staging for cutaneous squamous cell carcinoma.

PURPOSE: To compare American Joint Committee on Cancer (AJCC), International Union Against Cancer (UICC), and Brigham and Women's Hospital (BWH) tumor (T) staging systems for cutaneous squamous cell carcinoma and validate BWH staging against prior data. PATIENTS AND METHODS: Primary tumors diagnosed from 2000 to 2009 at BWH (n = 1,818) were analyzed. Poor outcomes (local recurrence [LR], nodal metastasis [NM], and disease-specific death [DSD]) were analyzed by T stage with regard to each staging system's distinctiveness (outcome differences between stages), homogeneity (outcome similarity within stages), and monotonicity (outcome worsening with increasing stage). RESULTS: AJCC and UICC T3 and T4 were indistinct with overlapping 95% CIs for 10-year cumulative incidences of poor outcomes, but all four BWH stages were distinct. AJCC and UICC high-stage tumors (T3/T4) were rare at 0.3% and 3% of the cohort, respectively. Most poor outcomes occurred in low stages (T1/T2; AJCC: 86% [95% CI, 77% to 91%]; UICC: 70% [61% to 79%]) resulting in heterogeneous outcomes in T1/T2. Conversely, in BWH staging, only 5% of tumors were high stage (T2b/T3), but they accounted for 60% (95% CI, 50% to 69%) of poor outcomes (70% of NMs and 83% of DSDs) indicating superior homogeneity and monotonicity as previously defined. Cumulative incidences of poor outcomes were low for BWH low-stage (T1/T2a) tumors (LR, 1.4% [95% CI, 1% to 2%]; NM, 0.6% [95% CI, 0% to 1%]; DSD, 0.2% [95% CI, 0% to 0.5%]) and higher for high-stage (T2b/T3) tumors (LR, 24% [95% CI, 16% to 34%]; NM, 24% [95% CI, 16% to 34%]; and DSD, 16% [95% CI, 10% to 25%], which validated an earlier study of an alternative staging system. CONCLUSION: BWH staging offers improved distinctiveness, homogeneity, and monotonicity over AJCC and UICC staging. Population-based validation is needed. BWH T2b/T3 tumors define a high-risk group requiring further study for optimal management.

Disclosure control using partially synthetic data for large-scale health surveys, with applications to CanCORS.

Statistical agencies have begun to partially synthesize public-use data for major surveys to protect the confidentiality of respondents' identities and sensitive attributes by replacing high disclosure risk and sensitive variables with multiple imputations. To date, there are few applications of synthetic data techniques to large-scale healthcare survey data. Here, we describe partial synthesis of survey data collected by the Cancer Care Outcomes Research and Surveillance (CanCORS) project, a comprehensive observational study of the experiences, treatments, and outcomes of patients with lung or colorectal cancer in the USA. We review inferential methods for partially synthetic data and discuss selection of high disclosure risk variables for synthesis, specification of imputation models, and identification disclosure risk assessment. We evaluate data utility by replicating published analyses and comparing results using original and synthetic data and discuss practical issues in preserving inferential conclusions. We found that important subgroup relationships must be included in the synthetic data imputation model, to preserve the data utility of the observed data for a given analysis procedure. We conclude that synthetic CanCORS data are suited best for preliminary data analyses purposes. These methods address the requirement to share data in clinical research without compromising confidentiality.

Representativeness of participants in the cancer care outcomes research and surveillance consortium relative to the surveillance, epidemiology, and end results program.

BACKGROUND: The research goals of the Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium are to determine how characteristics and beliefs of patients, providers, and health care organizations influence the treatments and outcomes of individuals with newly diagnosed lung and colorectal cancers. As CanCORS results will inform national policy, it is important to know how they generalize to the United States population with these cancers. RESEARCH DESIGN: This study assessed the representativeness of the CanCORS cohort of 10,547 patients with lung cancer (LC) or colorectal cancer (CRC) enrolled between 2003 and 2005. We compared characteristics (sex, race, age, and disease stage) with the Surveillance, Epidemiology, and End Results (SEER) population of 234,464 patients with new onset of these cancers during the CanCORS recruitment period. RESULTS: The CanCORS sample is well matched to the SEER Program for both cancers. In CanCORS, 41% LC/47% CRC were female versus 47% LC/49% CRC in SEER. African American, Hispanic, and Asian cases differed by no more than 5 percentage points between CanCORS and SEER. The SEER population is slightly older, with the percentage of patients older than 75 years 33.1% LC/37.3% CRC in SEER versus 26.9% LC/29.4% in CanCORS, and also has a slightly higher proportion of early stage patients. We also found that the CanCORS cohort was representative within specific SEER regions that map closely to CanCORS sites. CONCLUSIONS: This study demonstrates that the CanCORS Consortium was successful in enrolling a demographically representative sample within the CanCORS regions.

