BACKGROUND: Combination or multi-agent therapy including immune checkpoint inhibitors has shifted the landscape of the treatment of advanced/metastatic renal cell carcinoma. There are several approved immune checkpoint inhibitor (ICI) combinations featuring antibodies against programmed cell death protein 1 (PD-1) receptor or its ligand 1 (PD-L1) combined with other immune checkpoint inhibitors, multi-targeted tyrosine kinase inhibitors (TKIs), or other agents active in renal cell carcinoma. OBJECTIVE: This study aims to compile the evidence of available first-line combination therapies compared to sunitinib monotherapy in advanced renal cell carcinoma. METHODS: A systematic literature search was conducted according to the PRISMA statement to identify all randomized Phase III clinical trial data in previously untreated metastatic renal cell carcinoma featuring an immune checkpoint inhibitor combination compared against sunitinib. A two-stage selection process was utilized to determine eligible studies. Of a total of 124 studies and 94 additional abstracts, 6 studies were considered for final analysis. These studies were evaluated for progression free survival (PFS), overall survival (OS), Grade III or higher adverse events (AEs), objective response rate (ORR), and complete response rate (CRR). RESULTS: 6 studies with 5,121 patients met our search criteria. For OS, ICI combination therapy was favored over sunitinib with an estimated combined hazard ratio of 0.74 (0.67–0.81 95% CI). For PFS, ICI combination therapy was favored over sunitinib with an estimated combined hazard ratio of 0.65 (0.52–0.82, 95% CI). The combination of nivolumab and ipilimumab had the longest duration of response and less incidence of grade III or higher adverse events compared to the combination of anti-PD-1/PD-L1 with TKI. The combination of anti-PD-1/PD-L1 with TKI had higher rates of overall response and longer PFS than the combination of nivolumab/ipilimumab. CONCLUSIONS: This meta-analysis supports the recommendation of immune checkpoint inhibitor combination therapy over sunitinib monotherapy for previously untreated advanced renal cell carcinoma by virtue of improved PFS and OS. The choice of which ICI combination therapy to use may be guided by patient-specific characteristics including IMDC risk status, adverse effect profile, and need for early response.

Rural Appalachia continues to be an underserved region of healthcare disparity created by barriers including access to care and geographic isolation. In particular, cancer screening is well below national averages. We aim to identify factors contributing toward an unfavorable utilization pattern for cervical and colorectal cancer screening. Population based data from across the state of West Virginia was queried through our EHR to identify cervical and colorectal diagnoses from 2009-2020. Patient demographics, cancer stage at diagnosis, and incidence of screening were obtained through diagnosis/procedural codes. Tertiary care centers within the state were used as reference points. Analyses were performed using chi square tests to assess differences in categorical variables and logistic regression was used to identify trends and associations of screening based on distance from the patients' zip code of residence to the tertiary care centers. 194 patients were diagnosed with cervical cancer from 2009-2020. 52% of patients were identified as previously having undergone a screening pap smear prior to their diagnosis of cervical cancer. Age of the patients at diagnosis was <30yo (10%), 30-49yo (40%), and >50yo (50%). Patients who presented at <30yo were significantly more likely to have been screened compared to those ages 30-49 and >50yo (80%, 53%, and 45%, respectively) (p = 0.018). Patients presenting with FIGO stages 1-2 were more likely to have undergone a screening test compared to those presenting at FIGO stages 3-4 (76% vs 30%, respectively) (OR = 7.47, p = 0.002). Median distance from the patients' residence to the closest tertiary care center was 63 (IQR: 36-91) miles. Patients that lived within 71 miles of a tertiary care center were significantly more likely to have undergone a screening test (58%, p = 0.037). 1,965 patients were diagnosed with colorectal cancer from 2009-2020. 46% of patients were identified as having previously undergone a screening colonoscopy prior to diagnosis. Patients whose age at diagnosis was between 50-60yo underwent a screening test (49%) versus 60-70yo (45%), and >70yo (37%). Median distance from the patient's residence to the closest tertiary care center was 56 (IQR: 26-84) miles. Patients who lived further than 56 miles of a tertiary care center were less likely to undergo a screening colonoscopy (OR = 0.55, p<0.001). For each mile distance from a tertiary care center there was a 0.7% drop in likelihood to undergo a screening test. Rural Appalachia has a comparatively low screening rate of cervical and colorectal cancers. We found that patients who live in geographically underserved regions, isolated from tertiary care in the state of West Virginia, had lower overall rates of screening. This information may help to better mobilize resources to improve screening rates and cancer-specific outcomes in the region.

