Competing Risks Methods Research Articles

Late mortality after autologous blood or marrow transplantation in childhood: a Blood or Marrow Transplant Survivor Study-2 report

AbstractAutologous blood or marrow transplantation (BMT) is a curative option for several types of childhood cancer. However, there is little information regarding the risk of late mortality. We examined all-cause mortality, relapse-related mortality (RRM), and nonrelapse-related mortality (NRM) in 2-year survivors of autologous BMT performed before age 22 between 1980 and 2010 at 1 of 2 US transplant centers. Vital status information was collected using medical records, National Death Index, and Accurint databases. Overall survival was calculated using Kaplan-Meier techniques. Cumulative incidence of mortality used competing risk methods. Standardized mortality ratio (SMR) was calculated using age-, sex-, and calendar-specific mortality rates from Centers for Disease Control and Prevention. Cox regression analysis was used to determine predictors of all-cause late mortality. Among the 345 2-year survivors, 103 deaths were observed, yielding an overall survival of 70.3% 15 years post-BMT. The leading causes of death included primary disease (50.0%), subsequent neoplasm (21.4%), and infection (18.2%). Overall, the cohort was at a 22-fold increased risk of late mortality (SMR, 21.8; 95% CI, 17.9-26.3), compared with the general population. Mortality rates remained elevated among the 10-year survivors (SMR, 20.6; 95% CI, 9.9-37.2) but approached those of the general population ≥15 years post-BMT. The 10-year cumulative incidence of RRM (14.3%) exceeded that of NRM (10.4%). The 10-year cumulative mortality rate declined over time (<1990, 35.1%; 1990-1999, 25.6%; 2000-2010, 21.8%; P = .05). In conclusion, childhood autologous BMT recipients have an increased risk of late mortality, compared with the general population. The late mortality rates have declined over the past 3 decades.

Prostate sparing cystectomy for bladder cancer: A two-center study

PurposeTo assess long-term functional and oncologic outcomes of prostate sparing cystectomy (PSC) as a sexuality-preserving alternative to radical cystectomy in a selected group of bladder cancer (BC) patients. Materials and methodsBetween 1995 and 2014, 185 BC patients underwent PSC according to one of two standardized procedures at two centers. All patients had received extensive evaluation to rule out prostate cancer and BC at the bladder neck and prostatic urethra (PU), including prostate specific antigen blood analysis, transrectal ultrasound and/or prostate biopsies, PU biopsies and/or PU frozen section analysis. All patients received an orthotopic ileal neobladder. Overall survival (OS) was assessed by Kaplan–Meier estimates. Cumulative incidence of cancer specific mortality, any recurrence and loco-regional recurrence were calculated using competing-risk methods. Finally, functional outcomes (voiding, continence and erectile function) were evaluated. Results185 patients (cTa-3N0M0) with a mean age of 57 years (SD: 9) were included. Median follow-up was 7.5 years (IQR: 5.6–10.8). Five-year OS was 71% and 5-year cumulative incidence of recurrence was 31%. Twenty patients (10.8%) had a loco-regional recurrence, two recurrences were in the PU. During follow-up, prostate cancer was detected in six patients (3.2%). Erectile function was preserved in 86.1% of patients, complete daytime and nighttime continence in 95.6% and 70.2%, respectively. ConclusionThis two-center study shows that in men with BC in whom the prostate and PU were proven free of malignancy, PSC would represent a valid treatment option with excellent functional outcome. Oncologic outcomes were comparable to what is known from radical cystoprostatectomy series.

Location of Metastases in Contemporary Prostate Cancer Patients Affects Cancer-Specific Mortality.

To quantify the prognostic effect of the location of prostate cancer (PCa) metastases on cancer-specific mortality (CSM) and the rate of other-cause mortality (OCM) in contemporary newly diagnosed metastatic PCa (mPCa) patients. Within the Surveillance Epidemiology and End Results database (2004-2014), we focused on newly diagnosed mPCa patients. Data were stratified according to the site of metastases and age group. Cumulative incidence smoothed plots were generated for CSM and OCM at 5 years after diagnosis, according to the competing-risks methods. Multivariable competing-risks analyses tested the effect of the location of PCa metastases on CSM. Among 18,404 patients with mPCa, the majority had exclusively bone metastases (63.6%). At 5 years, CSM rate was 59.7% and OCM rate was 14%. According to the location of metastases, CSM rates were 44.5%, 57.9%, 67.1%, 62.7%, 66%, and 76.3% for exclusively lymph node (LN), exclusively bone, bone plus LN, exclusively visceral, visceral plus LN, and visceral plus LN and bone disease, respectively. In multivariable competing-risks models, PCa-specific mortality rate was 1.58-, 1.79-, 1.91-, 2.10-, and 2.47-fold higher in patients with exclusively bone, bone plus LN, exclusively visceral, visceral plus LN, and visceral plus bone and LN involvement compared to those with exclusively LN metastases (all P< .001). Patients with concomitant visceral, bone, and LN metastases have the worst prognosis. Similarly, when either bone or visceral metastases coexist with concomitant LN metastases, CSM rates are higher than when no concomitant LN metastases are present.

