Time-dependent analysis of the impact on early cytomegalovirus reactivation of HLA mismatch and acute graft-versus-host disease after allogeneic hematopoietic cell transplantation from related donors in acquired aplastic anemia

Cytomegalovirus (CMV) reactivation is an important issue in allogeneic hematopoietic cell transplantation (HCT). The incidence of early CMV reactivation is notably high in HLA-mismatched HCT. However, the interactions between HLA mismatch and acute graft-versus-host disease (aGvHD), a time-dependent event, make it methodologically challenging to evaluate the independent impact on CMV reactivation of the two variables. We retrospectively analyzed 355 patients with acquired aplastic anemia who received related donor transplants using a unified antithymocyte globulin-based platform. Patients were divided into group 1 (6/6 HLA match), group 2 (1-2/6 HLA allelic mismatch), and group 3 (3/6 HLA allelic mismatch). The impact of covariates was analyzed through two models: (1) time-dependent Cox and (2) dynamic landmarking analysis. The time-dependent Cox model showed that the HLA mismatch of 3/6 alleles (hazard ratio (HR) =1.852, P = .004) and aGvHD (HR = 1.009, P = .019) were independent risk factors for CMV reactivation. With the dynamic landmarking analysis, a higher HLA disparity correlated to increased early CMV reactivation (HR = 1.606, P = .001) at all time points. Developing aGvHD following HCT was generally associated with a higher incidence of CMV reactivation (HR = 1.623, P = .013), though its impact decreased with successive later landmark time points. In conclusion, our data suggest that the higher HLA disparity and aGvHD increases susceptibility to early CMV reactivation. In particular, the dynamic landmarking analysis demonstrated the time-varying effect of aGvHD on CMV reactivation, and HLA mismatch showed a profound impact over time following HCT.

Types of second primary cancer influence overall survival in cutaneous melanoma

BACKGROUND

Favorable survival in malignant cutaneous melanoma (melanoma) has increased the likelihood of second primary cancer (SPC). We assess the influence of patient characteristics at diagnosis of first melanoma and the type of SPC (second melanoma and other SPC) on overall survival.

METHODS

We used the Swedish Cancer Registry data to assess overall survival in melanoma for the period 1990 to 2015. Kaplan-Meier curves were plotted and hazard ratios (HRs) were estimated with Cox regression models by considering SPC diagnosis as a time-dependent variable.

RESULTS

A total of 46,726 patients were diagnosed with melanoma, and 15.3% of them developed SPC, among which, two thirds were other SPCs. Second melanomas were diagnosed early (31% during the first year) compared to non-melanoma SPCs (9.5%). Survival for women with second melanoma or other SPC (56 and 21% alive after 25 years of follow-up, respectively) exceeded the male rates (21 and 10%, respectively) but all these figures were lower than for females (60% alive) or males (48%) without SPC. Time dependent analysis showed vastly increased HRs for cancer types that are fatal also as first cancers, but SPC-specific HRs remained relatively uniform, irrespective of SPC diagnosed soon or late after first melanoma. In early-onset melanoma, SPC diagnosis after 10 years may not negatively influence overall survival.

CONCLUSIONS

As the overall survival of patients with many types of SPCs is unfavorable, advice about health lifestyle should benefit smoking patients and early detection methods may be recommended for SPCs of the breast, prostate and colorectum.

Optimizing the timing of diagnostic testing after positive findings in lung cancer screening: a proof of concept radiomics study

BACKGROUND

The timeliness of diagnostic testing after positive screening remains suboptimal because of limited evidence and methodology, leading to delayed diagnosis of lung cancer and over-examination. We propose a radiomics approach to assist with planning of the diagnostic testing interval in lung cancer screening.

