Declarations of Independence: How Embedded Multicollinearity Errors Affect Dosimetric and Other Complex Analyses in Radiation Oncology

The statistical technique of multiple regression, commonly referred to as "multivariable regression," is often used in clinical research to quantify the relationships between multiple predictor variables and a single outcome variable of interest. The foundational theory underpinning multivariable regression assumes that all predictor variables are independent of one another. In other words, the effect of each independent variable is measured by its contribution to the regression equation while all other variables remain unchanged. In the presence of correlations between two or more variables, however, it is impossible to change one variable without a consequent change in the variable(s) it is linked to. This condition, known as "multicollinearity," can introduce errors into multivariable regression models by affecting estimates of the regression coefficients that quantify the relationship between individual predictor variables and the outcome variable. Errors that arise due to violations of the multicollinearity assumption are of special interest to radiation oncology researchers. Because of high levels of correlation among variables derived from points along individual organ dose-volume histogram (DVH) curves, as well as strong intercorrelations among dose-volume parameters in neighboring organs, dosimetric analyses are particularly subject to multicollinearity errors. For example, dose-volume parameters for the heart are strongly correlated not only with other points along the heart DVH curve but are likely also correlated with dose-volume parameters in neighboring organs such as the lung. In this paper, we describe the problem of multicollinearity in accessible terms and discuss examples of violations of the multicollinearity assumption within the radiation oncology literature. Finally, we provide recommendations regarding best practices for identifying and managing multicollinearity in complex data sets.

Influence of Radiation Fractionation on Immune Repertoire Diversity in Solid Tumor Patients

PURPOSE/OBJECTIVE(S)

Radiation (RT)-induced lymphopenia (RIL) occurs in up to 75% of patients undergoing RT and is associated with worse tumor control and survival across a spectrum of solid tumors. Patients undergoing hypofractionated RT are at lower risk of RIL compared with patients treated with more prolonged RT courses. However, it is unknown whether immune repertoire diversity is similarly affected by fractionation scheme in patients undergoing RT. This prospective study analyzed RT-induced changes in immune repertoire diversity in patients treated with conventionally (CFRT) vs hypofractionated RT (HFRT).

MATERIALS/METHODS

RNA-based T and B cell receptor sequencing was performed on peripheral lymphocytes collected prospectively before RT and within 4 weeks of the last RT fraction from 23 patients (18 men, 5 women, median age 67 y) with primary solid tumors undergoing CFRT (≤3 Gy/day x ≥10 days, n = 13) or HFRT (≥5 Gy/day x ≤5 days, n = 10). Absolute lymphocyte counts (ALC; cells/μL) were obtained from clinical laboratory data. The number of unique CDR3 receptors (uCDR3) and Shannon entropy were used to monitor changes in T (TCR Vβ) and B (BCR IgH) receptor diversity.

RESULTS

ALC decreased after RT in 90% (20/22) of patients (mean pre-RT ALC 1830 vs 1040 post-RT, p <0.001). Mean % ALC loss was greater in CFRT vs HFRT patients (44.3 vs. 35.2%). After RT, entropy in IgH and Vβ decreased in 18/23 (78%) and 17/23 (74%) patients, respectively; uCDR3 in IgH and Vβ decreased in 14/23 (61%) and 15/23 (65%). Among patients with concordant decreases in ALC and uCDR3, a moderate correlation between magnitude of ALC loss and uCDR3 levels in the T-cell receptor Vβ was observed (r = 0.64, p = 0.02). For both receptor species studied (IgH and Vβ), HFRT patients were more likely to have an increase in either entropy or uCDR3 in the face of decreased ALC (36 vs 15%, X2 p = 0.03). Furthermore, while decreases in entropy were observed among the CFRT patients for both IgH (median entropy 10.4 vs 9.4, p = 0.06) and Vβ (9.7 vs 8.1, p = 0.02), entropy did not significantly change following RT in the HFRT patients (IgH 10.6 vs 10.4, p = 0.74 and Vβ 10.9 vs 10.8, p = 0.24).

CONCLUSION

RT-induced changes in immune repertoire diversity are variably reflected in the peripheral ALC. Both HFRT and CFRT depleted circulating lymphocytes, but patients undergoing HFRT were more likely to experience increases in T and B cell diversity metrics despite lymphopenia. It is therefore possible that relative sparing of repertoire diversity among patients undergoing HFRT could increase the likelihood of tumor antigen recognition by peripheral blood lymphocytes. As immune repertoire diversity is associated with the likelihood of response to immunotherapy, these findings also have implications for RT-immunotherapy combinations. Further study is required to understand the relationship between RT exposure to circulating lymphocyte populations and immune repertoire diversity.

Evaluation of Adjuvant Chemotherapy Survival Outcomes Among Patients With Surgically Resected Pancreatic Carcinoma With Node-Negative Disease After Neoadjuvant Therapy

Importance

Neoadjuvant therapy (NAT) is rarely associated with a complete histopathologic response in patients with pancreatic ductal adenocarcinoma (PDAC) but results in downstaging of regional nodal disease. Such nodal downstaging after NAT may have implications for the use of additional adjuvant therapy (AT).

Objectives

To examine the prognostic implications of AT in patients with node-negative (N0) disease after NAT and to identify factors associated with progression-free (PFS) and overall survival (OS).

Design, Setting, and Participants

A retrospective review was conducted using data from 2 high-volume, tertiary care academic centers (University of Pittsburgh Medical Center and the Medical College of Wisconsin). Prospectively maintained pancreatic cancer databases at both institutes were searched to identify patients with localized PDAC treated with preoperative therapy and subsequent surgical resection between 2010 and 2019, with N0 disease on final histopathology.

Exposures

Patients received NAT consisting of chemotherapy with or without concomitant neoadjuvant radiation (NART). For patients who received NART, chemotherapy regimens were gemcitabine or 5-fluoururacil based and included stereotactic body radiotherapy (SBRT) or intensity-modulated radiation therapy (IMRT) after all intended chemotherapy and approximately 4 to 5 weeks before anticipated surgery. Adjuvant therapy consisted of gemcitabine-based therapy or FOLFIRINOX; when used, adjuvant radiation was commonly administered as either SBRT or IMRT.

Main Outcomes and Measures

The association of AT with PFS and OS was evaluated in the overall cohort and in different subgroups. The interaction between AT and other clinicopathologic variables was examined on Cox proportional hazards regression analysis.

Results

In this cohort study, 430 consecutive patients were treated between 2010 and 2019. Patients had a mean (SD) age of 65.2 (9.4) years, and 220 (51.2%) were women. The predominant NAT was gemcitabine based (196 patients [45.6%]), with a median duration of 2.7 cycles (IQR, 1.5-3.4). Neoadjuvant radiation was administered to 279 patients (64.9%). Pancreatoduodenectomy was performed in 310 patients (72.1%), and 160 (37.2%) required concomitant vascular resection. The median lymph node yield was 26 (IQR, 19-34); perineural invasion (PNI), lymphovascular invasion (LVI), and residual positive margins (R1) were found in 254 (59.3%), 92 (22.0%), and 87 (21.1%) patients, respectively. The restricted mean OS was 5.2 years (95% CI, 4.8-5.7). On adjusted analysis, PNI, LVI, and poorly differentiated tumors were independently associated with worse PFS and OS in N0 disease after NAT, with hazard ratios (95% CIs) of 2.04 (1.43-2.92; P < .001) and 1.68 (1.14-2.48; P = .009), 1.47 (1.08-1.98; P = .01) and 1.54 (1.10-2.14; P = .01), and 1.90 (1.18-3.07; P = .008) and 1.98 (1.20-3.26; P = .008), respectively. Although AT was associated with prolonged survival in the overall cohort, the effect was reduced in patients who received NART and strengthened in patients with PNI (AT × PNI interaction: hazard ratio, 0.55 [95% CI, 0.32-0.97]; P = .04).

Conclusions and Relevance

The findings of this cohort study suggest a survival benefit for AT in patients with N0 disease after NAT and surgical resection. This survival benefit may be most pronounced in patients with PNI.

Severe Radiation-Induced Lymphopenia Attenuates the Benefit of Durvalumab After Concurrent Chemoradiotherapy for NSCLC

Introduction

Durvalumab after concurrent chemoradiation (CCRT) for NSCLC improves survival, but only in a subset of patients. We investigated the effect of severe radiation-induced lymphopenia (sRIL) on survival in these patients.

Methods

Outcomes after CCRT (2010–2019) or CCRT followed by durvalumab (2018–2019) were reviewed. RIL was defined by absolute lymphocyte count (ALC) nadir in samples collected at end of CCRT; sRIL was defined as nadir ALC less than 0.23 × 10⁹/L (the lowest tertile). Progression-free survival (PFS) and overall survival (OS) were calculated by the Kaplan-Meier method. Cox proportional hazard modeling evaluated associations between clinical variables and survival.