Prevalence, prognosis, and treatment implications of retropharyngeal nodes in unknown primary head and neck carcinoma

Objective(1) To estimate the prevalence of radiographically positive Retro-Pharyngeal Lymph Nodes (RPLN) in unknown primary carcinoma of the head and neck and (2) to determine the prognostic implications of radiographically positive RPLN and other radiographic features (3) to identify patients at low risk for retropharyngeal metastasis. Materials and methodsThe medical records of all 68 eligible patients treated at the Princess Margaret Cancer Centre between 2000 and 2014 were retrospectively reviewed for demographic, clinical, pathologic, and radiologic data. Radiologic data included: RPLN, extra capsular spread (ECS), neck staging and cystic/necrotic or matted neck nodes. LRR, DR, DFS and OS were estimated using the competing risk methods and the Kaplan-Meier method. ResultsSeven patients had concerning RPLN (10.3%). Forty-four patients were p16 positive (65%). RPLN status did not have any effect on LRR, DFS, DR and OS. Radiological ECS and p16 (neg.) status were found to be significant predictors of LRR (p = 0.023; p = 0.014). Matted nodes, radiological ECS and p16 (neg.) status were found to be significant predictors of DFS (p = 0.012; p < 0.001; p = 0.014). Matted nodes and radiological ECS were found to be significant predictors of OS (p = 0.017; p = 0.0036). Only radiological ECS was found to be a significant predictor of distant recurrence (p = 0.0066). Conclusions10% of CUP patients will harbor radiological positive RPLN. A large proportion of CUP patients are positive for p16. Radiologic features such as ECS and matted nodes can predict worse outcomes.

The natural history of large renal masses followed on observation

PurposeThe safety and feasibility of active surveillance in comorbid patients with renal masses ≥4.0cm is uncertain. The aim of this study is to describe our institutional experience with the observation of large renal masses. Materials and methodsOne hundred patients were identified with renal masses ≥ 4.0cm that were followed on observation for at least 6 months without surgical intervention between 1994 and 2016. Linear regression was conducted to determine predictors for renal mass growth and competing risk methods were used to estimate the probability of progression in the setting of death from other causes. ResultsMedian age at diagnosis was 73 years and 73% of patients had a Charlson Comorbidity index ≥ 4. At presentation, the median mass size was 4.9cm. The median growth rate was 0.4cm/y and there were no significant predictors of growth. Surveillance was discontinued in 34 patients who underwent delayed intervention. Median follow up for metastasis-free survivors was 4 years. In total, 10 patients developed metastatic disease, 3 died from kidney cancer and 30 patients died from other causes. The 5-year probability of other cause mortality was 22% (95% CI: 14%–32%) compared to 6% (95% CI: 2%–13%) for metastatic progression of kidney cancer. ConclusionIn highly comorbid patients, the observation of large renal masses has low likelihood for metastatic progression relative to the risk of nonkidney cancer related death. This data supports the use of surveillance as an acceptable strategy for highly selected patients with competing risks from other serious illnesses.

Population-based study of mortality and major amputation following lower limb revascularization.

The aim of this study was to estimate separate risks of major lower limb amputation and death following revascularization for peripheral artery disease (PAD) using competing risks analysis. Routinely collected data from Hospital Episode Statistics (HES) were used to identify patients who underwent endovascular or open lower limb revascularization for PAD in England from 2005 to 2015. The primary outcomes were major lower limb amputation and death within 5 years of revascularization. Cox proportional hazards and Fine-Gray competing risks regression were used to examine the competing risks of these outcomes. Some 164 845 patients underwent their first lower limb revascularization for PAD during the study interval. Most were men (64·6 per cent) and the median age was 71 (i.q.r. 62-78) years. Following endovascular revascularization, the 5-year cumulative incidence of amputation was 4·2 per cent in patients with intermittent claudication and 18·0 per cent in those with a record of tissue loss. The corresponding rates were 10·8 and 25·3 per cent respectively after open revascularization, and 8·1 and 25·0 per cent after combined procedures. The 5-year cumulative incidence of death varied from 24·5 to 39·8 per cent, depending on procedure type. Competing risks methods consistently produced lower estimates than standard methods. The 5-year risk of major amputation following lower limb revascularization for PAD appears lower than estimated previously. Patients undergoing revascularization for tissue loss and those who require an open procedure are at highest risk of limb loss.