OBJECTIVE: Stereotactic radiosurgery (SRS) can effectively control brain metastasis (BRM) from non-small-cell lung cancer (NSCLC), although intracranial recurrence from untreated micrometastatic tumor deposits is common without whole-brain radiotherapy. Our goal was to determine if immunotherapy improves distant intracranial progression-free survival (DI-PFS) compared with other systemic therapies in patients treated with SRS. METHODS: All patients from 2011 to 2019 treated with SRS without previous whole-brain radiotherapy for NSCLC BRM were reviewed. DI-PFS for the entire cohort, and subgroups of patients, was estimated and compared using the Kaplan-Meier/log-rank method. RESULTS: One hundred and thirty-six SRS sessions used to treat 99 patients were reviewed; 98 (72%) for previously untreated BRM and 38 (28%) for recurrent BRM. 35% received immunotherapy (77% concurrent with SRS), 46% received chemotherapy (75% concurrent), and 18% received epidermal growth factor receptor/anaplastic lymphoma kinase (ALK) targeted therapy (85% concurrent). At median follow-up of 13.7 months, 49% developed distant intracranial recurrence. One-year DI-PFS was improved with any use of immunotherapy (58% vs. 39%; P = 0.03) and concurrent immunotherapy versus chemotherapy or targeted therapy (67% vs. 37% vs. 39%, respectively; P = 0.01). In the immunotherapy cohort, 1-year DI-PFS was improved for programmed death-ligand 1 expression >= 50% versus 1%49% versus 0% (80% vs. 49% vs. 19%, respectively; P< 0.01), and Lung Immune Prognostic Index 0-1 versus 2 (63% vs. 34%; P = 0.03). CONCLUSIONS: Immunotherapy concurrent with SRS, particularly in patients with high programmed death-ligand 1 expression or low Lung Immune Prognostic Index, is associated with improved DI-PFS and no increased risk of radiation necrosis compared with other systemic therapies for NSCLC.

BACKGROUND Early palliative care (PC) physician involvement alongside standard oncologic care has been recommended by the American Society of Clinical Oncology (ASCO) guidelines for all advanced cancer patients, although adherence to these guidelines is variable. Radiation oncologists (ROs) could help facilitate early PC referral for patients treated with palliative radiation, particularly those with brain metastasis (BRM), and the aim of this study was to evaluate the circumstances of PC referral at our institution to better understand the multidisciplinary approaches to facilitate it. METHODS Patients diagnosed with BRM from non-small cell lung cancer (NSCLC) from 2012 to 2018 whose primary RO and MO were at our institution were reviewed. Overall survival and time to PC consultation from the first oncologic visit following BRM diagnosis was determined using the KaplanMeier method. Mann-Whitney U and Chi-Squared assessed for predictive factors for shorter time to PC consultation. For these factors, the overall survival, rate of PC consultation, and PC setting was used to determine utilization of early PC. RESULTS Among 103 eligible patients, only 48% underwent a PC consultation in their lifetime, with the initial evaluation being as an outpatient for 37%, and within 1 month of death for 35%. Median survival from BRM diagnosis was 9.0 months. The median time from oncologic appointment to PC referral was 2.8 months, and from initial PC consultation to death was 1.6 months. Only more recent BRM diagnosis (2016-2018 vs. 2012-2015) was associated with shorter time to PC consultation (1.0 vs. 5.6 months, P=0.013), increased PC consult rate (60% vs. 42%, P=0.105), and increased outpatient PC consultation (56% vs. 26%, P=0.037). CONCLUSIONS The majority of patients did not undergo early PC consultation, though utilization has improved over time. As ROs are commonly involved in BRM management, they may be in a position to proactively support early PC consultations in this patient population.