METHODS

From an institute-based lung cancer screening cohort, we retrospectively selected 92 patients with pulmonary nodules with diameters ≥ 3 mm at baseline (61 confirmed as lung cancer by histopathology; 31 confirmed cancer-free). Four groups of region-of-interest-based radiomic features (n = 310) were extracted for quantitative characterization of the nodules, and eight features were proven to be predictive of cancer diagnosis, noise-robust, phenotype-related, and non-redundant. A radiomics biomarker was then built with the random survival forest method. The patients with nodules were divided into low-, middle- and high-risk subgroups by two biomarker cutoffs that optimized time-dependent sensitivity and specificity for decisions about diagnostic workup within 3 months and about repeat screening after 12 months, respectively. A radiomics-based follow-up schedule was then proposed. Its performance was visually assessed with a time-to-diagnosis plot and benchmarked against lung RADS and four other guideline protocols.

RESULTS

The radiomics biomarker had a high time-dependent area under the curve value (95% CI) for predicting lung cancer diagnosis within 12 months; training: 0.928 (0.844, 0.972), test: 0.888 (0.766, 0.975); the performance was robust in extensive cross-validations. The time-to-diagnosis distributions differed significantly between the three patient subgroups, p < 0.001: 96.2% of high-risk patients (n = 26) were diagnosed within 10 months after baseline screen, whereas 95.8% of low-risk patients (n = 24) remained cancer-free by the end of the study. Compared with the five existing protocols, the proposed follow-up schedule performed best at securing timely lung cancer diagnosis (delayed diagnosis rate: < 5%) and at sparing patients with cancer-free nodules from unnecessary repeat screenings and examinations (false recommendation rate: 0%).

CONCLUSIONS

Timely management of screening-detected pulmonary nodules can be substantially improved with a radiomics approach. This proof-of-concept study's results should be further validated in large programs.

Assessing the effects of time-dependent restrictions and control actions to flatten the curve of COVID-19 in Kazakhstan

This article presents the assessment of time-dependent national-level restrictions and control actions and their effects in fighting the COVID-19 pandemic. By analysing the transmission dynamics during the first wave of COVID-19 in the country, the effectiveness of the various levels of control actions taken to flatten the curve can be better quantified and understood. This in turn can help the relevant authorities to better plan for and control the subsequent waves of the pandemic. To achieve this, a deterministic population model for the pandemic is firstly developed to take into consideration the time-dependent characteristics of the model parameters, especially on the ever-evolving value of the reproduction number, which is one of the critical measures used to describe the transmission dynamics of this pandemic. The reproduction number alongside other key parameters of the model can then be estimated by fitting the model to real-world data using numerical optimisation techniques or by inducing ad-hoc control actions as recorded in the news platforms. In this article, the model is verified using a case study based on the data from the first wave of COVID-19 in the Republic of Kazakhstan. The model is fitted to provide estimates for two settings in simulations; time-invariant and time-varying (with bounded constraints) parameters. Finally, some forecasts are made using four scenarios with time-dependent control measures so as to determine which would reflect on the actual situations better.

Application of time-dependent modeling for the exposure-efficacy analysis of ceritinib in untreated ALK-rearranged advanced NSCLC patients

PURPOSE

Ceritinib 750 mg/day was approved for the treatment of patients with untreated anaplastic lymphoma kinase (ALK)-rearranged non-small cell lung cancer (NSCLC) based on ASCEND-4 study. The objective of this article is to introduce the use of time-dependent modeling approach in the updated exposure-efficacy analysis of ceritinib for the first-line indication.

METHODS

Exposure-efficacy analyses, including data from 156 patients, were first conducted using time-independent logistic regression model for response of complete or partial response and Cox regression model for progression-free survival (PFS). The exposure measure used was average C_trough, which is defined as the geometric mean of all evaluable C_trough for each patient. To further investigate the impact of exposure measure on exposure-efficacy analyses, a time-dependent modeling approach was used, where exposure at different time intervals was associated with the corresponding response endpoints in a longitudinal manner.

RESULTS

With exposure measure being average C_trough, it was observed that higher exposure was associated with reduced efficacy in terms of response (odds ratio = 0.77) and PFS [hazard ratio (HR) = 1.12]. These time-independent models do not account for the impact of time-varying concentration due to dose modifications. Subsequently, a new time-dependent modeling approach was used, where exposure and efficacy were associated longitudinally in the analyses. The results showed that the odds ratio of response became 1.07, and the HR of PFS became 1.04, indicating no apparent reverse relationship between exposure and efficacy across the exposure range studied.

