Hematocrit control and thrombotic risk in patients with polycythemia vera treated with ruxolitinib in clinical practice

Polycythemia vera (PV) is a myeloproliferative neoplasm characterized by unregulated red blood cell production resulting in elevated hemoglobin and/or hematocrit levels. Patients often have symptoms such as fatigue, pruritus, and painful splenomegaly, but are also at risk of thrombosis, both venous and arterial. Ruxolitinib, a selective Janus kinase inhibitor, is approved by the US Food and Drug Administration as second-line cytoreductive treatment after intolerance or inadequate response to hydroxyurea. Although ruxolitinib has been widely used in this setting, limited data exist in the literature on ruxolitinib treatment patterns and outcomes among patients with PV in routine clinical practice. We report a retrospective, observational, cohort study of patients treated for PV with ruxolitinib across three US centers (academic and regional practice) from December 2014-December 2019. The study included 69 patients, with a median follow-up duration of 3.7 years (95% CI, 2.9-4.4). Our data demonstrate very high rates of hematocrit control (88% of patients by three months and 89% by six months); few patients required dose adjustments or suspension. No arterial thromboses were observed; however, the follow-up duration does not allow for the generation of meaningful conclusions from this. Three patients had thrombotic events; one was in the setting of a second malignancy, one post-operative, and a third related to prolonged immobility. We also found that 28% of patients initiated ruxolitinib as a result of poorly controlled platelet counts, second only to hydroxyurea intolerance (46%) as a reason to start therapy. In clinical practice, ruxolitinib continues to be effective in controlling hematocrit levels after three and six months of treatment in patients and is associated with low thrombotic risk.

Young Age as a Predictor of Chemotherapy Recommendation and Treatment in Breast Cancer: A National Cancer Database Study

INTRODUCTION

Breast cancer, although the second most common malignancy in women in the United States, is rare in patients under the age of 40 y. However, this young patient population has high recurrence and mortality rates, with chemotherapy frequently used as adjuvant treatment. We aimed to determine whether age is an independent predictor of chemotherapy recommendation and subsequent treatment and the relationship to Oncotype Dx (ODX) recurrence score (RS).

METHODS

The National Cancer Database was retrospectively reviewed from 2010-2016 to identify women with early-stage (pT1-pT3, pN0-pN1mic, M0), hormone receptor positive, human epidermal growth factor receptor 2 negative breast cancer who underwent ODX RS testing.

RESULTS

Of 95,382 patients who met the inclusion criteria, risk groups using the traditional ODX RS cutoffs were 59% low, 33% intermediate, and 8% high. Using Trial Assigning Individualized Options for Treatment RS cutoffs, risk groups were 23% low, 62% intermediate, and 15% high. Chemotherapy recommendation decreased as age at diagnosis increased (P < 0.001). Increasing age was associated with decreased odds of chemotherapy recommendation in univariate models both continuously (odds ratio: 0.98, 95% confidence interval 0.97-0.98; P < 0.001) and categorically by decade (P < 0.001). Age by decade remained an independent prognosticator of chemotherapy recommendation (P < 0.001), adjusted for risk groups.

CONCLUSIONS

Chemotherapy recommendation and treatment differs by age among patients with early-stage hormone receptor positive breast cancer who undergo ODX testing. While molecular profiling has been shown to accurately predict the benefit of chemotherapy, younger age at diagnosis is a risk factor for discordant use of ODX RS for treatment strategies in breast cancer; with patients aged 18-39 disproportionately affected.

Opioid Monitoring Using Urine Toxicology Screens in Outpatient Oncology Palliative Medicine

PURPOSE

There is a paucity of real-world data on opioid screening and urine toxicology testing in outpatient oncology palliative medicine.

METHODS

This was a retrospective analysis of adult patients with cancer completing ≥ one outpatient palliative medicine visit and the Edmonton Symptom Assessment Scale (ESAS). Patient demographics, the Screener and Opioid Assessment for Patients with Pain-Short Form (SOAPP-SF), ESAS, medications, and urine toxicology screens (UTSs) were collected at baseline and follow-up visits. The primary end point was the frequency and type(s) of noncompliant UTSs (ie, presence of a nonprescribed substance or absence of a prescribed substance). Secondarily, risk factors for noncompliant UTSs were evaluated using univariate and multivariable logistic regression.