Guideline-based cancer screening is crucial for early detection and improved patient outcomes. However, there are many barriers to effective cancer screening including insurance coverage, healthcare access, patient education/compliance, and physician non-adherence to established best practices. One solution to these barriers is to implement comprehensive community-based cancer screening events to provide increased access to care. From 2014-2019 we implemented a total of 35 comprehensive cancer screening events across 13 community sites. All screenings were performed by physician volunteers and were free-of-cost. Pre-screening questionnaires were completed to evaluate qualifying screenings, compliance with primary care, and incidence of past screenings. Screening sites included skin (total skin check), head and neck (laryngoscopy), breast (mammogram and exam), cervical (PAP spear), colorectal (FIT test), prostate (PSA and DRE), and lung (low dose CT). A report of results was given to the patient with counseling done at the time of abnormal screening results. Additionally, reports were sent to primary care physicians and appropriate sub-specialty referrals were placed. A total of 4835 cancer screenings were performed on 1972 individual patients. Median age was 65 with 58.7% female and 41.3% male. A total of 1040 (21.5%) of individual screenings had an abnormal result. Abnormal results were highest with skin (40.9%) and lung (36.7%) subsites (P<0.05). Abnormal results were lowest with colorectal (3.1%) and cervical (6.6%) subsites. Patient satisfaction surveys were completed with a majority indicating excellent/good quality of screenings. Large scale community-based cancer screenings are feasible and yield a significant proportion of abnormal results. Further investigation is underway to evaluate follow up compliance, false positive/negative rate, treatment related outcomes, patient satisfaction and cost effectiveness.Tabled 1Abstract 2873; TableScreeningabnormalnormaltotal% abnormalSkin659951161040.9%H&N6341948213.1%Mammogram5315120426.0%Breast exam324364686.8%Cervical436126556.6%Colorectal144324463.1%DRE7831639419.8%PSA4338342610.1%Lung CT559515036.7%Total10403795483521.5% Open table in a new tab

ALKS 4230 is a fusion protein of circularly permuted interleukin-2 (IL-2) & IL-2Rα designed to selectively bind to the intermediate-affinity IL-2 receptor. ALKS 4230 is being investigated as monotherapy & in combination with pembro in pts with solid tumors. We report extended follow-up from combination therapy and new ALKS 4230 monotherapy data. ARTISTRY-1 (NCT02799095) is an ongoing 3-part phase I/II study. In Parts A (dose escalation) and B (expansion), ALKS 4230 is administered as IV monotherapy on days 1-5 of 14- or 21-day cycles. In Part C (combination therapy), ALKS 4230 is administered via the same 5-day regimen q21d with pembro on day 1. Outcomes presented include safety, PK/pharmacodynamics, RP2D, and antitumor activity (RECIST v1.1 & iRECIST) from Parts B & C (response data cut: 4/29/2020; all other data: 3/23/2020). In Part B, 9 pts (5 melanoma, 4 RCC) received 6 μg/kg (RP2D). The most common adverse events (AEs) were grade ≤2 (fevers, chills, hypotension [not requiring vasopressors]). No treatment-related deaths occurred; no AEs led to treatment discontinuation. One pt with metastatic urethral melanoma (which had previously recurred following adjuvant nivolumab) achieved partial response (PR) to 6 μg/kg ALKS 4230 monotherapy with normalization of serum LDH. In Part C, 46 pts (13 tumor types) received ALKS 4230 (3 μg/kg) and pembro. The most common tumor types (≥5 each) were colorectal, ovarian (OC), & sarcoma. Addition of pembro did not alter the PK/pharmacodynamics of ALKS 4230; no new toxicities were observed. Among pts with ≥1 scan, 1 heavily pretreated pt with OC achieved complete response (per RECIST) & continues therapy >12 months; PR was seen in 2 other OC pts (1 pt awaiting confirmatory scan) who received therapy for 7 & 3 months, respectively, both ongoing; iPR was seen in 1 pt with triple-negative breast cancer (on therapy >12 months). All 4 responders were checkpoint inhibitor naive. ALKS 4230 is a promising novel agent given its tolerability & efficacy profile, which includes single agent activity & durable responses even in a pretreated pt population. Future research of mono & combination therapy with ALKS 4230 is warranted.

77 Background: Patients with advanced cancer benefit from early involvement of palliative care. Nonetheless, the ideal method of palliative care integration remains to be determined. Prior studies proposed automatic referral criteria and embedding palliative care teams within specialty clinics. Methods: We studied the impact of an embedded palliative care team that saw patients in an academic oncology clinic based on automatic referral criteria. Patients seen in this clinic on a specific day had access to the “embedded” model, whereas patients seen on two other days could access a separate palliative care clinic upon oncologist referral (usual care). We abstracted data from the medical records of 118 patients who were cared for in this oncology clinic and died during the 3 years following implementation of the embedded model. Results: Compared with those with access to usual care (n = 88), patients with access to the embedded model (n = 30) encountered palliative care as outpatients more often (p < 0.001) and twice as long before death (mean 223 versus 106 days, p = 0.001). Hospice enrollment rates were similar (p = 0.717) but duration was twice as long (mean 53.5 versus 25.3 days, p = 0.03), and enrollment greater than 7 days before death—a core Quality Oncology Practice Initiative metric—was significantly higher in the embedded model (OR 5.60, p = 0.034). Place of death (p = 0.505) and end-of-life chemotherapy (OR 0.361, p = 0.204) did not differ significantly. Conclusions: A model of embedded palliative care with automatic referral criteria, compared with usual practice, was associated with significant improvements in utilization and timing of palliative care and hospice.

