On the Importance of Accounting for Competing Risks in Pediatric Cancer Trials Designed to Delay or Avoid Radiotherapy: I. Basic Concepts and First Analyses

Cryopreserved versus non-cryopreserved stem cell autografts in multiple myeloma a restrospective cohort study

The use of non-cryopreserved hematopoietic stem cells (HSC) can be an alternative to the traditional cryopreserved infusions of HSCs in autologous stem cell transplantation (aHSCT). After high-dose melphalan conditioning (HDM), we sought to compare time to engraftment, overall survival, and safety in multiple myeloma (MM) patients undergoing a first aHSCT after high-dose melphalan conditioning (HDM). We conducted a cohort study from March 2018 to December 2019. Of all autologous transplants performed during this period, 105 were for MM as the first consolidation. Fifty-one patients received a cryopreserved graft; the remaining 54 patients received a fresh infusion. General clinical characteristics were similar between these two groups. Cell viability was higher in non-cryopreserved grafts (95% vs. 86% p < 0.01). Four deaths occurred during hospitalization in the cryopreserved group, one in the non-cryopreserved group. The cumulative incidence of neutrophil and platelet engraftment on D + 25 was higher in the non-cryopreserved compared to the cryopreserved group (98% vs 90% p < 0.01 and 96.2% vs 72.54% p < 0.01 respectively). Additionally, the hospital length of stay was reduced by 4 days for patients for the non-cryopreserved cohort. In summary, the use of non-cryopreserved HSCs after HDM is safe and effective compared to patients who received a cryopreserved graft.

Comparison of two treatments in the presence of competing risks

Competing risks data arise frequently in clinical trials, and a common problem encountered is the overall homogeneity between two groups. In competing risks analysis, when the proportional subdistribution hazard assumption is violated or two cumulative incidence function (CIF) curves cross; currently, the most commonly used testing methods, for example, the Gray test and the Pepe and Mori test, may lead to a significant loss of statistical testing power. In this article, we propose a testing method based on the area between the CIF curves (ABC). The ABC test captures the difference over the whole time interval for which survival information is available for both groups and is not based on any special assumptions regarding the underlying distributions. The ABC test was also extended to test short-term and long-term effects. We also consider a combined test and a two-stage procedure based on this new method, and a bootstrap resampling procedure is suggested in practice to approximate the limiting distribution of the combined test and two-stage test. An extensive series of Monte Carlo simulations is conducted to investigate the power and the type I error rate of the methods. In addition, based on our simulations, our proposed TS, Comb, and ABC tests have a relatively high power in most situations. In addition, the methods are illustrated using two different datasets with different CIF situations.

Risk factors for hospital and long-term mortality of critically ill elderly patients admitted to an intensive care unit

Background. Data on long-term outcomes of elderly (≥65 years) patients in ICU are sparse. Materials and Methods. Adult patients (n = 1563, 45.4% elderly) admitted over 28 months were analyzed by competing risks regression model to determine independent factors related to in-hospital and long-term mortality. Results. 414 (26.5%) and 337 (21.6%) patients died in-hospital and during the 52 months following discharge, respectively; the elderly group had higher mortality during both periods. After discharge, elderly patients had 2.3 times higher mortality compared to the general population of the same age-group. In-hospital mortality was independently associated with mechanical ventilation (subdistribution hazard ratio (SHR) 2.74), vasopressors (SHR 2.56), neurological disease (SHR 1.77), and Mortality Prediction Model II score (SHR 1.01) regardless of age and with malignancy (SHR, hematological 3.65, nonhematological 3.4) and prior renal replacement therapy (RRT, SHR 2.21) only in the elderly. Long-term mortality was associated with low hemoglobin concentration (SHR 0.94), airway disease (SHR 2.23), and malignancy (SHR hematological 1.11, nonhematological 2.31) regardless of age and with comorbidities especially among the nonelderly. Conclusions. Following discharge, elderly ICU patients have higher mortality compared to the nonelderly and general population. In the elderly group, prior RRT and malignancy contribute additionally to in-hospital mortality risk. In the long-term, comorbidities (age-related), anemia, airway disease, and malignancy were significantly associated with mortality.