RESULTS

Of 189 evaluable patients (632 clinic visits), 113 underwent ≥one UTSs, 125 SOAPP-SF, and 75 had both. The median age was 56 (range, 26-80) years, 56% were female, 58% were White, 40% were Black, 48% had stage IV disease, the median baseline pain score was 7, and the median SOAPP-SF was 3. Oxycodone was the most prescribed drug (n = 125). Of 113 patients who underwent UTSs, 54% (n = 61) had ≥one noncompliant result. Thirty-nine percent (n = 44) had a total of 128 noncompliant results for the presence of a nonprescribed substance; 29% (n = 33) had a total of 53 noncompliant results for the absence of a prescribed substance. SOAPP-SF Q4 (use of illegal drugs) (odds ratio [OR], 3.61; 95% CI, 1.81 to 7.19; P < .001) and prescription with nonopioid adjuvant medications (OR, 2.83; 95% CI, 1.12 to 7.19; P = .029) were associated with increased odds of a noncompliant UTS.

CONCLUSION

More than half of the tested population had noncompliant UTS. Screening and evaluating risk factors for nonmedical opioid use is critical in oncology palliative medicine.

Chaplains and telechaplaincy: best practices, strengths, weaknesses-a national study

Telechaplaincy is the use of telecommunications and virtual technology to deliver religious/spiritual care. It has been used for decades, but chaplains' understanding of telehealth lags behind other disciplines. The purpose of this study was to describe the use of telechaplaincy in the United States and chaplains' perceptions of the practice. Researchers surveyed chaplains through chaplain-certifying-body email-listservs, then conducted in-depth interviews with 36 participants identified through maximum variation sampling. Quantitative analysis and qualitative, thematic analysis were conducted. Quantitative results show that in 2019, approximately half of surveyed chaplains performed telechaplaincy. Rural chaplains were more likely to have practiced. Chaplains who had not practiced were more willing to try if they believed it was effective at meeting religious/spiritual needs. Qualitative findings describe chaplains' perceptions of strengths, weaknesses, and best practices.

Positive impact of oncology nurse navigation based on health insurance in head and neck cancer patients

e18545

Background: Health disparity in squamous cell carcinoma of the head and neck (SCCHN) is well-recognized. Privately insured SCCHN patients are observed to have better outcomes, but underlying factors are poorly understood. Oncology nurse navigators (ONN) are a well-established role focused on optimizing cancer care and patient experience. The effects of ONN on bridging health disparities have not been demonstrated. We sought to characterize health and outcome disparities in patients diagnosed with oropharyngeal cancers and to assess if the implementation of ONN bridges gaps in health disparities. Methods: Patients aged >18 diagnosed with node positive, non-metastatic oropharyngeal cancers at a large hybrid academic-community cancer center between June 2018 and September 2021 were analyzed retrospectively. Baseline demographic, socioeconomic and clinicopathologic factors including the frequency and percentage of patients receiving ONN were summarized. Clinical outcomes were assessed as a function of baseline factors and ONN. Overall survival (OS) and progression free survival (PFS) were evaluated with Kaplan Meier methods. Results: 184 patients were evaluated. Median age at diagnosis 62 (range 33-91). Median follow-up 24 months. The majority were HPV+ (73%). Navigation received in 64%. More males presented with cN1+ disease than females (57% v. 37%; p = 0.053). More Black patients presented with cN1+ disease than White patients (76% v. 50%; p = 0.017). More privately insured patients (92%) went to imaging after definitive treatment versus those with Medicaid (MD)/Self-pay (SP) (78%) or Medicare (MC)/Veteran’s affairs (VA) (78%) (p = 0.043), while more navigated patients went to imaging after definitive treatment than non-navigated patients (70% v. 40%; p = 0.003). Univariate modeling indicated clinical nodal stage, HPV status, race, and insurance status were associated with OS and PFS. In multivariable modeling adjusted for age, nodal stage, and HPV status, there was a disparity in OS associated with insurance type (MD/SP v. Private: HR 9.24, 95% CI = 2.34 to 36.41 and MC/VA; p = 0.004). There was a significant interaction of navigation and insurance type (p = 0.0157). In those not navigated, a difference in hazards exists between MD/SP or MC/VA compared with private insurance. This difference was not detected in those navigated. Similar interaction of navigation and insurance was noted in the analysis of PFS. Conclusions: We observed a disparity in patients with non-private insurance having an inferior PFS, OS and reduced rates of imaging after definitive treatment. An improved rate of post-treatment imaging was significantly associated with ONN. There was a trend towards benefit in outcomes for those receiving ONN but without private insurance suggesting that ONN may help mitigate disparities in the non-private insurance groups.