e16077 Background: Beginning in 2005 treatment of advanced renal cell carcinoma (RCC) saw a shift from cytokine to targeted therapies. Medication classes introduced in this period included Tyrosine kinase inhibitors, VEGF inhibitors, and mTOR inhibitors. Clinical trials leading to the approval of these medications often excluded non-clear cell histologies. However, retrospective and prospective data have shown there may be a benefit in non-clear cell histology. The aim of this review was to evaluate the impact of targeted therapies on survival with respect to the histologic subtypes. Methods: The Surveillance, Epidemiology, and End Results (SEER) database was used to identify patients with RCC diagnosed in 2000-2004 and 2007-2011. Only patients with distant disease at presentation were included. 12 month cause specific survival (CSS) was then calculated for histologic subtypes within the respective timeframes. Results: 12,239 patients met the above criteria. 5,776 patients were diagnosed from 2000-2004. The median age at diagnosis was 64. 65% were male. Histologic classification, (% of overall cases), and [12 month CSS] are as follows: Renal cell (64.3%) [28.3%], clear cell (27.1%) [52.5%], sarcomatoid (5.1%) [20.5%], papillary (2.1%) [50.3%], chromophobe (0.7%) [58.1%], collecting duct (0.5%) [21.9%], medullary (0.1%) [12.5%]. From 2007-2011 6,463 patients were diagnosed. The median age was 65. 67% were male. Histologic classification, (% of overall cases), and [12 month CSS] are as follows: Renal cell (47.8%) [28.6%], clear cell (41.1%) [57.7%], sarcomatoid (5.8%) [22.2%], papillary (3.5%) [44.7%], chromophobe (0.7%) [58.3%], collecting duct (0.6%) [20.0%], medullary (0.4%) [15.4%]. The only significant effect on 12 month CSS was for clear cell type: 52.5% (95% CI 49.9 – 54.9%) to 57.7% (95% CI 55.8-59.6%). Conclusions: While studies have shown activity of targeted therapies in non-clear cell histologies, only clear cell histology has seen a statistically significant improvement in CSS in this review. Notably, prognosis with regard to histologic subtype varied greatly and this is likely a reflection of a heterogeneous family of tumors which will require a more individualized approach in the future.

Cutaneous adnexal carcinomas (CAC) are rare neoplasms arising from hair follicles and sebaceous, apocrine, and eccrine glands. While in general prognosis is good, these tumors are thought to be locally aggressive. Due to the rare nature of these tumors, treatment paradigms have been adapted from skin cancer data. Tumors are generally treated with surgery, with some high-risk patients receiving adjuvant radiation. However, the role of radiation remains undefined and practice varies by institution. Here we aim to identify cohorts of patients that may be at higher risk and analyze outcomes. The Surveillance, Epidemiology, and End Results (SEER) database was used to identify patients with CAC with skin as the primary site. Patients diagnosed between 1998 and 2010 were included. Patients with known metastatic disease were excluded. Five-year cause-specific survival (CSS) was then calculated for multiple variables, including tumor size, nodal status, subsite, age, sex, race, and grade. Grade was divided into low grade (I and II) and high grade (III and IV). Frequency of radiation use was evaluated with regard to risk factors. A total of 3551 patients who met the above criteria were identified. The median age was 68 years, and 54.4% were male. Ethnicity was as follows: white 81.2%, black 6.1%, other 5.1%, and unknown 7.5%. Median follow-up was not reached. Grade, nodal status, and age were found to have statistically significant (11.4%, 23.0%, 4.6%, P<0.05) impacts on 5-year CSS. Patients with low- and high-grade disease were found to have 5-year CSS of 97.3% and 85.9%. Those who were node negative vs positive at presentation were noted to have 5-year CSS of 97.1% vs 74.1%. Last, patients stratified by age (70 years as cutoff) had 5-year CSS of 97.6% (<70) and 93.0% (≥70). In all, 7% of well-differentiated tumors and 18% of poorly differentiated tumors received radiation. Node-negative and node-positive patients received radiation in 5% and 58% of cases. While overall prognosis is good, we identified variables that were associated with a statistically inferior 5-year CSS. Nodal involvement and high tumor grade were most predictive, as noted in Table 1. Patients with these risk factors may derive benefit from comprehensive postoperative radiation as these tumors tend to be locoregionally aggressive. Despite its potential therapeutic value, our study shows radiation has been underutilized in these high-risk groups. While prospective data is the best way to confirm or refute our findings, it is always a major clinical challenge to conduct such a study in this rare set of tumors. Nevertheless, we need more data on other potential risk factors such as perineural invasion, extracapsular extension, and lymphovascular invasion.