Results

Of 309 patients, 192 (62%) received CCRT only and 117 (38%) CCRT plus durvalumab. Multivariable logistic regression analysis indicated that sRIL was associated with planning target volume (OR = 1.002, p = 0.001), stage IIIB disease (OR = 2.77, p = 0.04), and baseline ALC (OR = 0.36, p < 0.01). Durvalumab extended median PFS (23.3 versus 14.1 mo, p = 0.003) and OS (not reached versus 30.8 mo, p < 0.01). sRIL predicted poorer PFS and OS in both treatment groups. Among patients with sRIL, durvalumab did not improve survival (median = 24.6 mo versus 18.1 mo CCRT only, p = 0.079). On multivariable analyses, sRIL (OR = 1.81, p < 0.01) independently predicted poor survival.

Conclusions

Severe RIL compromises survival benefits from durvalumab after CCRT for NSCLC. Measures to mitigate RIL after CCRT may be warranted to enhance the benefit of consolidation durvalumab.

Neoadjuvant Radiotherapy After (m)FOLFIRINOX for Borderline Resectable Pancreatic Adenocarcinoma: A TAPS Consortium Study

BACKGROUND

The value of neoadjuvant radiotherapy (RT) after 5-fluorouracil with leucovorin, oxaliplatin, and irinotecan, with or without dose modifications [(m)FOLFIRINOX], for patients with borderline resectable (BR) pancreatic ductal adenocarcinoma (PDAC) is uncertain.

METHODS

We conducted an international retrospective cohort study including consecutive patients with BR PDAC who received (m)FOLFIRINOX as initial treatment (2012-2019) from the Trans-Atlantic Pancreatic Surgery Consortium. Because the decision to administer RT is made after chemotherapy, patients with metastases or deterioration after (m)FOLFIRINOX or a performance score ≥2 were excluded. Patients who received RT after (m)FOLFIRINOX were matched 1:1 by nearest neighbor propensity scores with patients who did not receive RT. Propensity scores were calculated using sex, age (≤70 vs >70 years), WHO performance score (0 vs 1), tumor size (0-20 vs 21-40 vs >40 mm), tumor location (head/uncinate vs body/tail), number of cycles (1-4 vs 5-8 vs >8), and baseline CA 19-9 level (≤500 vs >500 U/mL). Primary outcome was overall survival (OS) from diagnosis.

RESULTS

Of 531 patients who received neoadjuvant (m)FOLFIRINOX for BR PDAC, 424 met inclusion criteria and 300 (70.8%) were propensity score-matched. After matching, median OS was 26.2 months (95% CI, 24.0-38.4) with RT versus 32.8 months (95% CI, 25.3-42.0) without RT (P=.71). RT was associated with a lower resection rate (55.3% vs 72.7%; P=.002). In patients who underwent a resection, RT was associated with a comparable margin-negative resection rate (>1 mm) (70.6% vs 64.8%; P=.51), more node-negative disease (57.3% vs 37.6%; P=.01), and more major pathologic response with <5% tumor viability (24.7% vs 8.3%; P=.006). The OS associated with conventional and stereotactic body RT approaches was similar (median OS, 25.7 vs 26.0 months; P=.92).

CONCLUSIONS

In patients with BR PDAC, neoadjuvant RT following (m)FOLFIRINOX was associated with more node-negative disease and better pathologic response in patients who underwent resection, yet no difference in OS was found. Routine use of RT cannot be recommended based on these data.

Lymphocyte depletion rate as a biomarker of radiation dose to circulating lymphocytes during fractionated partial-body radiotherapy

Purpose

Radiation causes exponential depletion of circulating lymphocyte populations; in turn, radiation-induced lymphopenia is associated with worse survival for many solid tumors, possibly owing to attenuated antitumor immune responses. Identifying reliable and reproducible methods of calculating the radiation dose to circulating immune cells may facilitate development of techniques to reduce the risk and severity of radiation-induced toxic effects to circulating lymphocytes.

Methods and Materials

Patient-specific lymphocyte loss rates were derived from a clinical data set including 684 adult patients with solid tumors. Multivariable linear regression was used to model the relationship between the lymphocyte loss rate and physical parameters of the radiation plan that determine circulating blood dose.

Results

During partial-body radiation, lymphocyte loss rates are determined by physical parameters of the radiation plan that reflect radiation exposure to circulating cells, including target volume size, dose per fraction squared, and anatomic site treated. Differences in observed versus predicted lymphocyte loss rates may be partly explained by variations in concurrent chemotherapy regimens.

Conclusions

We describe a novel method of using patient-specific lymphocyte loss kinetics to approximate the effective radiation dose to circulating lymphocytes during focal fractionated photon radiation therapy. Clinical applications of these findings include the early identification of patients at particularly high risk of severe radiation-induced lymphopenia based on physical parameters of the radiation therapy plan.

Radiation-Induced Lymphopenia Risks of Photon Versus Proton Therapy for Esophageal Cancer Patients

Purpose

To assess possible differences in radiation-induced lymphocyte depletion for esophageal cancer patients being treated with the following 3 treatment modalities: intensity-modulated radiation therapy (IMRT), passive scattering proton therapy (PSPT), and intensity-modulated proton therapy (IMPT).

Methods and Materials

We used 2 prediction models to estimate lymphocyte depletion based on dose distributions. Model I used a piecewise linear relationship between lymphocyte survival and voxel-by-voxel dose. Model II assumes that lymphocytes deplete exponentially as a function of total delivered dose. The models can be fitted using the weekly absolute lymphocyte counts measurements collected throughout treatment. We randomly selected 45 esophageal cancer patients treated with IMRT, PSPT, or IMPT at our institution (15 per modality) to demonstrate the fitness of the 2 models. A different group of 10 esophageal cancer patients who had received PSPT were included in this study of in silico simulations of multiple modalities. One IMRT and one IMPT plan were created, using our standards of practice for each modality, as competing plans to the existing PSPT plan for each patient. We fitted the models by PSPT plans used in treatment and predicted absolute lymphocyte counts for IMRT and IMPT plans.

Results

Model validation on each modality group of patients showed good agreement between measured and predicted absolute lymphocyte counts nadirs with mean squared errors from 0.003 to 0.023 among the modalities and models. In the simulation study of IMRT and IMPT on the 10 PSPT patients, the average predicted absolute lymphocyte count (ALC) nadirs were 0.27, 0.35, and 0.37 K/μL after IMRT, PSPT, and IMPT treatments using Model I, respectively, and 0.14, 0.22, and 0.33 K/μL using Model II.

Conclusions

Proton plans carried a lower predicted risk of lymphopenia after the treatment course than did photon plans. Moreover, IMPT plans outperformed PSPT in terms of predicted lymphocyte preservation.

The association of age with acute toxicities in NRG oncology combined modality lower GI cancer trials

PURPOSE

Expected toxicity from chemoradiation (CRT) is an important factor in treatment decisions but is poorly understood in older adults with lower gastrointestinal (GI) malignancies. Our objective was to compare acute adverse events (AAEs) of older and younger adults with lower GI malignancies treated on NRG studies.

METHODS

Data from 6 NRG trials, testing combined modality therapy in patients with anal or rectal cancer, were used to test the hypothesis that older age was associated with increased AAEs. AAEs and compliance with protocol-directed therapy were compared between patients aged ≥70 and < 70. Categorical variables were compared across age groups using the chi-square test. The association of age on AAEs was evaluated using a covariate-adjusted logistic regression model, with odds ratio (OR) reported. To adjust for multiple comparisons, a p-value <0.01 was considered statistically significant.

RESULTS

There were 2525 patients, including 380 patients ≥70 years old (15%) evaluable. Older patients were more likely to have worse baseline performance status (PS 1 or 2) (23% vs. 16%, p = 0.001), but otherwise baseline characteristics were similar. Older patients were less likely to complete their chemotherapy (78% vs. 87%, p < 0.001), but had similar RT duration. On univariate analysis, older patients were more likely to experience grade ≥ 3 GI AAEs (36% vs. 23%, p < 0.001), and less likely to experience grade ≥ 3 skin AAEs (8% vs. 14%, p = 0.002). On multivariable analysis, older age was associated with grade ≥ 3 GI AAE (OR 1.93, 95% CI: 1.52, 2.47, p < 0.001) after adjusting for sex, race, PS, and disease site.

CONCLUSIONS

Older patients with lower GI cancers who underwent CRT were less likely to complete chemotherapy and had higher rates of grade 3+ GI AAEs. These results can be used to counsel older adults prior to treatment and manage expected toxicities throughout pelvic CRT.

A Pancreatic Cancer Multidisciplinary Clinic Eliminates Socioeconomic Disparities in Treatment and Improves Survival

AIMS

National studies have demonstrated disparities in the treatment and survival of pancreatic cancer patients based on socioeconomic status (SES). This study aimed to identify specific differences in perioperative management and outcomes based on patient SES and to study the role of a multidisciplinary clinic (MDC) in mitigating any variations.

METHODS

The study analyzed patients undergoing pancreaticoduodenectomy for pancreatic ductal adenocarcinoma in a large hospital system. The patients were categorized into groups of high and low SES and whether they were managed by the authors' pancreatic cancer MDC or not. The study compared differences in disease characteristics, receipt of multimodality therapy, perioperative outcomes, and recurrence-free and overall survival.