The effect of quality improvement interventions on inpatient cancer malnutrition documentation and coding in an academic medical center

38

Background: Malnutrition (MN) is common yet underdiagnosed in hospitalized cancer patients. Effective assessments can identify those who need nutritional care and help plan intervention. We examined the effect of quality improvement (QI) interventions on the dietitian documented MN (DDMN) and physician coded malnutrition (PCMN). We also determined if the registered dietitian (RD) and physician assessments of MN agreed. Methods: Electronic medical records (EMR) were reviewed for a consecutive cohort of inpatients with a solid tumor diagnosis staged I-IV and admitted to Atrium Health’s Carolinas Medical Center at least once between 1/1/2016 to 5/31/2019. Data were collected from the first admission EMR encounter closest to the cancer diagnosis date. RD assessments were reviewed for DDMN. PCMN diagnosis was based on MN ICD-10 codes in the discharge summary. MN was graded as mild, moderate, and severe. Two QI interventions were implemented during the study period: 1) 8/2016: RD message via EMR to query MD approval for MN diagnosis; 2) 4/2018: Clinical Documentation Integrity Team query MD by sending ASPEN criteria via an alert integrated into MD workflow. Agreement in MN identification was defined as the absence or presence of both DDMN and PCMN; agreement in severity was defined as the absence of DDMN and PCMN or the agreement in presence and severity of DDMN and PCMN. Cochran-Armitage tests for trend assessed prevalence and agreement across the three periods (N1=652; N2=2858; N3=1622) defined by the two sequential QI interventions. Results: N=5143; 52% males. Median age 63 (range 18-102) years. 70% White; 24% Black, 3% Latino. Commonest cancer diagnostic groups: Upper Gastrointestinal 22%, Thoracic (19%), Genitourinary 18%. 28% had stage IV disease. 11% (N=557) met criteria for DDMN and/or PCMN. Of the 557, 40% (N=223) met criteria for both DDMN and PCMN. DDMN (N=420) was mild 2%, moderate 19%, and severe 79%. On discharge, PCMN (N=360) was mild in 10%, moderate in 21%, and severe in 69%. The RD and MD agreed on the presence or absence (94%) and severity (93%) of MN. Significant trends were observed as DDMN prevalence increased from 3.1%, 8.1%, to 10.3% (p<.001), and PCMN prevalence from 0.5%, 7.8%, to 8.2% (p<.001). While rates of mild, moderate, and severe MN varied across the periods, statistically significant change in these distributions was not identified in DDMN (p=0.62) or PCMN (p=0.20) after the second QI intervention. Conclusions: MN was under-diagnosed compared to nutrition intervention studies. When MN was identified, it was moderate or severe in the majority. Evaluations by RD and MD were highly congruent for MN prevalence and severity. Implementation of nutrition-focused QI interventions improved documentation and coding of MN. Improved communication between the RD and the MD could improve the recognition and diagnosis of MN.

Pilot study of multi-gene pharmacogenetic testing for pain management in oncology palliative medicine

Aim: We evaluated the application and clinical impact of multi-gene pharmacogenetic testing in oncology palliative medicine. Patients & methods: In a single-arm pilot trial, cancer patients with uncontrolled pain were assessed in a palliative medicine clinic at baseline and received pharmacogenetic testing. Results were used as applicable up to the final visit (day 30). Pain scores, opioid prescribing, and use of pharmacogenetic test results were collected. Results: In 75 patients, the median baseline pain score was 7/10. Of 54 evaluable at the final visit, 28 required opioid modifications and 19 had actionable genotypes, mostly CYP2D6. Pain improvement (≥2-point reduction) was higher than historical data (56 vs 30%; p < 0.001). There were no differences in pain improvement between those with and without actionable genotypes (61 vs 53%). Conclusion: Multi-gene testing identified actionable genotypes and may improve cancer pain.