Dose-response in choroidal melanoma

PurposeIn choroidal melanoma the radiation threshold dose for local control remains largely unknown. The present study examined a group of patients that received a wide range of minimum tumor dose in order to investigate a dose-response relationship. A literature review is performed to compare our results with available evidence in brachytherapy and charged particle external beam radiotherapy. Materials and methodsA retrospective study was conducted on all choroidal melanomas treated with Strontium-90 (Sr-90) at the University Hospital of Leuven between 1983 and 2012. Local failure was defined as primary endpoint and was estimated according to the competing risk method. ResultsIn 135 patients, the minimum tumor dose (Dmin) ranged from 0 Gy to 287 Gy (median: 27.6 Gy). Multivariable analysis revealed Dmin ≥ 65 Gy (p = 0.04; HR = 0.09) and tumor distant from the optic disc (p < 0.001, HR = 0.09) to be independent variables favoring local control. The scleral dose, the tumor diameter and tumor height did not significantly affect local failure in multivariate analysis. ConclusionThis is the first study to examine a group of patients treated with a Dmin ranging from 0 Gy to >250 Gy. Treatment with a Dmin of 65 Gy is necessary to achieve durable tumor response. The dose-response data provided by our study could be used for the design of future trials examining the ideal dose for the treatment of choroidal melanoma with brachytherapy or charged particle external beam radiotherapy.

Analysis of central nervous system efficacy in the J-ALEX study of alectinib versus crizotinib in ALK-positive non-small-cell lung cancer

ObjectivesWe determined the central nervous system (CNS) efficacy of alectinib by calculating time to CNS progression and cumulative incidence rates (CIRs) of CNS progression, non-CNS progression and death in patients with anaplastic lymphoma kinase (ALK)-positive non-small-cell lung cancer (NSCLC) enrolled in the J-ALEX phase III study. Materials and methodsJapanese patients aged ≥20 years with ALK-positive NSCLC who were ALK inhibitor-naïve and chemotherapy-naïve, or who had received one previous chemotherapy regimen, were enrolled. Patients with treated or untreated asymptomatic CNS metastases were eligible. Treatment comprised oral alectinib 300 mg twice daily or crizotinib 250 mg twice daily until progressive disease, unacceptable toxicity, death or withdrawal. Imaging scans (computed tomography/magnetic resonance imaging) were taken at baseline and at regular intervals throughout the study. The CIRs for CNS progression, non-CNS progression and death were calculated for patients with and without baseline CNS metastases using a competing risks method. ResultsThe hazard ratio for time to CNS progression in patients with and without baseline CNS metastases was 0.51 (95% confidence interval [CI]: 0.16–1.64; P = 0.2502) and 0.19 (95% CI: 0.07–0.53; P = 0.0004), respectively. The CIRs of CNS progression and non-CNS progression were lower in the alectinib group than in the crizotinib group at all time points. The 1-year CIRs of CNS progression were 16.8% and 5.9% with crizotinib and alectinib, respectively, and the 1-year CIRs of non-CNS progression were 38.4% and 17.5%, respectively. Comparable findings were obtained in patients with or without baseline CNS metastases. ConclusionAlectinib appears to avert the progression of CNS metastases in patients with ALK-positive NSCLC and baseline CNS metastases, and to prevent the development of new CNS lesions in patients without baseline CNS disease.

Long-term outcomes in oral cavity squamous cell carcinoma with adjuvant and salvage radiotherapy after surgery.

Early-stage oral cavity squamous cell carcinoma (OCSCC) represents a heterogeneous group of patients, and locoregional recurrence rates are as high as 25% with surgery alone. Radiotherapy (RT) is typically reserved as part of salvage multimodality therapy after salvage surgery because it is generally thought that there is no significant detriment to salvage therapy. Our aim was to examine outcomes for recurrent OCSCC treated with salvage surgery and radiation and compare them to outcomes for patients treated with adjuvant postoperative RT upfront. We identified 425 patients with OCSCC treated with postoperative RT at our institution. The 5-year rates of local failure, locoregional failure (LRF), survival, and distant metastasis (DM) were the main outcome measures. We performed a landmark analysis and examined the same outcomes in the adjuvant versus salvage cohorts using Cox proportional hazards and Fine-Gray competing risk method. The adjuvant cohort had higher tumor (T) (P < 0.0001) and nodal (N) (P < 0.0001) stage than the salvage cohort's stage at initial presentation. On multivariate analysis, a strategy of salvage RT experienced poorer overall survival (OS) compared to upfront adjuvant RT (hazard ratio [HR] 1.84; 95% confidence interval [CI], 1.26-2.70; P = 0.002). Moreover, salvage surgery followed by RT patients experienced increasing risk of LRF (HR 1.56; 95% CI, 1.18-2.06; P = 0.002) and (DM) (HR 1.53; 95% CI, 1.08-2.17; P = 0.02) on multivariate analysis. Additional analysis was performed excluding salvage cohort with advanced disease at initial presentation (T3-T4 and N2). Salvage RT treatment selection for early-stage OCSCC continued to experience significantly poorer OS as compared to adjuvant RT (HR 1.48; 95% CI, 1.002-2.19; P = 0.049). Early-stage OCSCC patients who are observed and experienced recurrence requiring salvage therapy (surgery and RT) have worse oncologic outcomes than locally advanced patients receiving upfront adjuvant RT. Prospective randomized studies are needed to identify high-risk subset of early-stage OCSCC comparing adjuvant RT versus observation, followed by salvage surgery and RT at recurrence. 4. Laryngoscope, 2539-2545, 2018.