Merkel cell carcinoma (MCC) is a rare neuroendocrine malignancy of the skin with a predilection for aggressive behavior. Previous studies evaluating sentinel lymph node biopsy (SLNB) in MCC have shown positivity rates typically ranging from 22% to 48% in clinically N0 patients. The benefit of SLNB has also been well established in other similarly aggressive tumors, such as melanoma. As a result, sentinel lymph node biopsy (SLNB) is currently recommended for all clinically N0 patients, though its impact remains unclear. The aim of this review was to evaluate for a survival benefit in clinically N0 patients undergoing SLNB. The Surveillance, Epidemiology, and End Results (SEER) database was used to identify patients diagnosed with MCC from 2003 through 2012 with skin as the primary site. Patients up to age 75 years were included, provided they underwent a cancer-directed surgery. Patients with positive lymph nodes or distant disease were excluded. Cause-specific survival was then compared in patients who underwent SLNB as opposed to no pathologic nodal evaluation. A total of 603 patients were identified who met the above criteria. Patients undergoing SLNB had a median age of 65 years with a range of 38-75, 59% were male, and 95% were white. With regard to tumor size: 57% were ≤2 cm, 13% were 2-5 cm, and 30% were >5 cm. In all, 52% of patients received radiation. Patients with no pathologic nodal evaluation had a median age of 68 years with a range of 37-75, 60% were male, and 94% were white. With regard to tumor size, 34% were ≤2 cm, 19% were 2-5 cm, and 47% were >5 cm. A total of 48% of patients received radiation. Patients undergoing SLNB had an improved 5-year CSS of 91.3% (95% CI 86.4-94.5%) compared to the nonpathologic nodal evaluation group, with a 5-year CSS of 72.5% (95% CI 65.4-78.4%). When analyzing tumors ≤2cm, the improved outcome was maintained for SLNB patients vs nonpathologic the nodal evaluation group with a 5-year CSS of 94.70% (95% CI 88.4-97.6) compared to 79.3% (95% CI 66.6-87.6%). A CSS advantage was found for patients undergoing SLNB. This advantage remained significant when stratifying for tumors <2 cm. The reason for this advantage is not fully clear but could be in part from decreased false negatives for nodal staging, therapeutic effect of the SLNB, or a result of more complete care. While prospective data are needed to confirm these findings, the rare nature of these tumors renders this a clinical challenge. Regardless, our current findings support the standard of care of SLNB in clinically N0 patients.