Patient perspectives on breast cancer treatment plan and summary documents in community oncology care: a pilot program.

BACKGROUND: Although the routine use of treatment plans and summaries (TPSs) has been recommended to improve the quality of cancer care, limited data exist about their impact on quality, including patient satisfaction and coordination of care. METHODS: Patients received TPSs as part of the American Society of Clinical Oncology Breast Cancer Registry (BCR) pilot program of 20 community oncology practices. Participants were surveyed 2 to 4 weeks after receiving a TPS to evaluate their perceptions of the document. Patients who were receiving chemotherapy received the TPS as separate plan and summary documents (at the start and the end of treatment) and could complete 2 surveys. Others received a single integrated TPS. Eligible survey participants had stage 0 through III breast cancer and were enrolled in the BCR. RESULTS: Of 292 consented patients, 174 (60%) completed at least 1 survey. Of 157 patients who recalled receiving a TPS, 148 (94%) believed that the documents improved patient-physician communication, and 128 (82%) believed that they improved communication between physicians; 113 (72%) said the documents increased their peace of mind, whereas 2 (1%) had less peace of mind. Of 152 patients (97%) who still had their documents, 147 (97%) said they were useful, and 94 (62%) had given or planned to give the documents to another physician. All 63 patients who were surveyed after receiving a summary recommended that practices continue to provide TPSs to patients. CONCLUSIONS: Participants in this study expressed high satisfaction with TPSs. Additional research is needed to study the broad-scale implementation of the BCR and to evaluate the impact of routine use of TPSs on the quality of care delivered.

Patients' experiences with care for lung cancer and colorectal cancer: findings from the Cancer Care Outcomes Research and Surveillance Consortium.

PURPOSE: To assess patients' experiences with cancer care, ratings of their quality of care, and correlates of these assessments. PATIENTS AND METHODS: For 4,093 patients with lung cancer and 3,685 patients with colorectal cancer in multiple US regions and health care delivery systems, we conducted telephone surveys of patients or their surrogates in English, Spanish, or Chinese at 4 to 7 months after diagnosis. The surveys assessed ratings of the overall quality of cancer care and experiences with three domains of interpersonal care (physician communication, nursing care, and coordination and responsiveness of care). RESULTS: English-speaking Asian/Pacific Islander patients and Chinese-speaking patients and those in worse health reported significantly worse adjusted experiences with all three domains of interpersonal care, whereas white, black, and Hispanic patients reported generally similar experiences with interpersonal care. The overall quality of cancer care was rated as excellent by 44.4% of patients with lung cancer and 53.0% of patients with colorectal cancer, and these ratings were most strongly correlated with positive experiences with coordination and responsiveness of care (Spearman rank coefficients of 0.49 and 0.42 for lung and colorectal cancer, respectively). After multivariate adjustment, excellent ratings were less common for each cancer among black patients, English-speaking Asian/Pacific Islander patients, Chinese-speaking patients, and patients reporting worse health status (all P ≤ .05). CONCLUSION: Patients' reports and ratings of care differed significantly by race, language, and health status. Efforts to improve patients' experiences with cancer care should focus on problems affecting Asian and Pacific Islander patients and those in worse health.

Adjuvant chemotherapy use and adverse events among older patients with stage III colon cancer.