RESULTS

Of the 162 low-SES patients and 119 high-SES patients, 54% were managed in the MDC. Outside the MDC, low-SES patients were less likely to receive neoadjuvant chemotherapy and had less minimally invasive surgery, a longer OR time, less enhanced recovery participation, and more major complications (p < 0.05). No SES disparities were observed among the MDC patients. Despite similar tumor characteristics, the low-SES patients had inferior median overall survival (21 vs 32 months; p = 0.005), but the MDC appeared to eliminate this disparity. Low SES correlated with inferior survival for the non-MDC patients (17 vs 32 months; p < 0.001), but not for the MDC patients (24 vs 25 months; p = 0.33). These findings persisted in the multivariable analysis.

CONCLUSION

A pancreatic cancer MDC standardizes treatment decisions, eliminates disparities in surgical outcomes, and improves survival for low-SES patients.

Outcomes of Neoadjuvant Chemotherapy Versus Chemoradiation in Localized Pancreatic Cancer: A Case-Control Matched Analysis

BACKGROUND

Neoadjuvant therapy is increasingly used for patients with pancreatic ductal adenocarcinoma (PDAC). It is unknown whether neoadjuvant chemoradiotherapy is more effective than chemotherapy (NCRT vs. NAC). We aim to compare pathological and survival outcomes of NCRT and NAC in patients with PDAC.

PATIENTS AND METHODS

Single-center analysis of PDAC patients treated with NCRT or NAC followed by resection between December 2008 and December 2018 was performed. Average treatment effect (ATE) was estimated after case-control matching using Mahalanobis distance nearest-neighbor matching. Inverse probability weighted estimates (IPWE)-based ATE was estimated for disease-free survival (DFS) and overall survival (OS).

RESULTS

Among the 418 patients (mean age 66.8 years, 51% female) included in the study, 327 received NAC and 91 received NCRT. NCRT patients had higher rates of locally advanced disease, number of neoadjuvant chemotherapy cycles, more chemotherapy regimen crossover (gemcitabine and 5-FU based), and were more likely to undergo open surgical procedures and/or vascular resection (all p < 0.05). After matched analysis, NCRT was associated with a significant reduction in lymph node positive disease [ATE = (-)0.24, p = 0.007] and lymphovascular invasion [ATE = (-)0.20, p = 0.02]. While NCRT was associated with significantly improved DFS by 9.5 months (p = 0.006), it did not affect OS by IPWE-based ATE after adjusting for adjuvant therapy (ATE = 5.5 months; p = 0.32).

CONCLUSION

Compared with NAC alone, NCRT is associated with improved pathologic surrogates and disease-free survival, but not overall survival in patients with PDAC.

Demographic factors associated with missed follow-up among solid tumor patients treated at a large multi-site academic institution

Aim: To identify demographic predictors of patients who miss oncology follow-up, considering that missed follow-up has not been well studies in cancer patients. Methods: Patients with solid tumors diagnosed from 2007 to 2016 were analyzed (n = 16,080). Univariate and multivariable logistic regression models were constructed to examine predictors of missed follow-up. Results: Our study revealed that 21.2% of patients missed ≥1 follow-up appointment. African-American race (odds ratio [OR] 1.33; 95% CI: 1.17-1.51), Medicaid insurance (OR 1.59; 1.36-1.87), no insurance (OR 1.66; 1.32-2.10) and rural residence (OR 1.78; 1.49-2.13) were associated with missed follow-up. Conclusion: Many cancer patients miss follow-up, and inadequate follow-up may influence cancer outcomes. Further research is needed on how to address disparities in follow-up care in high-risk patients.

Baseline Karnofsky performance status is independently predictive of death within 30 days of intracranial radiation therapy completion for metastatic disease

Introduction

For patients with brain metastases, palliative radiation therapy (RT) has long been a standard of care for improving quality of life and optimizing intracranial disease control. The duration of time between completion of palliative RT and patient death has rarely been evaluated.

Methods

A compilation of two prospective institutional databases encompassing April 2015 through December 2018 was used to identify patients who received palliative intracranial radiation therapy. A multivariate logistic regression model characterized patients adjusting for age, sex, admission status (inpatient versus outpatient), Karnofsky Performance Status (KPS), and radiation therapy indication.

Results

136 consecutive patients received intracranial palliative radiation therapy. Patients with baseline KPS <70 (OR = 2.2; 95%CI = 1.6-3.1; p < 0.0001) were significantly more likely to die within 30 days of treatment. Intracranial palliative radiation therapy was most commonly delivered to provide local control (66% of patients) or alleviate neurologic symptoms (32% of patients), and was most commonly delivered via whole brain radiation therapy in 10 fractions to 30 Gy (38% of patients). Of the 42 patients who died within 30 days of RT, 31 (74%) received at least 10 fractions.

Conclusions

Our findings indicate that baseline KPS <70 is independently predictive of death within 30 days of palliative intracranial RT, and that a large majority of patients who died within 30 days received at least 10 fractions. These results indicate that for poor performance status patients requiring palliative intracranial radiation, hypofractionated RT courses should be strongly considered.

Radiosurgery dose reduction for brain metastases on immunotherapy (RADREMI): A prospective phase I study protocol

INTRODUCTION

Up to 20% of patients with brain metastases treated with immune checkpoint inhibitor (ICI) therapy and concomitant stereotactic radiosurgery (SRS) suffer from symptomatic radiation necrosis. The goal of this study is to evaluate Radiosurgery Dose Reduction for Brain Metastases on Immunotherapy (RADREMI) on six-month symptomatic radiation necrosis rates.

METHODS

This study is a prospective single arm Phase I pilot study which will recruit patients with brain metastases receiving ICI delivered within 30 days before SRS. All patients will be treated with RADREMI dosing, which involves SRS doses of 18 Gy for 0-2 cm lesions, 14 Gy for 2.1-3 cm lesions, and 12 Gy for 3.1-4 cm lesions. All patients will be monitored for six-month symptomatic radiation necrosis (defined as a six-month rate of clinical symptomatology requiring steroid administration and/or operative intervention concomitant with imaging findings consistent with radiation necrosis) and six-month local control. We expect that RADREMI dosing will significantly reduce the symptomatic radiation necrosis rate of concomitant SRS + ICI without significantly sacrificing the local control obtained by the present RTOG 90-05 SRS dosing schema. Local control will be defined according to the Response Assessment in Neuro-Oncology (RANO) criteria.

DISCUSSION

This study is the first prospective trial to investigate the safety of dose-reduced SRS in treatment of brain metastases with concomitant ICI. The findings should provide fertile soil for future multi-institutional collaborative efficacy trials of RADREMI dosing for this patient population.

TRIAL REGISTRATION

Clinicaltrials.gov identifier: NCT04047602 (registration date: July 25, 2019).

Comprehensive Analysis of the Kinetics of Radiation-Induced Lymphocyte Loss in Patients Treated with External Beam Radiation Therapy

Radiation-induced lymphopenia (RIL) is associated with worse survival in patients with solid tumors, as well as lower response rates to checkpoint inhibitors. While single-fraction total-body irradiation is known to result in exponential decreases in the absolute lymphocyte count (ALC), the kinetics of lymphocyte loss after focal fractionated exposures have not previously been characterized. In the current study, lymphocyte loss kinetics was analyzed among patients undergoing focal fractionated radiotherapy for clinical indications. This registry-based study included 419 patients who received either total-body irradiation (TBI; n = 30), stereotactic body radiation therapy (SBRT; n = 73) or conventionally fractionated chemoradiation therapy (CFRT; n = 316). For each patient, serial ALCs were plotted against radiotherapy fraction number. The initial three weeks of treatment for CFRT patients and the entirety of treatment for SBRT and TBI patients were fit to exponential decay in the form ALC(x) = ae^-bx, where ALC(x) is the ALC after x fractions. From those fits, fractional lymphocyte loss (FLL) was calculated as FLL = (1 - e^-b) * 100, and multivariable regression was performed to identify significant correlates of FLL. Median linearized R² when fitting the initial fractions was 0.98, 0.93 and 0.97 for patients receiving TBI, SBRT and CFRT, respectively. In CFRT patients, apparent ALC loss rate slowed after week 3. Fitting ALC loss over the entire CFRT course therefore required the addition of a constant term, "c". For TBI and SBRT patients, treatment ended during the pure exponential decay phase. Initial FLL varied significantly with treatment technique. Mean FLL was 35.5%, 24.3% and 10.77% for patients receiving TBI, SBRT and CFRT, respectively (P < 0.001). Significant correlates of FLL varied by site and included field size, dose per fraction, mean spleen dose, chemotherapy backbone and age. Finally, total percentage ALC loss during radiotherapy was highly correlated with FLL (P < 0.001). Lymphocyte depletion kinetics during the initial phase of fractionated radiotherapy are characterized by pure exponential decay. Initial FLL is strongly correlated with radiotherapy planning parameters and total percentage ALC loss. The two groups with the highest FLL received no concurrent chemotherapy, suggesting that ALC loss can be a consequence of radiotherapy alone. This work may assist in selecting patients for adaptive radiotherapy approaches to mitigate RIL risk.