The prognostic role of body mass index on mortality amongst the middle-aged and elderly: a competing risk analysis

AIMS

To determine the relationship between body mass index (BMI) including its 5-year changes and mortality, and compare the results obtained using Cox and competing risks models.

METHODS

Our study subjects included 2216 persons aged ≥49 years who participated in the Blue Mountains Eye Study, Australia between 1992 and 1994, and returned for further follow-up examinations between 1997 and 1999. We examined the relationship between BMI and mortality using cubic spline. The Cox and competing risks models were used to assess the associations between baseline BMI and its 5-year changes with all-cause and cause-specific mortality.

RESULTS

Amongst subjects aged ≤70 years, the relationship between BMI and all-cause mortality was U-shaped. For those aged >70 years, an L-shaped relationship was seen with no elevation in risk amongst the overweight/obese. Based on the competing risks model, obesity at baseline was associated with increased risk of cardiovascular death and reduction in BMI at 5-year was linked to an increase risk of cancer death amongst those aged ≤70 years. The cause-specific Cox model showed that reduction in BMI at 5-year was associated with cancer-death regardless of age, and with cardiovascular deaths among subjects aged ≤70 years. Cox regression model showed larger magnitude of effect with wider confidence interval as compared with competing risks model.

CONCLUSIONS

Conditions associated with obesity are more likely to affect mortality among subjects aged ≤70 years, but not among those aged over 70 years. Cox model shows larger magnitude of effect in comparison with competing risks model.

Empirical comparison of methods for analyzing multiple time-to-event outcomes in a non-inferiority trial: a breast cancer study

BACKGROUND

Subjects with breast cancer enrolled in trials may experience multiple events such as local recurrence, distant recurrence or death. These events are not independent; the occurrence of one may increase the risk of another, or prevent another from occurring. The most commonly used Cox proportional hazards (Cox-PH) model ignores the relationships between events, resulting in a potential impact on the treatment effect and conclusions. The use of statistical methods to analyze multiple time-to-event events has mainly been focused on superiority trials. However, their application to non-inferiority trials is limited. We evaluate four statistical methods for multiple time-to-event endpoints in the context of a non-inferiority trial.

METHODS

Three methods for analyzing multiple events data, namely, i) the competing risks (CR) model, ii) the marginal model, and iii) the frailty model were compared with the Cox-PH model using data from a previously-reported non-inferiority trial comparing hypofractionated radiotherapy with conventional radiotherapy for the prevention of local recurrence in patients with early stage breast cancer who had undergone breast conserving surgery. These methods were also compared using two simulated examples, scenario A where the hazards for distant recurrence and death were higher in the control group, and scenario B. where the hazards of distant recurrence and death were higher in the experimental group. Both scenarios were designed to have a non-inferiority margin of 1.50.

RESULTS

In the breast cancer trial, the methods produced primary outcome results similar to those using the Cox-PH model: namely, a local recurrence hazard ratio (HR) of 0.95 and a 95% confidence interval (CI) of 0.62 to 1.46. In Scenario A, non-inferiority was observed with the Cox-PH model (HR = 1.04; CI of 0.80 to 1.35), but not with the CR model (HR = 1.37; CI of 1.06 to 1.79), and the average marginal and frailty model showed a positive effect of the experimental treatment. The results in Scenario A contrasted with Scenario B with non-inferiority being observed with the CR model (HR = 1.10; CI of 0.87 to 1.39), but not with the Cox-PH model (HR = 1.46; CI of 1.15 to 1.85), and the marginal and frailty model showed a negative effect of the experimental treatment.

CONCLUSION

When subjects are at risk for multiple events in non-inferiority trials, researchers need to consider using the CR, marginal and frailty models in addition to the Cox-PH model in order to provide additional information in describing the disease process and to assess the robustness of the results. In the presence of competing risks, the Cox-PH model is appropriate for investigating the biologic effect of treatment, whereas the CR models yields the actual effect of treatment in the study.