Loss to follow-up and mortality among HIV-infected adolescents receiving antiretroviral therapy in Pune, India.

India has the highest number of HIV-infected adolescents in Asia, but little is known about their treatment outcomes. We assessed rates and factors associated with loss to follow-up (LTFU) and mortality among Indian adolescents. The analysis included adolescents (10-19 years old) starting antiretroviral therapy (ART) between 2005 and 2014 at BJ Government Medical College, Pune, India. LTFU was defined as missing more than three consecutive monthly visits. The competing-risks method was used to calculate subdistribution hazard ratios (SHRs) of predictors for LTFU, with death as the competing risk. Cox proportional hazard models were used to identify predictors of mortality. Of 717 adolescents starting ART, 402 with complete data were included in the analysis. Of these, 61% were male and 80% were perinatally infected, and the median baseline CD4 count was 174 cells/μL. LTFU and mortality rates were 4.4 and 4.9/100-person years, respectively. Cumulative LTFU incidence increased from 6% to 15% over 6 years. Age ≥ 15 years [adjusted SHR (aSHR) 2.44; 95% confidence interval (CI) 1.18-5.02] was a risk factor for LTFU. Cumulative mortality increased from 9.5% to 17.9% over 6 years. World Health Organization (WHO) stages III and IV [adjusted hazard ratio (aHR) 2.26; 95% CI: 1.14-4.48] and an increase in CD4 count by 100 cells/μL (aHR: 0.59; 95% CI: 0.43-0.83) were associated with mortality. A third of adolescents had been lost to follow-up or died by follow-up year 6. Older age was a risk factor for LTFU and advanced clinical disease for death. Strategies to improve retention counselling for older adolescents and closer clinical monitoring of all adolescents must be considered.

Missingness in the Setting of Competing Risks: from missing values to missing potential outcomes.

The setting of competing risks in which there is an event that precludes the event of interest from occurring is prevalent in epidemiological research. Unless studying all-cause mortality, any study following up individuals is subject to having a competing risk should individuals die during time period that the study covers. While there are prior papers discussing the need for competing risk methods in epidemiologic research, we are not aware of any review that discusses issues of missing data in a competing risk setting. We provide an overview of causal inference in competing risks as potential outcomes are missing, provide some strategies in dealing with missing (or misclassified) event type, and missing covariate data in competing risks. The strategies presented are specifically focused on those that may easily be implemented in standard statistical packages. There is ongoing work in terms of causal analyses, dealing with missing event type information, and missing covariate values specific to competing risk analyses. Competing events are common in epidemiologic research. While there has been a focus on why one should conduct a proper competing risk analysis, a perhaps unrecognized issue is in terms of missingness. Strategies exist to minimize the impact of missingness in analyses of competing risks.

A group sequential test for treatment effect based on the Fine-Gray model.

Competing risks endpoints arise when patients can fail therapy from several causes. Analyzing these outcomes allows one to assess directly the benefit of treatment on a primary cause of failure in a clinical trial setting. Regression models can be used in clinical trials to adjust for residual imbalances in patient characteristics, improving the power to detect treatment differences. But, none of the competing risks methods currently available for use in group sequential trials adjust for covariates. We propose a group sequential test for treatment effect that, because it is based on the Fine-Gray model, permits adjustment for covariates. Our derivations show that its sequence of test statistics has an asymptotic distribution with an independent increments structure, which allows standard techniques such as O'Brien-Fleming designs and error spending functions to be employed to meet type I error rate and power specifications. We demonstrate the test in a reanalysis of BMT CTN 0402, a phase III clinical trial that evaluated an experimental treatment for the prevention of adverse outcomes following blood and marrow transplant. Moreover, using a simulation study of randomized group sequential trials, we demonstrate that the proposed method preserves the type I error rate and power at their nominal levels in the presence of influential covariates.

Are competing risks models appropriate to describe implant failure?