Equal access to care and nurse navigation leads to equitable outcomes for minorities with aggressive large B-cell lymphoma

BACKGROUND

Aggressive large B-cell lymphomas (LBCLs) are curable, but previous studies have shown inferior outcomes in minorities. Nurse navigation programs can improve patient outcomes by providing patient support. This study presents the outcomes of White and minority patients with aggressive LBCL at an institution with an active nurse navigation program.

METHODS

The authors prospectively collected baseline characteristics, treatment regimens, and outcome data for patients with aggressive LBCL. Navigation encounters were characterized as low or high intensity. Overall survival (OS) and progression-free survival (PFS) were calculated with Kaplan-Meier methods. Baseline characteristics were compared with Fisher exact tests.

RESULTS

Two hundred four consecutive patients (47 minority patients and 157 White patients) were included. Results were presented as minorities versus Whites. There were no differences in prognostic scores (Revised International Prognostic Index score of 3-5, 43% vs 47%; P = .50), frontline chemotherapy (98% vs 96%; P = .68), or the incidence of relapsed/refractory disease (40% vs 38%; P = .74). For relapsed/refractory LBCL, similar proportions of patients underwent hematopoietic stem cell transplantation (32% vs 29%; P > .99) or chimeric antigen receptor T-cell therapy (16% vs 19%; P > .99). Enrollment in clinical trials was comparable (17% vs 14%; P = .64). More than 85% received nurse navigation, but minorities had higher intensity navigation encounters (42% vs 21%; P = .01). The 2-year OS rates were 81% and 76% for minorities and Whites, respectively (P = .27); the 2-year PFS rates were 62% and 65%, respectively (P = .78).

CONCLUSIONS

This study shows similar survival between Whites and minorities with aggressive LBCL, which was likely due to equal access to guideline-concordant therapy. Minorities received higher intensity navigation encounters, which may have helped them to overcome socioeconomic disadvantages.

Clinician's Experience and Self-Perceived Knowledge and Attitudes toward LGBTQ + Health Topics

Phenomenon: Research on clinicians' knowledge and attitudes toward lesbian, gay, bisexual, transgender, queer/questioning, and other sexual and gender minorities (LGBTQ+) health topics has relied mostly on small early-career samples. The influence of clinical experience on knowledge and attitudes has not been examined. The study purpose was to examine physicians' and advanced practice providers' (a) self-perceived knowledge and attitudes about LGBTQ + health topics, (b) the relationship between clinical experience and self-perceived knowledge and attitudes, and (c) preferences for LGBTQ + medical education.Approach: An online anonymous survey was emailed to 3667 physicians and advanced practice providers at a large southeastern multisite healthcare institution. Logistic regression determined associations between clinical experience, knowledge, and attitudes.Findings: There were 880 (24.0%) respondents. Most were physicians (70%). Thirty-eight percent had more than 15 years of experience. Fifty-four percent preferred online education. Few reported sophisticated knowledge of six LGBTQ + health topics (6%-10%). The majority felt that these were either important or very important for all physicians. After accounting for demographics, experience was not associated with knowledge or attitudes.Insights: Knowledge gaps yet positive attitudes suggest that additional training on LGBTQ + patients is warranted and likely to be well received. The lack of influence of experience on knowledge and attitudes may suggest that training is applicable to clinicians of all experience levels. Future research should determine if knowledge and attitudes toward LGBTQ + health topics influence culturally competent healthcare practices.

Overall and progression-free survival in young breast cancer patients with low muscle mass and density