CONTEXT: Randomized trials suggest adjuvant chemotherapy is effective for older patients with stage III colon cancer. However, older patients are less likely to receive this therapy than younger patients, perhaps because of concern about adverse effects. OBJECTIVE: To evaluate adjuvant chemotherapy use and outcomes for older patients with stage III colon cancer from well-defined population-based settings and health care systems. DESIGN: Observational study of adjuvant chemotherapy use and outcomes by age using Poisson regression to estimate the number of adverse events adjusted for demographic and clinical factors, including comorbid illness and specific elements of chemotherapy regimens documented with clinically detailed medical record reviews and patient and surrogate surveys. SETTING: Five geographically defined regions (Alabama, Iowa, Los Angeles County, northern California, and North Carolina), 5 integrated health care delivery systems, and 15 Veterans Affairs hospitals. PATIENTS: Six hundred seventy-five patients diagnosed with stage III colon cancer from 2003 through 2005 who underwent surgical resection and were followed up for as long as 15 months postdiagnosis. MAIN OUTCOME MEASURES: Chemotherapy regimen, dose, duration, and annualized mean number of adverse events stratified by age. RESULTS: Of 202 patients aged 75 years and older, 101 (50%) received adjuvant chemotherapy compared with 87% of 473 younger patients (difference, 37%; 95% confidence interval [CI], 30%-45%). Among patients who received adjuvant chemotherapy, 14 patients (14%) aged 75 years and older and 178 younger patients (44%) received a regimen containing oxaliplatin (difference, 30%; 95% CI, 21%-38%). Older patients were less likely to continue treatment, such that by 150 days, 99 patients (40%) aged 65 years and older and 68 younger patients (25%) had discontinued chemotherapy (difference, 15%; 95% CI, 7%-23%). Overall, 162 patients (24%) had at least 1 adverse clinical event, with more events among patients treated with vs without adjuvant chemotherapy (mean, 0.39 vs 0.16; difference, 0.23; 95% CI, 0.11-0.36; P < .001). Among patients receiving adjuvant chemotherapy, adjusted rates of late clinical adverse events were lower for patients 75 years and older (mean, 0.28) vs for younger patients (0.35 for ages 18-54 years, 0.52 for ages 55-64 years, and 0.45 for ages 65-74 years; P = .008 for any age effect). CONCLUSION: Among patients with stage III colon cancer who underwent surgical resection and received adjuvant chemotherapy, older patients in the community received less-toxic and shorter chemotherapy regimens, and those treated had fewer adverse events than younger patients.

Adverse events among the elderly receiving chemotherapy for advanced non-small-cell lung cancer.

PURPOSE: To describe chemotherapy use and adverse events (AEs) for advanced-stage, non-small-cell lung cancer (NSCLC) in community practice, including descriptions according to variation by age. METHODS: We interviewed patients with newly diagnosed, stages IIIB and IV NSCLC in the population-based cohort studied by the Cancer Care Outcomes Research and Surveillance Consortium, and we abstracted the patient medical records. AEs were medical events occurring during chemotherapy. Using logistic regression, we assessed the association between age and chemotherapy; with Poisson regression, we estimated event rate ratios and adjusted the analysis for age, sex, ethnicity, radiation therapy, stage, histology, and presence and grade of 27 comorbidities. RESULTS: Of 1,371 patients, 58% (95% CI, 55% to 61%) received chemotherapy and 35% (95% CI, 32% to 38%) had AEs. After adjustment, 72% (95% CI, 65% to 79%) of those younger than 55 years and 47% (95% CI, 42% to 52%) of those age 75 years and older received chemotherapy. Platinum-based therapies were less common in the older-age groups. Pretreatment medical event rates were 18.6% for patients younger than 55 years and were only 9.2% for those age 75 years and older (adjusted rate ratio, 0.49; 95% CI, 0.26 to 0.91). In contrast, older adults were more likely to have AEs during chemotherapy. The adjusted rate ratios compared with age younger than 55 years were 1.70 for 65- to 74-year-olds (95% CI, 1.19 to 2.43) and 1.34 for those age 75 years and older (95% CI, 0.90 to 2.00). CONCLUSION: Older patients who received chemotherapy had fewer pretherapy events than younger patients and were less likely to receive platinum-based regimens. Nevertheless, older patients had more adverse events during chemotherapy, independent of comorbidity. Potential implicit trade-offs between symptom management and treatment toxicity should be made explicit and additionally studied.

USING PROFILE LIKELIHOOD FOR SEMIPARAMETRIC MODEL SELECTION WITH APPLICATION TO PROPORTIONAL HAZARDS MIXED MODELS.