Background and purpose — The use of competing risks models is widely advocated in the arthroplasty literature due to a perceived bias in comparison of simple Kaplan–Meier estimates. Proponents of competing risk models in the arthroplasty literature appear to be unaware of the subtle but important differences in interpretation of net and crude failure estimated by competing risk and Kaplan–Meier methods respectively.Methods — Using a simple simulation we illustrate the differences between competing risks and Kaplan–Meier methods.Results — Competing risk and Kaplan–Meier methods estimate different survival quantities, i.e., crude and net failure respectively. Estimates of crude failure estimated using competing risk methods will be less than net failure as estimated using Kaplan–Meier methods.Interpretation — Kaplan–Meier methods are appropriate for describing implant failure, whereas crude survival estimated using competing risk methods estimates the risk of surgical revision as it depends on both implant failure and mortality. Both competing risk models and Kaplan–Meier methods are useful in arthroplasty, and both provide unbiased estimates of crude and net failure in the absence of any confounding or selection respectively. Surgeons and researchers should carefully consider whether the use of competing risks is always justified. Lower estimates of failure from competing risk models may be misleading to surgeons who are attempting to select the best implants with the lowest failure rates for their patients.

Competing Risk of Death With End-Stage Renal Disease in Diabetic Kidney Disease.

The concept of competing risks is particularly relevant to survival analyses of diabetic ESRD given the high likelihood of death prior to ESRD. Approaches such as Kaplan-Meier curves and Cox regression models operate on the assumption that there are no competing risks for the event of interest, yielding uninterpretable and generally biased estimates in the presence of competing risks. The cumulative incidence function and Fine-Gray regression are more appropriate methodologies for survival analysis when competing risks are present. We present an example taken from the Action to Control Cardiovascular Risk in Diabetes, a randomized trial of people with type 2 diabetes at high risk for cardiovascular disease. Participants were stratified according to baseline markers of kidney disease: (1) no kidney disease; (2) low estimated glomerular filtration rate; (3) microalbuminuria alone; and (4) macroalbuminuria. The macroalbuminuria group had the highest risk for ESRD and demonstrated the most marked difference between the Kaplan-Meier and cumulative incidence estimator. Cox and Fine-Gray regression models yielded similar risk estimates for baseline characteristics, with the exception of diabetes duration, which was significant in the Cox but not Fine-Gray model. We underscore the importance of competing risk methods, particularly when the competing risk is common, as is the case in diabetic kidney disease.

Decline in Late Mortality after Allogeneic Blood or Marrow Transplantation (alloBMT) Performed in Childhood: a Report from the BMT Survivor Study-2 (BMTSS-2)

Background: AlloBMT is a curative option for children with malignant and non-malignant diseases. Nonetheless, the high intensity of therapeutic exposures at a young age increases the risk of organ compromise and may lead to premature death. However, there remains a gap in knowledge regarding cause-specific late mortality experienced by children undergoing alloBMT. Furthermore, it is not clear whether mortality rates have changed over the past 4 decades. We address these gaps by examining premature death (all-cause, relapse-related mortality [RRM] and non-relapse-related mortality [NRM]) in patients undergoing alloBMT between 1974 and 2010 at City of Hope, University of Minnesota or University of Alabama at Birmingham. Methods: To be included in this analysis, the patients had to have received alloBMT before age 22, and survived for ≥2y after transplantation. Vital status information was ascertained as of Dec 31, 2016, using medical records, the National Death Index (NDI), and Accurint databases. Information on cause of death was obtained from the NDI Plus program and medical records. Using Kaplan-Meier techniques, we describe overall survival conditional on surviving 2+ and 20+y from alloBMT. We describe cumulative incidence of RRM and NRM using competing risk methods. We also describe all-cause late mortality for patients transplanted over 3 time periods: <1990; 1990-1999; 2000-2010. Standardized mortality ratio (SMR) was used to compare the mortality experienced by this cohort to the age- sex-, and calendar-specific mortality of the US general population. Proportional subdistribution hazards model (Fine-Gray) for competing risks was used for identifying predictors of late mortality, including demographics, primary disease, conditioning agents, disease status at alloBMT and GvHD prophylaxis. Results: In this cohort of 1,411 recipients of alloBMT performed in childhood and surviving ≥2y, the most common primary diagnoses were acute lymphoblastic leukemia (ALL: 25%), acute myeloid leukemia/myelodysplastic syndrome (AML/MDS: 24%), inborn errors of metabolism (IEM: 14%) and severe aplastic anemia (SAA: 11%). The distribution of the cohort over the 3 time periods was: <1990 (23%); 1990-1999 (29%); 2000-2010 (48%). After a median follow-up of 14.9y, we observed 295 deaths in this 2y survivor cohort, yielding an overall survival rate of 79.6% at 20y from alloBMT. The leading causes of premature death were infection (32.0%), primary disease (24.6%) and subsequent malignant neoplasms (18.4%). Overall, the cohort was at a 15-fold increased risk for premature death (SMR 15.3, 95% CI 13.6-17.2), when compared with the general population. Conditional on surviving 20y, the 5y overall survival was 96.0% for the entire cohort, and 98.4% for SAA, 96.0% for AML/MDS, 90.3% for ALL and 85.8% for IEM. Relative mortality declined with time from alloBMT, but remained significantly elevated at 25y after transplantation (SMR 1.9, 95% CI 1.1-2.9). The cumulative incidence of NRM exceeded that due to RRM (fig 1); the 20y cumulative incidence of NRM and RRM were 13.0% and 4.4%, respectively. The all-cause 10y cumulative mortality rate declined over the 3 eras (<1990: 18.9%, 1990-1999: 12.5%; >2000: 10.7%, p=0.001) (Fig 2). Adjusting for demographic and clinical variables, as well as conditioning, the hazard ratio (HR) of all-cause death decreased with more recent treatment eras (<1990: HR=1.0; 1990-1999: HR=0.7, 95%CI, 0.5-0.9, p=0.01; 2000-2010: HR=0.6, 95%CI, 0.4-0.8, p=0.004). Conclusions: This study demonstrates that compared with the general population, 2y survivors of alloBMT performed in childhood remain at an elevated risk of premature death, even after 25y from alloBMT. While the risk of relapse-related mortality plateaus with time, non-relapse-related mortality continues to increase and remains the major cause of late death in this population. However, we make an encouraging observation i.e., there has been a significant decline in all-cause mortality experienced by children undergoing alloBMT over the past 4 decades. [Display omitted] DisclosuresArora:Takeda Oncology: Consultancy.