12052

Background: Low muscle mass (skeletal muscle index, SMI) and density (skeletal muscle density, SMD) are associated with chemotherapy toxicity and shorter survival in women with breast cancer. This has not been studied specifically in women ≤ 40 years at diagnosis. They have different body compositions and more aggressive cancers than older women. We compared pre-treatment muscle measures in survivors and non-survivors and investigated their association with overall survival (OS) and progression-free survival (PFS). Methods: This case-control study included 112 women aged ≤ 40 years at diagnosis. Women with pre-treatment CT scans from 2009-2018 were identified; non-survivors were matched with survivors by age, year of diagnosis and disease characteristics. Body composition was determined by CT analysis. Measures were compared between the groups using Kruskal-Wallis tests. Kaplan-Meier methods summarized OS and PFS and the associations of muscle characteristics with OS and PFS were examined by univariate Cox proportional hazard models. Results: Median age was 35 years; median follow-up was 8.2 years. 75% had Stage II or III disease and 21% Stage IV disease. 33% were sarcopenic (SMI < 40) and 16% had low SMD (HU <37.8). Non-survivors had more intermuscular fat (IMAT), reduced SMD, and reduced skeletal muscle gauge (SMG). Sarcopenia was not associated with OS; however, sarcopenia was associated with shorter PFS. The median skeletal muscle gauge was 1973; low skeletal muscle gauge (SMG <1973) was associated with both shorter OS and PFS. Median IMAT was 1.6, and high IMAT (≥1.6) was associated with shorter OS and PFS. Conclusions: Low muscle mass (sarcopenia) at breast cancer diagnosis was associated with shorter PFS and low muscle density (low SMG and higher IMAT) was associated with shorter OS and PFS in women ≤ 40 years. These sub-optimal muscle characteristics may indicate an overall reduced state of health and/or decreased ability to tolerate treatment, thus reducing survival. Future research should determine the significance of muscle changes throughout treatment and establish standards for improved muscle health.[Table: see text]

Cancer inpatient malnutrition risk, documentation, and ICD-10 coding in an academic medical center

12107

Background: Malnutrition (MN) is common in hospitalized cancer patients but often underdiagnosed. We evaluated the prevalence of MN risk, dietitian documented MN (DDMN), and physician coded malnutrition (PCMN) in a consecutive cohort of cancer inpatients in an academic, community-based medical center. Methods: Electronic medical records (EMR) were reviewed for inpatients with a solid tumor diagnosis staged I-IV and admitted to Atrium Health Carolinas Medical Center at least once between 1/1/2016 to 5/21/2019. All data were collected from the first admission EMR encounter closest to the cancer diagnosis date. High MN risk was a score ≥2 on the Malnutrition Screening Tool (MST) completed by an RN at admission. Registered Dietitian (RD) assessments were reviewed for DDMN and grade (mild, moderate, severe). PCMN diagnosis was based on MN ICD-10 codes extracted from the medical coder’s discharge summary. Multivariate logistic regression models identified associations between clinic-demographic factors and the prevalence of DDMN and PCMN with stepwise selection. Results: N=5,143; 48% females. Median age 63 (range 18-102) years. 70% White; 24% Black, 3% Latino. Most common cancers: thoracic 19% and digestive system (14% other, 11% colorectal). 28% had known stage IV disease. The MST was completed in 79%. Among those with MST ≥2 (N=1,005; 25%), DDMN and PCMN prevalence was 30% and 22%, respectively. In the entire cohort, 8% had DDMN; 7% PCMN; 4% both. Prevalence of MN risk, DDMN, and PCMN by cancer site are in the Table. DDMN (N=420) was mild 2%, moderate 16%; severe 66%; unspecified 16%. On discharge, PCMN (N=360) was mild 10%; moderate 0%; severe 69%; unspecified 21%. Male gender (OR 1.27 [1.01, 1.59]), Black race (OR 1.57 [1.25, 1.98]), stage IV disease (v. I-III) (OR 3.08 [2.49, 3.82]), and primary site were all independent predictors of DDMN (all p<0.05); Black race (OR 1.46 [1.14, 1.87]), stage IV disease (OR 2.70 [2.15, 3.39]), and primary site were independent predictors of PCMN (all p<0.05). Conclusions: 25% of cancer inpatients were at high risk for MN. Primary site, disease stage, and race were independent predictors of a greater risk. MN appears to be under-diagnosed compared to population studies. This is the first study to report the prevalence of MN in a large cancer inpatient database with a representative population.[Table: see text]