Predictors of death for patients treated with palliative intent radiation using prospective databases

e18325

Background: Radiation therapy (RT) effectively palliates pain and other symptoms due to cancer and is frequently offered to patients with metastatic disease. Standard palliative RT schedules range from one to ≥10 fractions; however, prognostic variables related to survival following palliative RT are not well studied. A better understanding of prognostic factors in this setting will assist in selecting appropriate palliative RT regimens for patients with short life expectancies. Methods: Prospectively maintained institutional databases tracking morbidity and mortality and outcomes of a departmental peer review process were used to identify patients who received palliative RT between April 2015 - December 2018. Cox proportional hazards model was used to identify factors correlated with overall survival (OS) and mortality within 30 days (30DM) after completing palliative RT, including age, sex, admission status (inpatient versus outpatient), Karnofsky Performance Status (KPS), treated site and primary site. Results: Of 421 patients, 389 received palliative RT; 30DM rate was 25.7% (n = 100). Median age was 61, median KPS was 70, and 196 patients (50.4%) were female. The most common primary sites were thorax (N = 121, 31.1%) and genitourinary tract (N = 60, 15.4%), while the most commonly treated sites were brain (N = 136, 35.0%) and bone (N = 71, 18.3%). KPS and treatment site were strong independent predictors of both OS and 30DM. Patients with KPS < 70 had an HR 2.61 (95% CI 1.54 – 4.43, p < 0.0001) for 30DM and an HR of 2.22 (95% CI 1.57 – 3.13, p < 0.0001) for death at any time after RT. Treatment site was also independently associated with OS and 30DM (p = 0.01 and p = 0.03, respectively). Median OS durations (95%CI) were: abdomen/pelvis not reached, bone 340 days (133.9-546.1), spine 207 days (28.0 – 386.0), head and neck 184 days (55.2 – 312.8), brain 104 days (34.3 – 173.7) and thorax 76 days (10.3 – 141.7). Age, sex, admission status and primary site were not correlated with OS or 30DM. Conclusions: RT can effectively palliate distressing symptoms related to primary or metastatic cancers. However, early mortality is common following palliative RT and is associated with poor performance status and treated site. Hypofractionated RT courses should be considered in patients with such risk factors to reduce the burden of prolonged therapy on patients and their families at the end of life.

Comprehensive analysis of lymphocyte loss kinetics in patients treated with focal fractionated radiation therapy

e14191

Background: Radiation (RT)-induced lymphopenia (RIL) is associated with worse survival in patients with solid tumors as well as lower response rates to checkpoint inhibitors. This analysis aimed to model the kinetics of lymphocyte depletion during RT to assist in predicting RIL risk. Methods: This registry-based study included 419 patients who received either total body irradiation (TBI; n = 30), stereotactic body RT (SBRT; n = 73), or conventional chemoradiation (CRT; n = 316). For each patient, serial absolute lymphocyte counts (ALCs) were plotted against RT fraction number. The initial 3 weeks of treatment for conventionally irradiated patients and the entirety of treatment for SBRT and TBI patients were fit to exponential decay in the form ALC(x) = ae-bx. From those fits, percent per fraction loss in ALC (PFLAC) was calculated as PFLAC = (1 – e-b)*100, and multivariable regression was performed to find its significant predictors. Results: Curves were well fitted by exponential decay for all RT techniques (median linearized R2 0.98, 0.93, and 0.97 for patients treated with TBI, SBRT, and CRT, respectively). In CRT patients, apparent ALC loss rate slowed after week 3, potentially due to lymphocyte repopulation or other factors. TBI and SBRT patients completed RT before the end of the exponential decay phase, and their ALC loss rates remained unchanged throughout RT. Initial PFLAC varied significantly with treatment technique. Mean PFLAC was 35.5%, 24.3%, and 10.77% for patients treated with TBI, SBRT, and CRT, respectively (p < 0.001). Significant predictors of PFLAC varied by site and included field size, dose per fraction, mean spleen dose, chemotherapy backbone, and age. In pancreas cancer patients, gemcitabine was associated with a higher PFLAC (mean = 10.7) than 5-FU (mean = 8.3) after adjustment for covariates (p < 0.001). Finally, total % ALC loss during RT was highly correlated with PFLAC (p < 0.001). Conclusions: Lymphocyte depletion kinetics during the initial phase of fractionated RT are characterized by pure exponential decay. Initial PFLAC is strongly correlated with RT planning parameters and predicts total % ALC loss. The highest ALC loss rates were associated with RT-only regimens, implying that concurrent chemotherapy is not solely responsible for lymphopenia in patients receiving CRT. This work may also assist in selecting patients for adaptive RT approaches to mitigate RIL risk.

How often do metastatic brain disease patients prescribed ten fractions of whole brain radiation therapy die without completing treatment?

e23146

Background: Palliative whole brain radiation therapy (WBRT) is an established treatment modality for optimizing intracranial disease control in patients with metastatic brain disease, most commonly prescribed to a dose of 30 Gy in 10 fractions. The fidelity of WBRT course completion has not been rigorously evaluated. Methods: Two prospective institutional databases evaluating patients treated from April 2015 through December 2018 identified patients who received WBRT. Information analyzed included prescribed and delivered dose/fractionation, Karnofsky Performance Status (KPS), admission status, treatment duration, and time between WBRT completion and patient hospice referral/death. Results: 52 consecutive patients prescribed WBRT to 30 Gy in 10 fractions were evaluated. Twenty-six (50%) were inpatients, and only 17 (33%) had a KPS of 70 or greater (able to care for self). More than 30% failed to finish more than 5 of their 10 scheduled fractions before either death or hospice referral; only 29 of 52 patients (56%) finished WBRT as prescribed. Twenty-five patients (48%) died within 30 days of WBRT. Of the 29 patients who finished WBRT as prescribed, 9 (31%) died within 30 days of completing treatment. Of the 23 patients who failed to complete standard fractionation WBRT, nearly 70% died within 30 days of receiving their final radiation therapy fraction. Conclusions: These findings indicate that greater than 40% of patients prescribed the standard 10 fraction regimen of WBRT do not remain well enough to tolerate the full duration of treatment. With nearly half of patients dying within 30 days of standard fraction WBRT, there should be strong incentive to reduce the physical and financial burden on this patient population by prescribing WBRT with significantly shorter (i.e. single-fraction or 20 Gy in 5 fractions) fractionation schemes.

Aquaporin-4 Expression Patterns in Glioblastoma Pre-Chemoradiation and at Time of Suspected Progression

There has been controversy about the presence and potential role of aquaporin-4 (AQP4) in glioblastoma (GBM). We analyzed tissue from 22 patients with newly-diagnosed GBM as well as matching tissue from 17 of these cases who underwent repeat resection for suspected recurrence and performed immunohistochemical analysis for AQP-4 expression. While some degree of AQP4 expression was detected in all 22 cases (39 samples), there was no clear relationship between staining pattern and disease status (active versus inactive GBM) between baseline and time of repeat biopsy. In addition, there was no clear relationship between AQP4 expression and degree of edema.

Prospective trial of functional liver image-guided hepatic therapy (FLIGHT) with hepatobiliary iminodiacetic acid (HIDA) scans and update of institutional experience

373

Background: Functional liver image-guided hepatic therapy (FLIGHT) is a novel stereotactic body radiation therapy (SBRT) planning technique. A functional map, generated from hepatobiliary iminodiacetic acid (HIDA) scans, is used to maximize the functional residual capacity of liver receiving < 15 Gy (FRC15HIDA). We present initial results of a prospective trial evaluating FLIGHT vs standard planning and update our institutional experience. Methods: Eligible patients were ≥ 18 yo with 1o or 2o liver malignancy and Child-Pugh ≤ B7. Liver function was assessed with HIDA and blood chemistry at baseline, mid-treatment, and 3, 6, and 12 months post SBRT. Both standard and FLIGHT (optimized to avoid high functioning liver) plans were generated for each patient. Treating MDs were blinded to planning technique before selecting the treatment plan. The primary endpoint was to show > 5% increase in FRC15HIDA in 3/15 pts. Secondary endpoints included the rate FLIGHT plans were selected and changes in HIDA and other liver function tests. Prior institutional experience included 27 pts with FLIGHT planned retrospectively. Paired t-test was used to compare dosimetric endpoints for FLIGHT vs. standard plans, including: FRC15HIDA, mean liver dose, effective uniform dose (EUD), and functional EUD (FEUD). Results: Fifteen pts were enrolled. The primary endpoint was met, as 4/15 pts had > 5% improvement in FRC15HIDA (mean 5.2%, range -2.3-19.8%). Notably, the FLIGHT plan was selected in 11/15. The mean improvements in FRC15HIDA (5.2 vs 5.0%), mean liver dose (11.9 vs. 13.0%), EUD (5.1 vs 5.2%), and FEUD (6.9 vs 7.1%) were similar between prospective and retrospective cohorts (p > 0.5). In the entire cohort (n = 42), FLIGHT improved FRC15HIDA, mean liver dose, EUD, and FEUD ( p ≤ 0.001). There were > 5% improvements in FRC15HIDA in 15, mean liver dose in 31, EUD in 19, and FEUD in 27. Conclusions: FLIGHT with HIDA led to improvements in all analyzed dosimetric parameters. The extent of benefit was similar in both cohorts, and there was individual variation in the extent of benefit. Longer follow-up is required to determine the effect of FLIGHT on post-SBRT liver function. Clinical trial information: NCT03338062.