A current standard of care treatment for locally advanced but clinically node negative bladder cancer is radical cystectomy with neoadjuvant chemotherapy. Locoregional failure following cystectomy however remains a significant source of morbidity, and is associated with adverse pathologic features including positive surgical margins (SM+). The selection criteria for postoperative radiation therapy (PORT) and clinical effectiveness with PORT have not been well-established. The objective of this study is to evaluate clinical and pathologic features, and associated outcomes, in patients who received PORT for pT3-T4, pN0-1 M0 muscle invasive bladder cancer. The patients chosen for this retrospective review were compiled via the cancer registry at Penn State Hershey Medical Center from 1995-2015. Sixty-three patients were identified as having non-metastatic muscle invasive bladder cancer on initial presentation, who underwent cystectomy, and staged as pT3-pT4, pN0-N1. Analysis was done on a cohort of patients that received PORT with a cohort that did not. Ten patients received PORT using IMRT, with one of the patients unable to complete the entire course due to toxicity, and 53 patients did not. Radiation dose ranged from 34.2-58 Gy. The PORT group had 70% urothelial histology and 60% had SM+ compared to 92% and 6%, respectively, in the non-PORT group [both statistically significant (P = 0.038 and P = <0.001 respectively)]. The entire cohort local recurrence (LR) rate was 28%. The PORT group had a 40% LR rate with the majority of these cases being gastrointestinal in location. The overall distant recurrence rate was 25% with no significance between the groups. The entire cohort 5-yr disease free survival (DFS) rate was 30%, and the 5-yr overall survival (OS) rate was 41%. The 5-year DFS rates were 20% in the PORT group versus 32% in the non-PORT group [not statistically significant (hazard ratio [HR] 1.498, 95% confidence interval [CI] 0.539-4.1663, log-rank P = 0.361)]. The 5-yr OS rates were 30% in the PORT group versus 43% in the non-PORT group [not statistically significant ([HR] 0.863, [CI] 0.357-2.089, log-rank P = 0.75)]. Multivariate analysis showed that SM+ was significant for 5-yr DFS, while nodal status and stage were significant for 5-yr OS. Post-cystectomy local recurrence rates were high despite the use of PORT in patients with high risk features. When considering a 5-yr OS of about 40%, reducing pelvic failure remains a critical goal. Studies evaluating the clinical target volume, radiation dose, and patterns of failure should help refine the selection criteria of patients most likely to benefit from PORT.

137 Background: Early palliative care (PC) improves quality of life (QOL) and enhances end-of-life (EOL) care, but the optimal timing and most effective model for integrating PC into oncologic care is uncertain. To understand the impact of an integrated model with PC providers embedded with oncologists vs. usual care (UC) with referral at the discretion of the same oncologists, we examined the timing and delivery of PC and Quality Oncology Practice Initiative (QPOI) EOL metrics among patients with RCC and M in a single clinic. We hypothesized that integrated PC would result in more referrals, earlier contact with PC and better QOPI EOL metrics compared with UC. METHODS In a retrospective cohort study of patients with RCC and M in the Beth Israel Deaconess Biologics Clinic who expired between 10/1/12 and 12/31/14, we compared patients seen 2 days/week, when referral to PC was discretionary, with a third day when PC providers shared the clinic for real-time consultations. Patients were identified as meeting PC eligibility if they had recurrent, metastatic disease and were on active treatment or had a symptom severity of 7+ on Edmonton Symptom Assessment Scale (ESAS). Two oncologists saw all patients, regardless of day. RESULTS Seventy-six patients expired, 19 in the Integrated PC model and 57 with UC. Patients were similar with respect to diagnosis and demographics except for smoking. The integrated model substantially improved timing and location of PC. In the integrated PC model, 85% were seen by PC compared with 45% in UC (P = 0.002). All patients in the integrated model began PC as an outpatient compared with 36% in UC (P < 0.001). The mean number of days from first PC contact to death was 28 (SD = 54) for UC and 118 (SD = 120) with integrated PC (P < 0.001). The location of death did not differ significantly between models, occurring outside the hospital with hospice among 71% of patients in the integrated model and 53% in UC (P = 0.25). Results were similar in relative risk models adjusted for smoking. CONCLUSIONS A practice model that integrated PC with oncologic care was associated with more PC referrals, earlier contact, and a nonsignificant trend toward fewer deaths in hospital and ICU.

An improved understanding of kidney cancer (KC) tumor biology has led to major advancements in the treatment of patients with metastatic disease. While agents that target the VEGF and mTOR pathways prolong survival, resistance develops for most patients within the first year of therapy. Agents that lead to durable remissions are of urgent need to patients living with this disease. For two decades, the clinical experience with high-dose bolus IL-2 has served as proof of principle that immunotherapy can produce durable responses in a small percentage of KC patients. Basic and translational investigation to elucidate the molecular mechanisms that govern the interaction between a tumor and its host immune response has helped to explain why immunotherapies too often fail to achieve satisfactory results. In KC, obstacles to effective immunotherapy may include the physiological down-modulation of the immune response through the expression of programmed death ligand-1 (PD-L1) on tumor cells and cytotoxic T-lymphocyte antigen-4 (CTLA-4) on activated T cells, which serve to restrict the cytolytic function of tumor-infiltrating T lymphocytes, the proliferation of regulatory (CD4+ CD25+) T cells (T-regs) in response to nonspecific cytokine administration, and the immunosuppressive effects of elevated circulating VEGF levels and myeloid-derived suppressor cells (MDSC). These insights have encouraged investigators to pursue agents that block T-cell regulation (e.g., PD-1 and CTLA-4 antibodies), inhibit tumor-induced immunosuppression (e.g. PD-L1 antibody) and more specifically activate T-cells (eg, CD-137 antibody, IL-21), and dendritic cells (e.g. AGS-003). Several of these approaches have shown encouraging efficacy in early trials both as single agents and in combination with standard therapies. However, for “targeted immunotherapy” to reach its full potential, significant work will need to be done to improve patient selection by refining proposed predictive biomarkers and overcome resistance mechanisms through the development of combination regimens.