Knitting to improve cognition and reduce stress in cancer survivors: A pilot study

e24049

Background: Cancer-related cognitive impairment (CRCI), or “chemo brain,” is defined by mild cognitive changes before, during or after cancer treatment. Objective deficits on neuropsychological tests are mild, while distress about cognitive issues is often high and interventions are lacking. Skill learning, which relies upon implicit memory, promotes functional and structural neuroplasticity. This has not been explored in CRCI. We designed a standardized protocol for learning to knit over 8 weeks and recruited consecutive cancer patients with subjective cognitive concerns. Cognition and perceived stress were assessed before and after learning to knit. Methods: Sixteen cancer survivors (N=16) without any known central nervous system disease or prior knitting experience completed the study. All were female and >80% were Caucasian. Mean age was 54 (SD=7) years. Mean education was 16 (SD=2) years. Neuropsychological tests were selected based on sensitivity to frontal lobe functions and memory and resistance to practice effects. Cognition and perceived stress were measured before and after the intervention, summarized, and compared using paired t-tests. P<0.1 was considered statistically significant. Clinically significant cognitive change was defined by ≥1 standard deviation change between pre- and post-intervention standardized scores. Results: Baseline cognitive function was average across domains. Baseline perceived stress was moderately high. At post-testing, there were statistically significant improvements in perceived stress, cognitive flexibility, and psychomotor speed. After controlling for demographic factors, clinically significant improvements in attention and memory were seen in 38% and 25% of subjects, respectively. 69% had clinical improvement in at least one cognitive domain. Conclusions: Learning to knit produced statistically and clinically significant improvements in cognition and perceived stress in cancer survivors in 8 weeks. Most had clinical improvement in at least one cognitive domain. This skill-based pilot study suggests functional neuroplastic change is possible in selected female, college-educated cancer survivors in a short timeframe and without technology use. Future research on the impact of skill learning on cognition and stress in larger, more diverse samples of cancer survivors may yield much needed interventions for CRCI.[Table: see text]

Opioid screening and urine toxicology results in outpatient oncology palliative medicine

e24068

Background: Opioid misuse is a major public health issue. Given widespread opioid prescribing in cancer patients (pts), screening for potential misuse is critical. There is lack of real-world data on opioid screening and urine toxicology testing in outpatient oncology palliative medicine. Methods: This is a retrospective clinical analysis of adult cancer pts previously consented for a pharmacogenomics specimen collection study between August 2019-March 2020. Pts completing ≥ 1 outpatient palliative medicine visit with at least half undergoing urine toxicology screening (UTS) per standard practice were included. Pt demographics, medication(s), UTS results, symptoms using Edmonton Symptom Assessment Scale, and opioid screening using Screener and Opioid Assessment for Patients with Pain - Short Form (SOAPP-SF) were collected at baseline and follow up visits, if available. The primary endpoint was the frequency and type(s) of non-compliant (NC) UTS. Secondarily, risk factors for NC UTS were evaluated using univariate and multivariate logistic regression. Results: Of 189 pts (632 visits), 113 underwent UTS, 125 SOAPP-SF, and 75 had both. The median age was 56, 56% were female, 58% white, 40% black, 48% had stage IV disease, and median pain score was 7. More black pts (72%) underwent UTS compared to white pts (53%) (p = 0.001). The mean age of pts with a UTS was 53 compared to 59 in those without UTS (p = 0.002). Oxycodone was the most prescribed drug (N = 125). Median SOAPP-SF was 3 (range 0-11); 38% had a score ≥ 4 (considered high risk). About half (54%; N = 61) who underwent a UTS were NC. Of these, 32 had 1 NC UTS, whereas 29 had 2 or more. The most common reason was presence of a substance not prescribed (N = 44 pts and 128 results), whereas 33 pts (53 results) were NC for substance(s) not present but prescribed. Four had presence of marijuana only and 21 with marijuana plus another NC substance; presence of cocaine and alcohol were the 2nd and 3rd most frequent aberrant result. Of those with a NC UTS and SOAPP-SF score (N = 44), 59% had a score ≥ 4. In univariate analyses, SOAPP-SF ≥ 4 (p = 0.004), nausea (p = 0.05), depression (p = 0.02), anxiety (p = 0.01), and prescriptions for antidepressants (p = 0.006), acetaminophen (p = 0.03), and/or dronabinol (p = 0.04), were associated with NC UTS. In multivariate analyses, SOAPP-SF Q4 (use of illegal drugs) (OR 2.86, 95% CI 1.64 to 5.02; p < 0.001) and prescription with muscle relaxants (OR 2.90, 95% CI 1.19 to 7.09; p = 0.019) were associated with increased odds of a NC UTS. Conclusions: About half of those undergoing UTS were NC. SOAPP-SF Q4 and prescription with muscle relaxants were associated with a NC UTS. Overall, pt demographics (e.g. younger, more female, more black patients, severe pain) varied from the typical cancer population. Screening using SOAPP-SF, UTS, pain contracts, prescription drug monitoring databases, and evaluating pt-specific risk factors is important to reduce opioid misuse risk.