Neoadjuvant therapy duration and outcome of patients with resectable and borderline resectable pancreatic ductal adenocarcinoma (PDAC)

436

Background: Neoadjuvant chemotherapy (NA CT) may improve surgical selection for resectable (R) PDAC, and margin negative resection in borderline resectable (BR) PDAC. Optimal duration of NA CT is unknown, as is the role of XRT with modern chemotherapy. We compared survival outcomes by duration of NA CT and NA CT + XRT. Methods: Patients with R or BR PDAC who underwent NA CT with or without XRT and followed by curative resection were included in this analysis. Data was extracted from an IRB approved pancreatic cancer database at Indiana University. Disease Free (DFS) and Overall (OS) survival were calculated from the surgery date and compared between: 1) < 3 v ≥ 3 months NA CT and 2) NA CT with/without XRT. Results: Between Summer 2008 and Summer 2018, 116 patients received NA CT with or without XRT and completed surgical resection. Median (range) age was 63 years (36, 84), stages were R=47%, BR=53%. Most patients received modified FOLFIRINOX or FOLFIRINOX (59 %), or gemcitabine/nab-paclitaxel (13%) and 24% received XRT. There were four (3 %) pathologic complete responders, all in the ≥ 3 mo NA CT + XRT group. Percent node positive was lower in NA CT + XRT versus NA CT only (median 0% vs 7.4%, p < 001), but did not differ by duration of NA CT. With a median (range) follow-up time of 13.7 mo (0.7, 83.0), median OS was 22.5 mo (19.5, 29.8) with < 3 mo NA CT versus 16.3 (12.2, 18.9) with ≥ 3 mo NA CT (p = 0.02) and was 22.6 mo (17.0, 82.9) with NA CT + XRT versus 19.5 (13.1, 22.5) in NA CT only (p = 0.03). There was no difference in DFS by duration of NA CT or XRT. Conclusions: In this study, patients who received a shorter course of chemotherapy and radiation had improved mOS when calculated from the surgery date. While this unexpected results could reflect selection bias of therapy, further analysis will account for tumor stage at diagnosis, perioperative complications and use propensity score adjustment to examine/adjust for possible treatment selection bias.

Predictors of Nodal and Metastatic Failure in Early Stage Non-small-cell Lung Cancer After Stereotactic Body Radiation Therapy

INTRODUCTION/BACKGROUND

Many patients with early stage non-small-cell lung cancer (ES-NSCLC) undergoing stereotactic body radiation therapy (SBRT) develop metastases, which is associated with poor outcomes. We sought to identify factors predictive of metastases after lung SBRT and created a risk stratification tool.

MATERIALS AND METHODS

We included 363 patients with ES-NSCLC who received SBRT; the median follow-up was 5.8 years. The following patient and tumor factors were retrospectively analyzed for their association with metastases (defined as nodal and/or distant failure): gender; age; lobe involved; centrality; previous NSCLC; smoking status; gross tumor volume (GTV); T-stage; histology; dose; minimum, maximum, and mean GTV dose; and parenchymal lung failure. A metastasis risk-score linear-model using beta coefficients from a multivariate Cox model was built.

RESULTS

A total of 111 (27.3%) of 406 lesions metastasized. GTV and dose were significantly associated with metastases on univariate and multivariate Cox proportional hazards modeling (P < .001 and hazard ratio [HR], 1.02 per mL; P < .05 and HR, 0.99 per Gy, respectively). Histology, T-stage, centrality, lung parenchymal failures, and previous NSCLC were not associated with development of metastasis. A metastasis risk-score model using GTV and prescription dose was built: risk score = (0.01611 × GTV) - (0.00525 × dose [BED₁₀]). Two risk-score cutoffs separating the cohort into low-, medium-, and high-risk subgroups were examined. The risk score identified significant differences in time to metastases between low-, medium-, and high-risk patients (P < .001), with 3-year estimates of 81.1%, 63.8%, and 38%, respectively.

CONCLUSION

GTV and radiation dose are associated with time to metastasis and may be used to identify patients at higher risk of metastasis after lung SBRT.

Field size effects on the risk and severity of treatment-induced lymphopenia in patients undergoing radiation therapy for solid tumors

Purpose

Radiation-induced lymphopenia (RIL) is the result of direct toxicity to circulating lymphocytes as they traverse the irradiated field, occurs in 40% to 70% of patients who undergo conventional external beam radiation therapy, and is associated with worse outcomes in multiple solid tumors. As immunotherapy strategies evolve, a better understanding of radiation's effects on the immune system is needed in order to develop rational methods of combining RT with immunotherapy.

Methods and materials

This paper is a review of the available literature on the clinical significance and dosimetric predictors of radiation-induced toxicity to the immune system.

Results

An association between severe RIL and inferior survival has been described in multiple solid tumors, including glioma, lung cancer, and pancreatic cancer. RIL risk is correlated with field size, dose per fraction, and fraction number. SBRT and proton therapy techniques are associated with lower RIL risk.

Conclusions

The immune system should be considered an organ at risk during RT, and absolute lymphocyte count is an important biomarker of RT-induced immunotoxicity. Radiation dose and technique affect the risk and severity of RIL. Further research is needed to accurately characterize RT-induced immunotoxicity and develop strategies to prevent or mitigate this clinically significant side effect.

Functional liver image guided hepatic therapy (FLIGHT) with hepatobiliary iminodiacetic acid (HIDA) scans

PURPOSE

Hepatobiliary iminodiacetic acid (HIDA) scans provide global and regional assessments of liver function that can serve as a road map for functional avoidance in stereotactic body radiation therapy (SBRT) planning. Functional liver image guided hepatic therapy (FLIGHT), an innovative planning technique, is described and compared with standard planning using functional dose-volume histograms. Thresholds predicting for decompensation during follow up are evaluated.

METHODS AND MATERIALS

We studied 17 patients who underwent HIDA scans before SBRT. All SBRT cases were replanned using FLIGHT. The following dosimetric endpoints were compared for FLIGHT versus standard SBRT planning: functional residual capacity <15 Gy (FRC₁₅HIDA), mean liver dose (MLD), equivalent uniform dose (EUD), and functional EUD (FEUD). Receiver operating characteristics curves were used to evaluate whether baseline HIDA values, standard cirrhosis scoring, and/or dosimetric data predicted clinical decompensation.

RESULTS

Compared with standard planning, FLIGHT significantly improved FRC₁₅HIDA (mean improvement: 5.3%) as well as MLD, EUD, and FEUD (P < .05). Considerable interindividual variations in the extent of benefit were noted. Decompensation during follow-up was associated with baseline global HIDA <2.915%/min/m², FRC₁₅HIDA <2.11%/min/m², and MELD ≥11 (P < .05).

CONCLUSIONS

FLIGHT with HIDA-based parameters may complement blood chemistry-based assessments of liver function and facilitate individualized, adaptive liver SBRT planning.

Kinetics and dosimetric predictors of acute radiation-induced lymphopenia in pancreatic cancer

300

Background: Radiation (RT) induced lymphopenia (RIL) is an adverse prognostic factor in pancreatic cancer (PC) and is likely due to the irradiation of lymphocytes in the RT field. The goal of this study was to identify dosimetric predictors for high rates of absolute lymphocyte count (ALC) loss during RT for PC. Methods: This was a retrospective study of 34 PC patients in an institutional database who had received concurrent 5-FU or gemcitabine-based chemoradiation (50-54 Gy) and had ≥ 3 ALCs measured during RT. Baseline ALC was normal (>1000 cells/uL) in 28/34 (82%) and grade 3-4 RIL occurred in 24/34 (71%). ALC was plotted against fraction # and a best-fit line for each patient was created to determine per-fraction loss in ALC (PFLAC). Linear regression was used to correlate PFLAC with dosimetric parameters including mean dose to gut, liver, kidney, spleen, and cisterna chyli, as well as estimated dose to immune cells (EDIC), which calculates dose to immune cells according to the % of body lymphocytes contained in each organ. Results: All patients exhibited exponential loss in ALC during RT. Mean PFLAC was 6.8% (range 1.7-13.4); fraction # was strongly correlated with ALC (mean R2 = 0.89). Patients with >/= grade 3 lymphopenia had a significantly higher PFLAC than those with grade 0 - 2 lymphopenia (mean daily loss 7.8% in Gr 3-4 vs. 4.8% in Gr 0-2, p = 0.001; independent sample T test). Field size was not correlated with PFLAC for high (> 1 Gy) or low (< 0.5 Gy) isodose volumes. Mean whole body (r = 0.59, p < 0.001), bowel (r = 0.39, p = 0.012), liver (r = 0.42, p = 0.007), and cisterna chyli (r = 0.583, p = 0.004) doses were moderately correlated with PFLAC; mean kidney (r = 0.22, p = 0.11) and spleen (r = 0.26, p = 0.06) doses were weakly correlated with PFLAC. EDIC was more strongly correlated with PFLAC than any individual organ mean dose (r = 0.69, p < 0.001). Conclusions: Patients undergoing RT for PC experience a predictable RIL characterized by an exponential loss of lymphocytes per day. PFLAC is a useful method of characterizing RIL and facilitates evaluation of dosimetric predictors of RIL. We identified dose to cisterna chyli as a significant contributor to RIL in PC; however, EDIC has a stronger correlation with RIL severity than any single organ dose.