56 Background: Early palliative care (PC) has been shown to improve quality of life (QOL), enhance quality end-of-life (EOL) care, and reduce costs, yet many cancer centers lack resources to provide outpatient palliative care services. Data is needed to identify the optimal timing and most effective and efficient model for integrating PC into the care of cancer patients. In order to understand what components of PC have the greatest impact on patients, we examined the acceptability of early PC among patients with kidney cancer (RCC) and melanoma (M) and describe the content of PC visits. Methods: In July 2013, the outpatient PC team at Beth Israel Deaconess Medical Center, including physician (MD), social worker (SW) and chaplain (chap) were invited to see patients within a clinic that specializes in seeing patients with advanced RCC and M. All patients completed data on symptom burden using the Edmonton Symptom Assessment Scale, measures of QOL and degree of psycho-social and spiritual support. Referral to PC was based on these metrics and physician discretion. Results: Of the 21 patients seen by PC, 57% had RCC; 76% male; 86% White; 57% married. Mean age was 62 (range: 36-87). At the 1st PC visit: All had locally adv/metastatic disease; 48% were being treated with curative intent; 29% had not yet started treatment; 57% rated pain 0 on 0-10 scale (range: 0-9). First PC visit content: 76% psychosocial support (including building rapport); 67% symptom management; 33% included advance care planning (ACP) and decision-making support. Ninety-one percent were seen >1 by PC; 19% seen by SW or chap; median PC visits: 3. Eighty-one percent completed health care proxy. Additional data on patient outcomes will be presented. Conclusions: Nearly half of patients seen by PC were being treated with curative intent, suggesting acceptability of early PC integration in patients with RCC and M. Symptom management and psychosocial support dominated the early PC visits.

Case reports and animal studies suggest combining radiation and immunotherapy may generate responses outside the radiation field. We sought to explore this phenomenon in a systematic fashion by evaluating melanoma patients with brain metastases who were treated with both radiation and ipilimumab. We retrospectively reviewed the records of sixteen consecutive patients who received ipilimumab and 51 courses of brain-directed radiation at a single institution from 2008 to 2013. We examined responses of the largest "index" extra-cranial lesion after 31 of these courses where imaging was available before and after treatment and compared this response to two consecutive studies performed prior to radiation therapy when available. Overall survival was calculated from the last day of the first brain-directed radiation therapy course using the Kaplan-Meier method. Responses before and after radiation therapy were compared using McNemar's and the binomial tests for clustered data. Two-group exact tests for clustered data were used to explore the impact of treatment timing on response. Median survival after initial radiation was 14 months, with maximum survival of 50 months. After radiation therapy, the largest index extra-cranial lesion shrank in 35% of 31 instances with sequential imaging as compared to a 17% favorable response rate in the 25 instances with two scans prior to radiation (p = 0.20). There was also a 2.8-fold increased likelihood that the kinetics of index lesion response improved following radiation therapy that did not reach statistical significance (p = 0.07). Index lesions were more likely to respond when ipilimumab was administered within three months of radiation (p<0.01). Historical series of melanoma patients with brain metastases describe median survival of less than 6 months, and less than 4 months in patients treated with radiation therapy without surgery for these lesions. Our data suggest patients treated with ipilimumab and radiation therapy may fare better than these historical controls. There was a trend for favorable systemic response following radiation therapy worthy of further evaluation in studies powered to detect potential synergies between radiation and immunotherapy.