Cannabidiol (CBD) use among cancer survivors

12096

Background: Cannabidiol (CBD) is a non-psychoactive component of cannabis touted for various therapeutic effects. The Federal Drug Agency has only approved one prescription CBD product for treatment of severe epilepsy. On December 17, 2020 the Federal Trade Commission announced legal consequences for deceptively marketed CBD products in the rapidly expanding market of various CBD products; the products’ unsupported claims included CBD as a cancer treatment. Little is known about survivors use of CBD. This study explores the prevalence and nature of CBD use by cancer survivors. Methods: A link to an anonymous, electronic survey was posted on the Levine Cancer Institute and SherryStrong (Martin Truex Jr. Foundation: philanthropy for ovarian cancer) Twitter and Facebook social media platforms. Data were managed in REDCap, a secure, web-based, electronic data capture tool. Survey responses were summarized and described with frequencies and compared using Fisher’s Exact tests; p < 0.1 was considered statistically significant. Results: N = 295 self-selected respondents were White (95%), female (86%), middle aged (45-64 years) (58%) and in the US (95%). Ninety percent indicated current (85%) or past (15%) use of CBD product; a third of these participants (N = 102) identified as cancer survivors. Gynecologic (31%) and breast (30%) cancers were the most recorded malignancies, and 38% report active treatment. Most survivors indicated using CBD products daily (77%) for a year or less (79%) and spent @$30 a week on products (70%). Common uses for CBD were easing pain (66%), anxiety (50%), and sleep (50%)—14% reported treating or preventing cancer. 41% learned about CBD from family/friends, fewer learned from the Internet (21%) or local store (11%). Only 12% received information from a physician. Liquid drops (58%) and topicals (19%) were popular products and reported side effects were sparse—sedation and/or euphoria were indicated by 10% and 2%, respectively. Over 82% of cancer survivors indicated that CBD product helped their conditions. CBD use to ease anxiety and stress declined with age; 71% of young survivors (aged 18-44) sought anxiety relief versus 45% and 36% of middle age (aged 45-64) and seniors (aged 65+), respectively (p = 0.05), and 58% of young survivors pursued stress relief versus 39% of middle age and 21% of seniors (p = 0.08). More young (25%) and middle age (37%) survivors indicated spending over $30 on products weekly than seniors (7%) (p = 0.08). No differences were seen in CBD use between cancer survivors by gender or treatment status. Conclusions: Cancer survivors commonly use CBD, yet infrequently under the guidance of a physician. Survivors largely rely on word of mouth and internet information about CBD. Despite lack of standardization of production and labeling of CBD products, the majority of patients reported positive improvements in symptoms. Future research should explore strategies to educate cancer patients and providers in safe CBD use.

Potentially actionable pharmacogenetic variants and symptom control medications in oncology

PURPOSE

We estimated the prevalence of potentially actionable pharmacogenetic (PGx) variants related to symptom control medications (SCMs) based on institutional prescribing patterns and correlated presenting symptoms with SCM prescribing.

METHODS

This was a retrospective study of adult ambulatory cancer patients undergoing electronic distress screening (EDS) within 90 days of intake to the cancer hospital. We estimated the proportion prescribed SCM(s) with PGx evidence within 90 days of intake. Those with potentially actionable variants were estimated using population frequency data from 1000 genomes. The expected number at risk of altered drug response was estimated. The associations between symptom scores and SCM(s) were estimated with logistic regression and threshold analyses performed with receiver operating characteristic (ROC) curves.

RESULTS

Of 6985 patients, 3222 (46%) received ≥ one SCM. Of these, 2760 (86%) received SCM(s) with PGx evidence for CYP2B6, CYP2C19, CYP2D6, or SLC6A4; 2719 (84%) received a drug metabolized by CYP2D6, most commonly hydrocodone (40.4%), ondansetron (35.6%), oxycodone (24.2%), and/or tramadol (7.1%). Based on this, about one quarter were expected to have altered metabolism and/or drug response. One third were prescribed two or more SCMs with PGx evidence. About half reported at least one severe symptom, which significantly correlated with SCM prescribing (p < 0.001). Threshold scores were identified that highly correlated with SCM prescribing for anxiety, depression, nausea, neuropathy, pain, and sleep.