Principal component analysis identifies patterns of cytokine expression in non-small cell lung cancer patients undergoing definitive radiation therapy

Background/Purpose

Radiation treatment (RT) stimulates the release of many immunohumoral factors, complicating the identification of clinically significant cytokine expression patterns. This study used principal component analysis (PCA) to analyze cytokines in non-small cell lung cancer (NSCLC) patients undergoing RT and explore differences in changes after hypofractionated stereotactic body radiation therapy (SBRT) and conventionally fractionated RT (CFRT) without or with chemotherapy.

Methods

The dataset included 141 NSCLC patients treated on prospective clinical protocols; PCA was based on the 128 patients who had complete CK values at baseline and during treatment. Patients underwent SBRT (n = 16), CFRT (n = 18), or CFRT (n = 107) with concurrent chemotherapy (ChRT). Levels of 30 cytokines were measured from prospectively collected platelet-poor plasma samples at baseline, during RT, and after RT. PCA was used to study variations in cytokine levels in patients at each time point.

Results

Median patient age was 66, and 22.7% of patients were female. PCA showed that sCD40l, fractalkine/C3, IP10, VEGF, IL-1a, IL-10, and GMCSF were responsible for most variability in baseline cytokine levels. During treatment, sCD40l, IP10, MIP-1b, fractalkine, IFN-r, and VEGF accounted for most changes in cytokine levels. In SBRT patients, the most important players were sCD40l, IP10, and MIP-1b, whereas fractalkine exhibited greater variability in CFRT alone patients. ChRT patients exhibited variability in IFN-γ and VEGF in addition to IP10, MIP-1b, and sCD40l.

Conclusions

PCA can identify potentially significant patterns of cytokine expression after fractionated RT. Our PCA showed that inflammatory cytokines dominate post-treatment cytokine profiles, and the changes differ after SBRT versus CFRT, with vs without chemotherapy. Further studies are planned to validate these findings and determine the clinical significance of the cytokine profiles identified by PCA.

Survival after recurrence following curative intent resection of pancreatic adenocarcinoma

e15760

Background: Limited data exist on outcomes after recurrence after curative intent resection for pancreatic adenocarcinoma (PC). We analyzed the prognostic significance of time to recurrence and associated factors in patients with post-pancreatectomy recurrence. Methods: Patients with a documented recurrence were identified from a prospectively maintained database of all patients undergoing pancreatectomy for PC between 2009-2015 at Indiana University Simon Cancer Center. Patients were divided into early (≤12 mos after surgery) and late recurrence ( > 12 mos after surgery) groups. Demographic (age, race, sex) and clinical (CA19-9, bilirubin, tumor location, diabetes at presentation, ECOG performance status, margin status, adjuvant radiation, and adjuvant and metastatic chemotherapy) data were obtained by manual chart abstraction and the IUSCC cancer registry. The primary outcome was survival after recurrence (SAR- time from recurrence to death). Chi-square was used for univariate analysis (except t-test for age). SAR was estimated with the Kaplan-Meier method and 95% CI. Cox proportional-hazard model was used for multivariate analysis (MVA). Results: Of 437 patients undergoing surgical resection for PC, 235 had documented recurrence. More patients had an early recurrence (ER) as compared to late recurrence (LR; 58.3% vs 41.7%). Median SAR was significantly shorter in ER vs LR patients (6.2 vs 8.6 months; HR, 0.71; 95% CI, 0.53-0.95; P = 0.02). Median age was higher in ER vs LR groups (64.7 and 64.2 years, P = 0.002). Diabetes at presentation was more common in ER vs LR group (42.3% vs 30.6%, P = 0.09). ER patients were significantly less likely to have received adjuvant chemotherapy (70.7% vs 84.7%, P = 0.02) or adjuvant radiation (17.8% vs 40%, P < .001). MVA showed a trend towards longer SAR in LR vs ER group (adjusted HR = 0.75, 95% CI, 0.55-0.1.03, p = 0.08) however age, adjuvant chemotherapy, adjuvant radiation and diabetes at presentation were not significant predictors of SAR. Conclusions: Early recurrence after resection of PC is associated with poor SAR. Patients with early recurrence tend to be older, diabetic at presentation and were less likely to have received adjuvant chemotherapy or adjuvant radiation.

The Association Between Chemoradiation-related Lymphopenia and Clinical Outcomes in Patients With Locally Advanced Pancreatic Adenocarcinoma

OBJECTIVES

Lymphopenia is a common consequence of chemoradiation therapy yet is seldom addressed clinically. This study was conducted to determine if patients with locally advanced pancreatic cancer (LAPC) treated with definitive chemoradiation develop significant lymphopenia and if this affects clinical outcomes.

METHODS

A retrospective analysis of patients with LAPC treated with chemoradiation at a single institution from 1997 to 2011 was performed. Total lymphocyte counts (TLCs) were recorded at baseline and then monthly during and after chemoradiation. The correlation between treatment-induced lymphopenia, established prognostic factors, and overall survival was analyzed using univariate Cox regression analysis. Important factors identified by univariate analysis were selected as covariates to construct a multivariate proportional hazards model for survival.

RESULTS

A total of 101 patients met eligibility criteria. TLCs were normal in 86% before chemoradiation. The mean reduction in TLC per patient was 50.6% (SD, 40.6%) 2 months after starting chemoradiation (P<0.00001), and 46% had TLC<500 cells/mm. Patients with TLC<500 cells/mm 2 months after starting chemoradiation had inferior median survival (8.7 vs. 13.3 mo, P=0.03) and PFS (4.9 vs. 9.0 mo, P=0.15). Multivariate analysis revealed TLC<500 cells/mm to be an independent predictor of inferior survival (HR=2.879, P=0.001) along with baseline serum albumin (HR=3.584, P=0.0002), BUN (HR=1.060, P=0.02), platelet count (HR=1.004, P=0.005), and radiation planning target volume (HR=1.003, P=0.0006).

CONCLUSIONS

Severe treatment-related lymphopenia occurs frequently after chemoradiation for LAPC and is an independent predictor of inferior survival.

Efficacy of platinum chemotherapy agents in the adjuvant setting for adenosquamous carcinoma of the pancreas

BACKGROUND

Pancreatic adenosquamous carcinoma (PASC) accounts for only 1-4% of all exocrine pancreatic cancers and carries a particularly poor prognosis. This retrospective study was performed to determine whether inclusion of a platinum agent as part of adjuvant therapy is associated with improved survival in patients with resected PASC.

METHODS

Records of all patients who underwent pancreatic resection at Johns Hopkins Hospital from 1986 to 2012 were reviewed to identify those with PASC. Multivariable Cox proportional hazards modeling was used to assess for significant associations between patient characteristics and survival.

RESULTS

In total, 62 patients (1.1%) with resected PASC were identified among 5,627 cases. Median age was 68 [interquartile range (IQR), 57-77] and 44% were female. Multivariate analysis revealed that, among all patients (n=62), the following factors were independently predictive of poor survival: lack of adjuvant therapy [hazard ratio (HR) =3.6; 95% confidence interval (CI), 1.8-7.0; P<0.001], margin-positive resection (HR =3.5; 95% CI, 1.8-6.8; P<0.001), lymph node involvement (HR =3.5; 95% CI, 1.5-8.2; P=0.004), and age (HR =1.0; 95% CI, 1.0-1.1; P=0.035). There were no significant differences between patients who did and did not receive adjuvant therapy following resection (all P>0.05). A second multivariable model included only those patients who received adjuvant therapy (n=39). Lack of inclusion of a platinum agent in the adjuvant regimen (HR =2.4; 95% CI, 1.0-5.8; P=0.040) and larger tumor diameter (HR =1.3; 95% CI, 1.0-1.6; P=0.047) were independent predictors of inferior survival.

CONCLUSIONS

Addition of a platinum agent to adjuvant regimens for resected PASC may improve survival among these high-risk patients, though collaborative prospective investigation is needed.