Volume, Dose, and Fractionation Considerations for IMRT-based Reirradiation in Head and Neck Cancer: A Multi-institution Analysis

Limited data exist to guide the treatment technique for reirradiation of recurrent or second primary squamous carcinoma of the head and neck. We performed a multi-institution retrospective cohort study to investigate the effect of the elective treatment volume, dose, and fractionation on outcomes and toxicity. Patients with recurrent or second primary squamous carcinoma originating in a previously irradiated field (≥40Gy) who had undergone reirradiation with intensity modulated radiation therapy (IMRT); (≥40Gy re-IMRT) were included. The effect of elective nodal treatment, dose, and fractionation on overall survival (OS), locoregional control, and acute and late toxicity were assessed. The Kaplan-Meier and Gray's competing risks methods were used for actuarial endpoints. From 8 institutions, 505 patients were included in the present updated analysis. The elective neck was not treated in 56.4% of patients. The median dose of re-IMRT was 60Gy (range 39.6-79.2). Hyperfractionation was used in 20.2%. Systemic therapy was integrated for 77.4% of patients. Elective nodal radiation therapy did not appear to decrease the risk of locoregional failure (LRF) or improve the OS rate. Doses of ≥66Gy were associated with improvements in both LRF and OS in the definitive re-IMRT setting. However, dose did not obviously affect LRF or OS in the postoperative re-IMRT setting. Hyperfractionation was not associated with improved LRF or OS. The rate of acute grade ≥3 toxicity was 22.1% overall. On multivariable logistic regression, elective neck irradiation was associated with increased acute toxicity in the postoperative setting. The rate of overall late grade ≥3 toxicity was 16.7%, with patients treated postoperatively with hyperfractionation experiencing the highest rates. Doses of ≥66Gy might be associated with improved outcomes in high-performance patients undergoing definitive re-IMRT. Postoperatively, doses of 50 to 66Gy appear adequate after removal of gross disease. Hyperfractionation and elective neck irradiation were not associated with an obvious benefit and might increase toxicity.

Contemporary disengagement from antiretroviral therapy in Khayelitsha, South Africa: A cohort study.