Comparison of the dosimetries of 3-dimensions Radiotherapy (3D-RT) with linear accelerator and intensity modulated radiotherapy (IMRT) with helical tomotherapy in children irradiated for neuroblastoma

Background

Intensity modulated radiotherapy is an efficient radiotherapy technique to increase dose in target volumes and decrease irradiation dose in organs at risk. This last objective is mainly relevant in children. However, previous results suggested that IMRT could increase low dose, factor of risk for secondary radiation induced cancer. This study was performed to compare dose distributions with 3D-radiotherapy (3D-RT) and IMRT with tomotherapy (HT) in children with neuroblastoma. Seven children with neuroblastoma were irradiated. Treatment plans were calculated for 3D-RT, and for HT. For the volume of interest, the PTV-V_95% and conformity index were calculated. Dose constraints of all the organs at risk and integral dose were compared.

Results

The conformity index was statistically better for HT than for 3D-RT. PTV-V_95% constraint was reached in 6 cases with HT compared to 2 cases with 3D-RT. For the ipsilateral kidney of the tumor, the V_{12 Gy} constraint was reached for 3 patients with both methods. The values were lower with HT than with 3D-RT in two cases and higher in one case. The threshold was not reached for one patient with either technique, but the value was lower with HT than with 3D-RT. For the contralateral kidney of the tumors, the V_{12 Gy} constraint was reached for all patients with both methods. The values were lower with HT than with 3D-RT in 5 of 7 children, equal in one patient and higher in one patient. The organ-at-risk volumes receiving low doses were significantly lower with 3D-RT but larger for the highest doses, compared to those irradiated with HT. The integral doses were not different.

Conclusions

IMRT with HT allows a better conformity treatment, a more frequently acceptable PTV-V_95% than 3D-RT and, concomitantly, a better shielding of the kidneys. The integral doses are comparable between both techniques but consideration of differences in dose distribution between the two techniques, for the organs at risk, has to be taken in account when validating treatment.

Competing events in patients with malignant disease who are at risk for recurrent venous thromboembolism

Patients with malignant disease enrolled in trials of thrombotic disorders may experience competing events such as death. The occurrence of a competing event may prevent the thrombotic event from being observed. Standard survival analysis techniques ignore competing risks, resulting in possible bias and distorted inferences. To assess the impact of competing events on the results of a previously reported trial comparing low molecular weight heparin (LMWH) with oral anticoagulant (OAC) therapy for the prevention of recurrent venous thromboembolism (VTE) in patients with advanced cancer, we compare the results from standard survival analysis with those from competing risk techniques which are based on the cumulative incidence function (CIF) and Gray's test. The Kaplan-Meier method overestimates the risk of recurrent VTE (17.2% in the OAC group and 8.7% in the LMWH group). Risk of recurrence using the CIF is 12.0% and 6.0% in the OAC and LMWH groups, respectively. Both the log-rank test (p=0.002) and Gray's test (p=0.006) suggest evidence in favor of LMWH. The overestimation of risk is 30% in each treatment group, resulting in a similar relative treatment effect; using the Cox model the hazard ratio (HR) is 0.48 (95% confidence interval [CI], 0.30 to 0.78) and HR=0.47 (95% CI, 0.29 to 0.74) using the CIF model. Failing to account for competing risks may lead to incorrect interpretations of the probability of recurrent VTE. However, when the distribution of competing risks is similar within each treatment group, standard and competing risk methods yield comparable relative treatment effects.

Analysis and design of randomised clinical trials involving competing risks endpoints

BACKGROUND

In randomised clinical trials involving time-to-event outcomes, the failures concerned may be events of an entirely different nature and as such define a classical competing risks framework. In designing and analysing clinical trials involving such endpoints, it is important to account for the competing events, and evaluate how each contributes to the overall failure. An appropriate choice of statistical model is important for adequate determination of sample size.

METHODS

We describe how competing events may be summarised in such trials using cumulative incidence functions and Gray's test. The statistical modelling of competing events using proportional cause-specific and subdistribution hazard functions, and the corresponding procedures for sample size estimation are outlined. These are illustrated using data from a randomised clinical trial (SQNP01) of patients with advanced (non-metastatic) nasopharyngeal cancer.