Changes in circulating lymphocyte counts and tumor-infiltrating lymphocyte subpopulations among patients receiving endorectal brachytherapy for rectal adenocarcinoma

500

Background: Radiation-induced lymphopenia (RIL) is common in patients with rectal cancer and is associated with worse outcomes in rectal cancer and other solid tumors. We investigated whether endorectal brachytherapy (EB) is associated with milder RIL than external radiation treatment (XRT) in patients with rectal cancer and whether EB is associated with altered tumor-infiltrating lymphocyte (TIL) subpopulations. Methods: Records from 11 patients enrolled in a prospective study of EB for T2-3, N0-1 rectal cancer were reviewed; total lymphocyte counts (TLC) at baseline and 2 months after treatment were recorded. EB was given to tumor alone (6.5Gy x 4). XRT was given to the whole pelvis (median dose 50.4Gy). All patients underwent proctectomy after EB/XRT. TLCs from EB patients were compared to 62 rectal cancer patients receiving pelvic XRT + capecitabine. Proctectomy specimens were immunostained for TIL (CD3, CD4, CD8, FoxP3, and CD25) subtypes. Results: Median baseline TLC in EB patients was 1930 vs 1570 cells/uL in XRT patients (p>0.05). Two months after treatment, median TLC was 1550 in EB patients vs 520 in XRT patients (p<0.01). EB patients had higher absolute numbers of CD3, CD4, CD8, and FoxP3+ TILs (Table). Conclusions: EB spares circulating lymphocytes and is associated with increased TIL expression after therapy. Further investigation is needed to determine if there is a causal association between these phenomena and if the increased TIL expression in EB patients is associated with better outcomes in these patients. [Table: see text]

Efficacy of platinum chemotherapy agents in the adjuvant setting for adenosquamous carcinoma of the pancreas

269

Background: Pancreatic adenosquamous carcinoma (PASC) accounts for only 1-4% of exocrine pancreatic cancers and carries a particularly poor prognosis. This retrospective study was performed to determine whether inclusion of a platinum agent as part of adjuvant therapy is associated with improved survival in patients with PASC. Methods: Records of all patients who underwent pancreatic resection at our institution from 1986-2012 were reviewed to identify those with PASC. Multivariable Cox proportional hazards modeling was used to assess for significant associations between patient characteristics and survival. Results: In total, 62 patients with PASC were identified among 5,627 cases (1.1%). Median age was 68 (interquartile range [IQR], 57-77) and 44% were female. Multivariate analysis revealed that among all patients (n=62) the following factors were independently predictive of survival: lack of adjuvant therapy (hazard ratio [HR]=3.558, p=0.0002), positive margin (HR=3.466, p=0.0003), lymph node involvement (HR=3.482, p=0.004), and age (HR=1.030, p=0.035). Among patients who underwent adjuvant therapy (n=39), those who received a platinum agent as part of the adjuvant therapy regimen experienced a longer median survival of 19.1 months (95% CI, 13.8-24.4) compared to 10.7 months (95% CI, 7.9-13.5) for patients who never received a platinum agent (p=0.011, log-rank test). Given this improvement in survival on univariate analysis, a second multivariable model was constructed to elucidate factors associated with survival among patients who received adjuvant therapy in order to confirm the univariate result while controlling for potential confounding risk factors. Backward elimination performed using this second multivariable model revealed that inclusion of a platinum agent in the adjuvant regimen (median survival HR=0.408, p=0.040) and larger tumor diameter (HR=1.259, p=0.047) were independent predictors of survival for this cohort. Conclusions: Addition of a platinum agent to adjuvant regimens for resected PASC should be considered as a means to improve survival among these high risk patients, though collaborative prospective investigation is needed.

Effects of gemcitabine and stereotactic body radiotherapy on quality of life in locally advanced pancreatic cancer

278

Background: Existing literature on the impact of radiation therapy for locally advanced pancreatic cancer (LAPC) on quality of life (QoL) is limited and is specific to standard chemoradiation. We prospectively investigated patient-reported QoL after treatment with fractionated stereotactic body radiation therapy (SBRT). Methods: Forty-nine patients with LAPC treated were prospectively enrolled in a clinical trial at 3 institutions. Participants received a total of 33 Gy in 6.6 Gy daily fractions using SBRT either upfront (N=5) or after a single induction cycle of gemcitabine (N=44), followed by post-SBRT gemcitabine until evidence of disease progression. Two validated questionnaires, the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 and pancreatic cancer-specific QLQ-PAN26, were administered to patients prior to SBRT and at 4-weeks and 3 to 4-months following SBRT. Results: QoL questionnaires were available for 43 patients (88%) in the clinical trial at enrollment, of which 38 (88%) and 22 (51%) also had data at 4-weeks and 3 to 4-months following SBRT, respectively. There was no significant change in global QoL scores from enrollment at 4-weeks (p=0.682) and 4-months (p>0.999) following SBRT. Patients demonstrated a significant improvement in pancreatic pain (p=0.001), body image (p=0.007) and jaundice (p=0.001) scores from the pre-SBRT values at 4 weeks following SBRT. The initial 4-week results were similar when restricted to those patients with both 4-week and 4-month QoL data (p=0.020) and jaundice (p=0.004); however, these symptoms had returned to enrollment levels by the 4-month follow-up. An improvement in body image (p=0.016) and decline in role functioning (p=0.002) were observed in patients with 4-month follow-up QoL questionnaires. Conclusions: LAPC patients undergoing SBRT experience no impairment in global QoL and have short-term improvements in symptoms commonly bothersome to this population including pancreatic pain, body image, and jaundice. Patients also experience a delay in impaired role functioning several months after treatment. These results warrant further investigation of SBRT for LAPC. Clinical trial information: NCT01146054.

Patterns of palliative radiation near the end of life: A single-institution retrospective analysis

9636

Background: The care of patients who receive radiation therapy (RT) at the end of life (EOL) is under scrutiny to ensure effectiveness and value, with many patients not completing RT (Gripp, 2010; Toole, 2012). This retrospective analysis seeks to describe patterns of utilization of palliative RT, including rates of completion of RT offered at the EOL and the use of single fraction RT for bone metastases. Methods: Electronic medical records were used to create a database of 3,383 RT plans for brain, bone, lung, and other metastatic sites in patients treated at Johns Hopkins Hospital from 9/1/2007-7/15/2012. RT plans without palliative intent were excluded. T-tests and logistic regression compared patient and treatment characteristics between patients who died > 1 month versus ≤ 1 month after their last RT fraction. Results: A total of 983 patients were treated to 1,524 sites, with an average of 1.7 RT sites (SD 1.3) per patient. Of these, 872 (89%) patients had complete records and were included in analysis. At the time of analysis, 85% had died. The mean age of 62.1 years (SD 3.4) did not differ statistically based on time from RT to death. Death ≤ 1 month after RT was documented in 215 (24.7%) patients. Compared to patients living > 1 month after RT, patients receiving RT within the last month of life were more likely to be lung (17% versus 9%), less likely to be brain (34% versus 44%), and equally likely to be bone (45% versus 43%) sites. Patients who died ≤ 1 month after completing RT spent on average 5 days (16.6%) of the last month of life receiving RT, with no significant difference by disease site. Conclusions: Most patients receiving palliative RT finish therapy, with 25% dying ≤ 1 month after RT. Single fraction bone RT was relatively uncommon, with no significant difference in the rates of single fraction RT based on time from RT to death. These data provide a framework to match treatment patterns with national guidelines. Additionally, they provide context to model risk of death shortly after RT, which can aid in clinical decision-making. [Table: see text]

Is successful resection following neoadjuvant radiation therapy for borderline resectable pancreatic cancer dependent on improved tumor-vessel relationships?

4057

Background: Margin-negative (R0) surgical resection is the only potentially curative therapy for pancreatic cancer. For patients deemed borderline resectable (BL), neoadjuvant chemoradiotherapy (NRT) increases the likelihood of subsequent R0 resection and improves overall survival. Prognostic factors for achieving resection following NRT have yet to be clearly identified. Methods: Fifty consecutive patients with BL-PDAC evaluated by a multidisciplinary tumor board who received NRT from 2007-2012 were retrospectively identified. Computed tomography (CT) scans pre- and post-radiation and surgical specimens were centrally reviewed. Results: 29 patients underwent resection following NCRT, while 21 remained unresectable. Between the two groups, age, gender, mean RT dose, and proportion of pancreatic head tumors were not significantly different. Smaller tumor volume and lack of the following factors was associated with selection for resection: superior mesenteric/portal vein encasement (p=0.01), superior mesenteric artery involvement (p=0.02), ascites (p=0.01), and questionable/overt metastases (p=0.01). Notably, celiac artery involvement/encasement, common hepatic artery encasement, and percentage change in tumor volume were not significant predictors of resection (all p>>0.05). Interestingly, tumor volume and degree of individual vessel involvement did not significantly change from scans before and after NCRT (all p>>0.05). Median OS was 22.9 vs.13.0 months in resected and unresected patients, respectively (p<0.001). Of resected patients, 93% had negative margins, 28% had positive nodes, 27% demonstrated <10% viable tumor, and 12% had pathologic complete response at surgery. Dpc4 expression was retained in 68% of specimens with viable tumor. Conclusions: Although the apparent radiographic extent of vascular involvement does not change significantly after NRT, subsequent R0 resection rates are high, nodal involvement is low, and outcomes are similar to resected patients who receive adjuvant therapy. Resection attempts should not be deferred solely based on lack of improvement in tumor-vessel interactions.