BackgroundRetention in care is an essential component of meeting the UNAIDS “90-90-90” HIV treatment targets. In Khayelitsha township (population ~500,000) in Cape Town, South Africa, more than 50,000 patients have received antiretroviral therapy (ART) since the inception of this public-sector program in 2001. Disengagement from care remains an important challenge. We sought to determine the incidence of and risk factors associated with disengagement from care during 2013–2014 and outcomes for those who disengaged.Methods and findingsWe conducted a retrospective cohort study of all patients ≥10 years of age who visited 1 of the 13 Khayelitsha ART clinics from 2013–2014 regardless of the date they initiated ART. We described the cumulative incidence of first disengagement (>180 days not attending clinic) between 1 January 2013 and 31 December 2014 using competing risks methods, enabling us to estimate disengagement incidence up to 10 years after ART initiation. We also described risk factors for disengagement based on a Cox proportional hazards model, using multiple imputation for missing data. We ascertained outcomes (death, return to care, hospital admission, other hospital contact, alive but not in care, no information) after disengagement until 30 June 2015 using province-wide health databases and the National Death Registry. Of 39,884 patients meeting our eligibility criteria, the median time on ART to 31 December 2014 was 33.6 months (IQR 12.4–63.2). Of the total study cohort, 592 (1.5%) died in the study period, 1,231 (3.1%) formally transferred out, 987 (2.5%) were silent transfers and visited another Western Cape province clinic within 180 days, 9,005 (22.6%) disengaged, and 28,069 (70.4%) remained in care. Cumulative incidence of disengagement from care was estimated to be 25.1% by 2 years and 50.3% by 5 years on ART. Key factors associated with disengagement (age, male sex, pregnancy at ART start [HR 1.58, 95% CI 1.47–1.69], most recent CD4 count) and retention (ART club membership, baseline CD4) after adjustment were similar to those found in previous studies; however, notably, the higher hazard of disengagement soon after starting ART was no longer present after adjusting for these risk factors. Of the 9,005 who disengaged, the 2 most common initial outcomes were return to ART care after 180 days (33%; n = 2,976) and being alive but not in care in the Western Cape (25%; n = 2,255). After disengagement, a total of 1,459 (16%) patients were hospitalized and 237 (3%) died. The median follow-up from date of disengagement to 30 June 2015 was 16.7 months (IQR 11–22.4). As we included only patient follow-up from 2013–2014 by design in order to maximize the generalizability of our findings to current programs, this limited our ability to more fully describe temporal trends in first disengagement.ConclusionsTwenty-three percent of ART patients in the large cohort of Khayelitsha, one of the oldest public-sector ART programs in South Africa, disengaged from care at least once in a contemporary 2-year period. Fifty-eight percent of these patients either subsequently returned to care (some “silently”) or remained alive without hospitalization, suggesting that many who are considered “lost” actually return to care, and that misclassification of “lost” patients is likely common in similar urban populations.A challenge to meeting ART retention targets is developing, testing, and implementing program designs to target mobile populations and retain them in lifelong care. This should be guided by risk factors for disengagement and improving interlinkage of routine information systems to better support patient care across complex care platforms.

Clinical Outcomes for Frameless Image-Guided Stereotactic Radiation Therapy to Intact Brain Metastases in Five Daily Hypofractionated Treatments

Image-guided multi-lead collimator (MLC) linac-based hypofractionated stereotactic radiation (FSRT) is increasingly practiced for brain metastases. We report our outcomes for 5-fraction FSRT including the rate of radiation necrosis. 113 patients with 184 brain metastases who were treated with FSRT from Oct 2009-Dec 2014 were retrospectively reviewed. Overall Survival (OS) and progression free survival (PFS) rates were calculated using the Kaplan-Meier method. Local control rates were determined with Fine and Gray’s competing risk method with death as the competing risk. Predictive factors were identified using Cox regression multivariable analysis. All patients were treated in a thermoplastic mask using a cone-beam CT image-guided linac equipped with a MLC and six-degrees-of-freedom treatment couch. All patients were treated with a 5 daily fraction dose scheme to the gross tumor volume (GTV) with a 2 mm planning target volume (PTV). 30 Gy in 5 fractions was the most common prescription. The median follow-up was 11 months. Non-small cell lung cancer the most frequently treated histology (46%) followed by breast cancer (21%) and melanoma (9%). FSRT was given for a single brain metastasis in 67 (59%) patients and for multiple metastases in 46 patients (41%). 77 patients (68%) were treated as upfront and 36 (32%) as salvage following whole brain radiation. 50 patients (44%) had treated lesions >2 cm in maximum diameter. 43 targets (23%) received 25-27.5Gy, 111 (60%) received 30 Gy, and 30 (16%) received >30 Gy in 5 fractions. The PFS rates at 6 months, 12 months and 24 months were 65%, 38%, and 15%, respectively (median, 9.0 months). The OS rates at 6 months, 12 months and 24 months were 69% 47% and 20%, respectively (median OS, 11.8 months). The local failure rates at 6 months, 12 months and 24 months were 8.2%, 22% and 35%, respectively (median, 11.43 months). With regards to other patterns of failure, cumulative incidence rates of leptomeningeal disease at 6 months, 12 months and 24 months were 9.8%, 14% and 17% respectively (median, 4.4 months). 28 targets in 25 patients (15%) developed radiation necrosis, of which 17 were symptomatic (9%). Post-treatment necrosis was a positive prognostic factor (HR 2.1, p=0.003) for OS. Upfront therapy as opposed to salvage (HR 0.54, p=0.02) and those treated with doses > 30Gy (HR 0.55, p=0.03) were predictive for local control. Frameless image-guided MLC-based linac FSRT for intact brain metastasis results in long-term local control with an acceptable rate of RN.