CONCLUSION

The drug effect on efficacy in clinical trials could be better characterized using time-dependent exposure-response models.

The association of sleep duration and quality with all-cause and cause-specific mortality in the Women's Health Initiative

BACKGROUND/OBJECTIVE

Many studies have shown a U-shaped association of sleep duration with mortality; however, this association is difficult to interpret owing to possible reverse causation, residual confounding, and measurement issues. We used data from the Women's Health Initiative to examine the associations of sleep duration, insomnia, and use of sleep aids with death from cardiovascular disease (CVD), cancer, "other" causes, and all causes combined.

METHODS

Cox proportional hazards models were used in the analysis of baseline data and in time-dependent analyses of repeated measures to estimate associations of sleep-related factors with mortality. Among 158,203 women with information regarding sleep, 30,400 total deaths, 8857 CVD deaths, 9284 cancer deaths, and 11,928 other deaths were ascertained over a median of 17.8 years.

RESULTS

In both baseline and time-dependent analyses, both short (≤5 h) and long sleep (≥9 h) durations were associated with increased risk of total, CVD, and "other" deaths, but not with cancer deaths. Insomnia showed no association with mortality, whereas use of sleep medications was associated with an increased mortality risk.

CONCLUSIONS

While our findings showed a small but robust association of sleep duration with mortality in postmenopausal women, studies including objective measurements of sleep quality and efficiency are needed to clarify these associations.

Optimal adjuvant endocrine treatment of ER+/HER2+ breast cancer patients by age at diagnosis: A population-based cohort study

BACKGROUND

Prior randomised controlled trials on adjuvant hormonal therapy included HER2_any patients; however, a differential effect of aromatase inhibitors (AIs) versus tamoxifen (TAM) may have been missed in ER+/HER2+ patients that comprise 7-15% of all breast cancer patients. In addition, a woman's hormonal microenvironment may influence sensitivity to TAM and AIs in the adjuvant setting, which changes during menopausal transition, a process that takes years. We studied the efficacy of AIs versus TAM in ER+/HER2+ breast cancer patients grouped by age at diagnosis as a proxy for menopausal status using treatment and outcome data from the nationwide population-based Netherlands Cancer Registry (NCR).

PATIENTS AND METHODS

All women diagnosed between 2005 and 2007 with endocrine-treated, T_anyN_anyM₀, ER+/HER2+ breast cancer were identified through the NCR (n = 1155). Patients were divided by age at diagnosis: premenopausal (≤45 years; n = 326), perimenopausal (45<years≤55; n = 304) and postmenopausal (>55 years; n = 525). A time-dependent variable, indicating whether AI or TAM was received for >50% of endocrine treatment duration, was applied to subdivide groups by predominant treatment received. Recurrence-free survival (RFS) and overall survival (OS) were assessed using Kaplan-Meier survival estimation and Cox regression. Hazard ratios (HRs) were adjusted for chemotherapy, trastuzumab, age at diagnosis, N-status, grade, pT-stage and ovarian ablation.

RESULTS

During follow-up, 237 recurrences and 182 deaths occurred. Perimenopausal women derived significant RFS and OS benefit from AI compared with TAM, HR 0.47 (95% CI 0.25-0.91; P = 0.03) and HR 0.37 (95% CI 0.18-0.79; P = 0.01), respectively, whereas premenopausal women derived no benefit from AI compared with TAM. Treatment effects differed significantly between these age groups (interaction P = 0.03 and P = 0.02, respectively). Among postmenopausal women a small but non-significant AI benefit was observed.

CONCLUSION

AI treatment, preferably without any TAM treatment, was associated with the best RFS and OS outcome in ER+/HER2+ perimenopausal breast cancer patients.