Efficacy of platinum chemotherapy agents in the adjuvant setting for adenosquamous carcinoma of the pancreas

e15028

Background: Pancreatic adenosquamous carcinoma (PASC) is a rare morphological subtype of pancreatic adenocarcinoma. PASC accounts for only 1-4% of exocrine pancreatic cancers and carries a particularly poor prognosis. Due to the rarity of PASC, studies of therapies specifically targeting this histopathologic entity are exceedingly limited. Based on efficacy exhibited by platinum agents against squamous carcinoma of other body sites, addition of these agents to adjuvant regimens may benefit patients with PASC. This retrospective study was performed among the largest series of patients with PASC identified to date in order to determine whether addition of platinum agents improves survival. Methods: Records of all patients who underwent pancreatic resection at our institution from 1986-2012 were reviewed to identify those with PASC. Demographic, surgical, pathologic, adjuvant therapy, and survival data were collected. Patients were divided into non-platinum (NPG) and platinum (PG) groups based on whether or not they received a platinum agent as part of adjuvant therapy. Results: In total, 62 patients with PASC were identified among 5,627 cases (1.1%). Fourteen patients received a platinum agent in the adjuvant setting (PG), while 48 did not (NPG). These two groups were comparable in regard to median age (65 vs. 69 yrs, p=0.34), gender (36 vs. 46% female, p=0.50), performance status (78 vs. 79% ECOG 0, p=0.98), histologic grade (86 vs. 77% grade 3, p=0.49), positive resection margins (14 vs. 29%, p=0.26), lymph node involvement (71 vs. 79%, p=0.54), median tumor diameter (4.0 vs. 4.3 cm, p=0.64), and proportion receiving radiotherapy (57 vs. 42%, p=0.31). PG patients received a median of 5.5 cycles (IQR, 3.3-6.0) of platinum chemotherapy, with 10 patients (71%) receiving cisplatin-based regimens and 4 (29%) receiving oxaliplatin-based regimens. PG patients experienced significantly longer median survival (19.1 months, 95% CI 12.8-25.4) compared to NPG patients (9.8 months, 95% CI 7.3-12.4) (p=0.024). Conclusions: Addition of a platinum agent to adjuvant regimens for resected PASC may improve survival among these high risk patients, though collaborative prospective investigation is needed.

First report of the correlation of PET Response Criteria in Solid Tumors (PERCIST) criteria and pathologic change in patients with rectal cancer treated with neoadjuvant radiation

261

Background: High-dose-rate endorectal brachytherapy (Endo-HDR) is a promising technique using a short course of radiation therapy in the neoadjuvant setting for rectal cancer. While clinical exam and standard imaging (CT and MRI) suggest outstanding radiation response in these patients, they cannot reliably predict pathologic complete response (pCR) or high tumor regression grade (TRG). In this study, we investigate the utility of PERCIST criteria to predict pathologic change. Methods: We treated nine patients on a prospective clinical trial investigating Endo-HDR as a single agent neoadjuvant therapy (6.5 Gy x 4 fractions, total dose of 26 Gy) for stage II-III rectal cancer patients. A pre-radiation PET/CT scan was compared to a post-treatment PET/CT scan performed one month following the completion of radiation. All nine patients proceeded to surgical resection with the pathologic specimen analyzed for residual disease. TRG (Mandard et al.) was used to quantify response to treatment. A 50% decrease in SUVpeak by PERCIST criteria was correlated with a TRG 1 or 2 (none or minimal residual cancer cells on the final pathologic specimen). Results: As demonstrated in our table, 3/5 (60%) of patients achieving a TRG 1 or 2 had a SUVpeak that decreased by 50% or more; contrarily, zero of 4 (0%) of patients without a TRG 1 or 2 had a SUVpeak of 50% or more. Despite a trend, the small patient sample limited statistical significance (p=0.17). A minimum decrease of 50% in SUVpeak resulted in a positive predictive value of 100%, a negative predictive value of 67%, a specificity of 100%, and a sensitivity of 60% for predicting TRG 1 or 2. Conclusions: Our results demonstrate the first report of utilizing PERCIST criteria in rectal cancer treated with neoadjuvant Endo-HDR. Reliable clinical predictors following Endo-HDR for excellent pathologic response may identify patients in whom a non-operative rectal cancer treatment is possible. [Table: see text]

Prognostic factors for achieving resection following neoadjuvant radiation therapy for borderline resectable pancreatic adenocarcinoma

285

Background: Margin-negative (R0) surgical resection is the only potentially curative therapy for pancreatic cancer. For patients deemed borderline resectable (BL), neoadjuvant chemoradiotherapy (NCRT) increases the likelihood of subsequent R0 resection and improves overall survival. Prognostic factors for achieving resection following NCRT have yet to be clearly identified. Methods: 50 consecutive patients diagnosed with BL pancreatic cancer by a multidisciplinary tumor board from 2008-12 were retrospectively identified. Pre- and post-NCRT CT scans and surgical specimens were centrally reviewed by a blinded radiologist and pathologist, respectively. Results: 29 patients underwent resection following NCRT, while 21 remained unresectable. Between the two groups, age, gender, mean RT dose, and proportion of pancreatic head tumors were not significantly different. Lack of the following factors was favorably associated with resection: SMV/PV encasement (p=0.01), SMA involvement (p=0.02), ascites (p=0.01), and questionable/overt metastases (p=0.01). Notably, celiac artery involvement/encasement, common hepatic artery encasement, and percentage change in tumor volume were not significant predictors of resectability (all p>0.05). Additionally, tumor volume and degree of individual vessel involvement did not significantly change from scans before and after NCRT (all p>0.05). Median OS was 22.9 vs.13.0 months in resected and unresected patients, respectively (p<0.001). Of resected patients, 93% had negative margins, 28% had positive nodes, 27% demonstrated <10% viable tumor, and 12% had pathologic complete response at surgery. Conclusions: Certain radiographic features appear more strongly associated with resectability after NCRT than others. Despite the fact that tumor-vessel interactions do not change significantly due to NCRT, subsequent R0 resection rates are high, nodal involvement is low, nearly 1/3 of patients have minimal residual tumor, and outcomes are improved. Further studies are needed to elucidate novel biomarkers or functional imaging predictors for successful resection following neoadjuvant therapy.

Phase II study of erlotinib combined with adjuvant chemoradiation and chemotherapy for resectable pancreatic cancer

269

Background: Amplification and overexpression of the EGFR gene and surface protein have been described in up to 80% of pancreatic tumors, making EGFR an attractive target in developing new adjuvant therapy for pancreatic adenocarcinoma (PDAC). Inhibition of EGFR has been suggested to provide additive activity when combined with capecitabine and radiation. Here we evaluate the antitumor activity and toxicity profile of erlotinib combined with adjuvant chemoradiotherapy (CRT) and chemotherapy. Methods: 50 patients with resected stage I/II PDAC were enrolled in a phase II trial of adjuvant erlotinib and capecitabine administered concurrently with IMRT (50.4 Gy), followed by 4 cycles of erlotinib and gemcitabine. Results: Median length of follow-up was 18.2 months (IQR, 13.8-27.1). Seventy-nine percent of tumors were of the pancreatic head, 85% had nodal involvement, and 17% had positive margins. Median RFS was 15.6 months (95% CI, 14.1-17.1), local RFS 21.1 months (95% CI, 17.1-25.1), and OS 24.4 months (95% CI, 17.1-31.6). Local recurrence was only observed in 19% patients and synchronous recurrence in 8%. Patients with maximum tumor diameter of ≤3 cm showed superior RFS (17.9 vs. 14.0 months; P=0.049), as did patients with cutaneous reaction to erlotinib (16.3 vs. 9.3 months, P=0.021). Superior OS was associated with less than median (32.3U/mL) pre-CRT CA19-9 values (28.2 vs. 19.0 months, P=0.012). During CRT, 31% patients experienced grade 3 toxicity and 2% grade 4, while 31% patients required a treatment break/stopped treatment early. During post-CRT chemotherapy, 35% patients experienced grade 3 toxicity and 8% grade 4, while 30% required a dose reduction. Conclusions: Results of this phase II trial suggest erlotinib combined with standard adjuvant CRT and chemotherapy provides excellent local disease control and reasonable tolerability compared with existing adjuvant regimens. Patients with maximum tumor diameter of ≤3 cm, cutaneous reaction to erlotinib, and less than median pre-CRT CA19-9 values appear to especially benefit from this new adjuvant regimen. Clinical trial information: NCT00962520.