We consider selection of nested and non-nested semiparametric models. Using profile likelihood we can define both a likelihood ratio statistic and an Akaike information for models with nuisance parameters. Asymptotic quadratic expansion of the log profile likelihood allows derivation of the asymptotic null distribution of the likelihood ratio statistic including the boundary cases, as well as unbiased estimation of the Akaike information by an Akaike information criterion. Our work was motivated by the proportional hazards mixed effects model (PHMM), which incorporates general random effects of arbitrary covariates and includes the frailty model as a special case. The asymptotic properties of its parameter estimate has recently been established, which enables the quadratic expansion of the log profile likelihood. For computation of the (profile) likelihood under PHMM we apply three algorithms: Laplace approximation, reciprocal importance sampling, and bridge sampling. We compare the three algorithms under different data structures, and apply the methods to a multi-center lung cancer clinical trial.

Reducing bias in parameter estimates from stepwise regression in proportional hazards regression with right-censored data.

When variable selection with stepwise regression and model fitting are conducted on the same data set, competition for inclusion in the model induces a selection bias in coefficient estimators away from zero. In proportional hazards regression with right-censored data, selection bias inflates the absolute value of parameter estimate of selected parameters, while the omission of other variables may shrink coefficients toward zero. This paper explores the extent of the bias in parameter estimates from stepwise proportional hazards regression and proposes a bootstrap method, similar to those proposed by Miller (Subset Selection in Regression, 2nd edn. Chapman & Hall/CRC, 2002) for linear regression, to correct for selection bias. We also use bootstrap methods to estimate the standard error of the adjusted estimators. Simulation results show that substantial biases could be present in uncorrected stepwise estimators and, for binary covariates, could exceed 250% of the true parameter value. The simulations also show that the conditional mean of the proposed bootstrap bias-corrected parameter estimator, given that a variable is selected, is moved closer to the unconditional mean of the standard partial likelihood estimator in the chosen model, and to the population value of the parameter. We also explore the effect of the adjustment on estimates of log relative risk, given the values of the covariates in a selected model. The proposed method is illustrated with data sets in primary biliary cirrhosis and in multiple myeloma from the Eastern Cooperative Oncology Group.

Satisfaction of the uncertainty principle in cancer clinical trials: retrospective cohort analysis.

OBJECTIVE: To assess whether publicly funded adult cancer trials satisfy the uncertainty principle, which states that physicians should enroll a patient in a trial only if they are substantially uncertain which of the treatments in the trial is most appropriate for the patient. This principle is violated if trials systematically favour either the experimental or the standard treatment. DESIGN: Retrospective cohort study of completed cancer trials, with randomisation as the unit of analysis. SETTING: Two cooperative research groups in the United States. STUDIES INCLUDED: 93 phase III randomised trials (103 randomisations) that completed recruitment of patients between 1981 and 1995. MAIN OUTCOME MEASURES: Whether the randomisation favoured the experimental treatment, the standard treatment, or neither treatment; effect size (outcome of the experimental treatment compared with outcome of the standard treatment) for each randomisation. RESULTS: Three randomisations (3%) favoured the standard treatment, 70 (68%) found no significant difference between treatments, and 30 (29%) favoured the experimental treatment. The average effect size was 1.20 (95% confidence interval 1.13 to 1.28), reflecting a slight advantage for the experimental treatment. CONCLUSIONS: In cooperative group trials in adults with cancer, there is a measurable average improvement in disease control associated with assignment to the experimental rather than the standard arm. However, the heterogeneity of outcomes and the small magnitude of the advantage suggest that, as a group, these trials satisfy the uncertainty principle.

Reporting of fertility events by men and women in rural Gambia.

We conducted a survey of male and female fertility in rural villages in The Gambia and compared men and women's reports of recent pregnancy events in the aggregate and of children ever born for matched couples. Despite widespread polygyny and sex differences in fertility, men's and women's reports were similar. Small sex differences in reports of recent stillbirths and neonatal deaths were found. For matched couples, husbands reported 0.23 more children ever born than their wives on average, but discordant reporting had little effect on recent marital fertility rates. Modeling of discordant reports indicates that fertility reports are more likely to be underestimated by both men and women for their earliest marriages. Reliable fertility data can be collected from men in this population.