Purpose: This phase II study assessed the antitumor activity of tremelimumab, a fully human, anti–CTL-associated antigen 4 monoclonal antibody, in patients with melanoma. Experimental Design: Patients with refractory/relapsed melanoma received 15 mg/kg tremelimumab every 90 days. After 4 doses, patients with tumor response or stable disease were eligible to receive ≤4 additional doses. Primary endpoint was best overall tumor response assessed by an independent endpoint review committee, and secondary endpoints included duration of response, overall survival, progression-free survival, and safety. Results: Of 251 patients enrolled, 246 (241 response-evaluable) received tremelimumab. Objective response rate was 6.6% (16 partial responses); duration of response was 8.9 to 29.8 months. Eight (50%) objective responses occurred in patients with stage IV M1c disease, and 11 (69%) were ongoing at last tumor assessment. Eight (3.3%) patients achieved responses in target lesions (Response Evaluation Criteria in Solid Tumors) despite progressive disease within the first cycle. All 8 survived for >20 months; 5 (63%) remained alive. Clinical benefit rate (overall response + stable disease) was 21% (16 partial responses and 35 stable disease), and median overall survival was 10.0 months. Progression-free survival at 6 months was 15%, and survival was 40.3% at 12 months and 22% at 24 months. Common treatment-related adverse events were generally mild to moderate, and grade 3/4 adverse events included diarrhea (n = 28, 11%), fatigue (n = 6, 2%), and colitis (n = 9, 4%). There were 2 (0.8%) treatment-related deaths. Conclusions: Tremelimumab showed an objective response rate of 6.6%, with all responses being durable ≥170 days since enrollment, suggesting a potential role for tremelimumab in melanoma. Clin Cancer Res; 16(3); 1042–8

16071 Background: Sunitinib and other antiangiogenesis inhibitors have emerged from clinical trials as promising therapies for mRCC. However, patients in “real-world” clinical practice may differ from those in trials, potentially leading to differences in efficacy. The goal of this study is to describe baseline characteristics and treatment efficacy among patients receiving sunitinib outside of clinical trials. Methods: A retrospective review of medical records was performed in 2 large tertiary oncology centers. Patients were included in the study if they were ≥ 18 years of age, had a histological diagnosis of mRCC, and received ≥ 1 prescription for sunitinib. Trial patients were excluded. Descriptive analyses were conducted to characterize patients by various demographic and clinical factors as of sunitinib initiation. Objective response rate (ORR) was assessed based on RECIST criteria (confirmed or unconfirmed by a 2nd assessment). Patient observation was truncated at 15 months to mimic published phase 3 trial results. Results: Preliminary results are presented below. The median duration of sunitinib treatment was 5.4 months (range 2.0–15.0). Brain metastasis, ECOG ≥ 2, and prior cytokines, all exclusion criteria in the phase 3 trial, were observed in the clinical practice and were associated with a lower ORR. The overall ORR (23.1%, 95% CI=9.9–36.3%) is 25% less than the 31% ORR reported in the phase 3 trial. (Motzer, N Engl J Med, 2007) ORR varied considerably across factors, suggesting that some subgroups may not respond well to sunitinib. Data collection is ongoing; updated results will be presented. Conclusions: Patients receiving sunitinib for mRCC outside of trials may have a different risk factor profile and lower ORR than trial patients, suggesting that clinical trial results may not be generalizable. No of patients (%) ORR % All patients 39 23.1 Age, yrs 30-59 21 (53.9) 28.6 60-89 18 (46.2) 16.7 Gender Female 11 (28.2) 36.4 Male 28 (71.8) 17.9 Number of metastatic sites 1 11 (28.2) 27.3 2 13 (33.3) 23.1 ≥ 3 15 (38.5) 20.0 Bone Yes 11 (28.2) 18.2 No 28 (71.8) 25.0 Brain Yes 8 (20.5) 12.5 No 31 (79.5) 25.8 Liver Yes 4 (10.3) 50.0 No 35 (89.7) 20.0 Lung Yes 24 (61.5) 16.7 No 15 (38.5) 33.3 Lymph nodes Yes 18 (46.2) 22.2 No 21 (53.9) 23.8 ECOG performance status 0 17 (43.6) 35.3 1 12 (30.8) 16.7 2 3 (7.7) 0.0 Not available 7 (18.0) 14.3 Prior nephrectomy Yes 35 (89.7) 22.9 No 4 (10.3) 25.0 Prior cytokine therapy Yes 7 (18.0) 14.3 No 32 (82.1) 25.0 Prior radiation therapy Yes 7 (18.0) 14.3 No 32 (82.1) 25.0 Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration GlaxoSmithKline Genentech, Novartis, Onyx/Bayer, Pfizer GlaxoSmithKline Pfizer Genentech, GlaxoSmithKline, Novartis, Onyx/Bayer, Pfizer, Wyeth