RESULTS

In this trial, treatment has no effect on the competing event of loco-regional recurrence. Thus the effects of treatment on the hazard of distant metastasis were similar via both the cause-specific (unadjusted csHR = 0.43, 95% CI 0.25 - 0.72) and subdistribution (unadjusted subHR 0.43; 95% CI 0.25 - 0.76) hazard analyses, in favour of concurrent chemo-radiotherapy followed by adjuvant chemotherapy. Adjusting for nodal status and tumour size did not alter the results. The results of the logrank test (p = 0.002) comparing the cause-specific hazards and the Gray's test (p = 0.003) comparing the cumulative incidences also led to the same conclusion. However, the subdistribution hazard analysis requires many more subjects than the cause-specific hazard analysis to detect the same magnitude of effect.

CONCLUSIONS

The cause-specific hazard analysis is appropriate for analysing competing risks outcomes when treatment has no effect on the cause-specific hazard of the competing event. It requires fewer subjects than the subdistribution hazard analysis for a similar effect size. However, if the main and competing events are influenced in opposing directions by an intervention, a subdistribution hazard analysis may be warranted.

Primary postoperative chemotherapy without radiotherapy for treatment of brain tumours other than ependymoma in children under 3 years: results of the first UKCCSG/SIOP CNS 9204 trial

BACKGROUND

Radiotherapy is an effective adjuvant treatment for brain tumours arising in very young children, but it has the potential to damage the child's developing nervous system at a crucial time - with a resultant reduction in IQ leading to cognitive impairment, associated endocrinopathy and risk of second malignancy. We aimed to assess the role of a primary chemotherapy strategy in avoiding or delaying radiotherapy in children younger than 3 years with malignant brain tumours other than ependymoma, the results of which have already been published.

METHODS

Ninety-seven children were enrolled between March 1993 and July 2003 and, following diagnostic review, comprised: medulloblastoma (n=31), astrocytoma (26), choroid plexus carcinoma [CPC] (15), CNS PNET (11), atypical teratoid/rhabdoid tumours [AT/RT] (6) and ineligible (6). Following maximal surgical resection, chemotherapy was delivered every 14 d for 1 year or until disease progression. Radiotherapy was withheld in the absence of progression.

FINDINGS

Over all diagnostic groups the cumulative progression rate was 80.9% at 5 years while the corresponding need-for-radiotherapy rate for progression was 54.6%, but both rates varied by tumour type. There was no clear relationship between chemotherapy dose intensity and outcome. Patients with medulloblastoma presented as a high-risk group, 83.9% having residual disease and/or metastases at diagnosis. For these patients, outcome was related to histology. The 5-year OS for desmoplastic/nodular medulloblastoma was 52.9% (95% confidence interval (CI): 27.6-73.0) and for classic medulloblastoma 33.3% (CI: 4.6-67.6); the 5-year EFS were 35.3% (CI: 14.5-57.0) and 33.3% (CI: 4.6-67.6), respectively. All children with large cell or anaplastic variants of medulloblastoma died within 2 years of diagnosis. The 5-year EFS for non-brainstem high-grade gliomas [HGGs] was 13.0% (CI: 2.2-33.4) and the OS was 30.9% (CI: 11.5-52.8). For CPC the 5-year OS was 26.67% (CI: 8.3-49.6) without RT. This treatment strategy was less effective for AT/RT with 3-year OS of 16.7% (CI: 0.8-51.7) and CNS PNET with 1-year OS of 9.1% (CI: 0.5-33.3).

INTERPRETATION

The outcome for very young children with brain tumours is dictated by degree of surgical resection and histological tumour type and underlying biology as an indicator of treatment sensitivity. Overall, the median age at radiotherapy was 3 years and radiotherapy was avoided in 45% of patients. Desmoplastic/nodular sub-type of medulloblastoma has a better prognosis than classic histology, despite traditional adverse clinical features of metastatic disease and incomplete surgical resection. A subgroup with HGG and CPC are long-term survivors without RT. This study highlights the differing therapeutic challenges presented by the malignant brain tumours of early childhood, the importance of surgical approaches and the need to explore individualised brain sparing approaches to the range of malignant brain tumours that present in early childhood.