Comparison of Single- and Three-Fraction Schedules of Stereotactic Body Radiation Therapy for Peripheral Early-Stage Non–small Cell Lung Cancer: A Single Institution Experience

Stereotactic Body Radiation Therapy (SBRT) is an alternative treatment option for medically inoperable patients with early-stage non-small cell lung cancer (NSCLC). The optimal dose and fractionation for NSCLC remain unclear. The aim of this study is to compare the clinical outcome of patients with early-stage NSCLC who underwent either single-fraction (SF) or three-fraction (TF) SBRT at a single institution over 8 years. Between February 2007 and November 2015, patients with peripherally located early-stage NSCLC who underwent SBRT were included in this study. Data were retrospectively reviewed and collected in an institutional review board-approved database. Overall survival and progression-free survival were analyzed using the Kaplan-Meier method and log-rank tests. Local, nodal, and distant failures were assessed using the competing risks method and Gray’s tests. Potential prognostic factors were examined using Cox regression. Toxicity was evaluated using the Common Terminology Criteria for Adverse Events (version 4.0). R software (version 3.3.2) was used for statistical analyses. Of 158 patients with 162 total lung tumors, 68 lesions received 30-34 Gy (median 30 Gy) in 1 fraction, while the remaining 94 lesions received 48-60 Gy (median 60 Gy) in 3 fractions. After a median follow-up of 17.1 months for SF-SBRT and 19.1 months for TF-SBRT groups (p=0.54), there was no statistically significant difference in overall survival (p=0.93), local failure (p=0.96), nodal failure (p=0.82), and distant failure (p=0.65) at 18 months. Among the incidences of nodal or distant failure (n=21 for SF-SBRT, n=28 for TF-SBRT), median time to failure was 11.5 months and 10.2 months for the SF- and TF-SBRT groups, respectively (p=0.45). No grade 3+ toxicity in the SF-SBRT group was reported, while only a single case of grade 3 pulmonary embolism in the TF-SBRT group was reported 10 months after the SBRT. A large tumor size was associated with poor overall survival (HR=1.22, p=0.045), and male gender was associated with a higher risk for nodal (HR=2.81, p=0.013) and distant failures (HR=2.00, p=0.036). The SF- and TF-SBRT groups showed no significant difference in overall survival, local failure, nodal failure, distant failure, and the onset of nodal or distant failure. Both SF-SBRT and TF-SBRT regimens were well tolerated. Small tumor size was a favorable prognostic factor for overall survival, whereas male gender was an adverse prognostic factor for nodal and distant failures. The similarity in clinical outcomes between these two schedules is notable since patients with worse prognosis or performance status tend to receive SF-SBRT.

Salvage Radiotherapy Leads to Worse Overall Survival Compared to Adjuvant Radiotherapy in Oral Cavity Carcinoma: A Matched Analysis

To compare oncologic outcomes between adjuvant radiotherapy and salvage radiotherapy in patients with oral cavity carcinoma (OCC). Between 12/1989 and 4/2011, we reviewed 425 patients with non-metastatic oral cavity cancer were treated with postoperative RT at our institution: 301 patients received adjuvant RT while 124 received salvage RT after tumor recurrence and salvage surgery. We estimated overall survival (OS) for the adjuvant cohort using the KM method for OS and cumulative incidence of LF, LRF, and DM using the Gray method. We subsequently performed a landmark analysis from 12 months post the initial surgery to reduce immortal time bias and compare the same outcomes in the adjuvant versus salvage cohort using the Cox proportional hazards regression model and Fine and Gray competing risks method. The adjuvant cohort had significantly higher T stage than the salvage cohort, T1: 23% vs 66%; T2: 34% vs 23%; T3: 9% vs 3%; T4: 34% vs 2% p<0.0001. The adjuvant cohort also had significantly more locally advanced N stage than the salvage cohort, N0: 33% vs 93%; N1: 25% vs 2%; N2: 41% vs 3%, N3: 1% vs 0%. The median follow up for survivors was 77 months from the landmark, the 5 year LF, LRF, DM, and OS rates in the adjuvant cohort were 21.8%/28.7%/19%,/57% respectively. On MVA, patients who received salvage RT experienced significantly poorer OS compared to those that received adjuvant RT after adjusting for age and T stage (HR 1.84, 95% CI 1.26-2.70, p=0.002). Salvage RT was also associated with significantly lower rates of LRC (HR 1.56, 95% CI 1.18-2.06, p=0.002) and DM (HR 1.53, 95% CI 1.08-2.17, p=0.02) on MVA. Salvage RT when adjusted for other factors results in significantly decreased LRC, higher rates of DM, and decreased OS as compared to adjuvant RT in OCC. Our study suggests that aggressive upfront loco-regional therapy represents the best opportunity to cure oral cavity carcinoma.