Longitudinally collected CTCs and CTC-clusters and clinical outcomes of metastatic breast cancer. Breast Cancer Res Treat. 2017;161:83-94. [PMID: 27771841 DOI: 10.1007/s10549-016-4026-2]

PURPOSE

Circulating tumor cell (CTC) is a well-established prognosis predictor for metastatic breast cancer (MBC), and CTC-cluster exhibits significantly higher metastasis-promoting capability than individual CTCs. Because measurement of CTCs and CTC-clusters at a single time point may underestimate their prognostic values, we aimed to analyze longitudinally collected CTCs and CTC-clusters in MBC prognostication.

METHODS

CTCs and CTC-clusters were enumerated in 370 longitudinally collected blood samples from 128 MBC patients. The associations between baseline, first follow-up, and longitudinal enumerations of CTCs and CTC-clusters with patient progression-free survival (PFS) and overall survival (OS) were analyzed using Cox proportional hazards models.

RESULTS

CTC and CTC-cluster counts at both baseline and first follow-up were significantly associated with patient PFS and OS. Time-dependent analysis of longitudinally collected samples confirmed the significantly unfavorable PFS and OS in patients with ≥5 CTCs, and further demonstrated the independent prognostic values by CTC-clusters compared to CTC-enumeration alone. Longitudinal analyses also identified a link between the size of CTC-clusters and patient OS: compared to the patients without any CTC, those with 2-cell CTC-clusters and ≥3-cell CTC-clusters had a hazard ratio (HR) of 7.96 [95 % confidence level (CI) 2.00-31.61, P = 0.003] and 14.50 (3.98-52.80, P < 0.001), respectively.

CONCLUSIONS

In this novel time-dependent analysis of longitudinally collected CTCs and CTC-clusters, we showed that CTC-clusters added additional prognostic values to CTC enumeration alone, and a larger-size CTC-cluster conferred a higher risk of death in MBC patients.

Transient analysis of the Erlang A model

We consider the Erlang A model, or [Formula: see text] queue, with Poisson arrivals, exponential service times, and m parallel servers, and the property that waiting customers abandon the queue after an exponential time. The queue length process is in this case a birth-death process, for which we obtain explicit expressions for the Laplace transforms of the time-dependent distribution and the first passage time. These two transient characteristics were generally presumed to be intractable. Solving for the Laplace transforms involves using Green's functions and contour integrals related to hypergeometric functions. Our results are specialized to the [Formula: see text] queue, the M / M / m queue, and the M / M / m / m loss model. We also obtain some corresponding results for diffusion approximations to these models.

β-Blocker use and all-cause mortality of melanoma patients: results from a population-based Dutch cohort study

BACKGROUND

Results from preclinical and observational studies suggest that β-adrenoreceptor inhibition might influence disease progression of melanoma.

PATIENTS AND METHODS

Patients ⩾18years with cutaneous melanoma (Breslow thickness >1mm) registered in the Eindhoven Cancer Registry between January 1, 1998 and December 31, 2010, who were also registered with PHARMO record linkage system (RLS), were eligible. Randomly selected patients using β-blockers from PHARMO record linkage system (RLS) matched on age and gender served as a control cohort. Adjusted time-dependent and time-fixed Cox proportional hazard models were employed to estimate the hazard ratio of all-cause mortality. Five-year relative survival rates for all-cause mortality were calculated to estimate disease specific survival.

RESULTS

203 of 709 eligible patients used β-blockers after melanoma diagnosis. The use of β-blockers was not associated with the risk of dying (adjusted hazard ratio (HR) 0.82, 95% confidence interval (CI) 0.55-1.24). Neither duration of exposure nor β-blocker dosage showed significant influence on survival. Five-year relative survival for β-blocker users was lower than in non-users amongst melanoma patients (80.9% and 83.7%, respectively) but higher among the β-blocker control group compared to the general population (101.4%).

CONCLUSION

Our results do not show a statistically significant impact of β-blocker exposure on overall survival of melanoma patients, regardless of the timing, duration or dosage of β-blocker use.