CONCLUSION

About half presented with significant symptom burden, which highly correlated with SCM prescribing. Most received SCMs with PGx evidence. Preemptive PGx testing for these variants should be evaluated in prospective trials to evaluate the impact on symptom control.

Pain Improvement After Healing Touch and Massage in Breast Cancer: an Observational Retrospective Study

BACKGROUND

Healing Touch (HT) and Oncology Massage (OM) are nonpharmacologic pain interventions, yet a comparative effectiveness study has not been conducted for pain in breast cancer.

PURPOSE

This breast cancer subgroup analysis compared the effectiveness of HT vs. OM on pain.

SETTING

The research occurred at an outpatient setting at an academic hybrid, multi-site, community-based cancer institute and Department of Supportive Oncology across four regional locations.

PARTICIPANTS

Breast cancer outpatients along the cancer continuum who experienced routine clinical, nonexperimentally manipulated HT or OM.

RESEARCH DESIGN

The study was an observational, retrospective, comparative effectiveness post hoc subanalysis of a larger dataset. Patients reporting pain < 2 were excluded. Pre- and posttherapy pain scores and differences were calculated. Logistic regression modeled posttherapy pain by modality, adjusting for pretherapy pain. The proportions experiencing ≥ 2-point (clinically significant) pain reduction were compared with chi-square tests.

INTERVENTION

The study focused on the first session of either HT or OM.

MAIN OUTCOME MEASURES

Pre- and posttherapy pain (range: 0 = no pain to 10 = worst possible pain).

RESULTS

A total of 407 patients reported pre- and posttherapy pain scores, comprised of 233 (57.3%) who received HT and 174 (42.8%) who received OM. Pretherapy mean pain was higher in HT (M=5.1, ± 2.3) than OM (M=4.3, ± 2.1) (p < .001); posttherapy mean pain remained higher in HT (M=2.7, ± 2.2) than OM (M=1.9, ± 1.7) (p < .001). Mean difference in pain reduction was 2.4 for both HT and OM. Both HT (p < .001) and OM (p < .001) were associated with reduced pain. Proportions of clinically significant pain reduction were similar (65.7% HT and 69.0% OM, p = .483). Modality was not associated with pain improvement (p = .072).

CONCLUSIONS

Both HT and OM were associated with clinically significant pain improvement. Future research should explore attitudes toward the modalities and potential influence of cancer stage and treatment status on modality self-selection.

Abstract 1920: Response to hypomethylating agents based on cytidine deaminase expression, genetic polymorphisms, and NPM1 mutation status in acute myeloid leukemia

Response to hypomethylating agents (HMAs) in patients (pts) with acute myeloid leukemia (AML) is variable. Data on biological predictors of response is limited. Cytidine deaminase (CDA) inactivates HMAs. Increased CDA activity may lead to HMA resistance. Using core pathway analysis, we identified nucleophosmin 1 (NPM1) indirectly influences CDA expression. We hypothesized that responses to HMAs occur in the setting of decreased CDA expression regulated by NPM1 and investigated the relationship between NPM1 status, CDA expression, single nucleotide polymorphisms (SNPs) in CDA, and HMA response in pts with AML.

AML pts with banked samples who received HMA-based therapy between 1/2014 to 12/2018 were reviewed. Responses following at least 2 cycles of HMA were categorized as responders (R) or non-responders (NR). Pt, disease, NPM1 status, and treatment characteristics were summarized. CDA gene and protein expression was examined in bone marrow and peripheral blood samples at diagnosis using qRT-PCR and CDA sandwich ELISA, respectively. CDA gene expression levels were normalized to the housekeeping gene, 18s, and the comparative CT method was used to assess expression. Comparisons based on response and NPM1 mutation status were performed using the Mann-Whitney U test. 17 SNPs previously shown to alter CDA activity were selected for analysis. SNPs were determined using real-time PCR with allele specific probes; longer insertions/deletions were identified by sanger sequencing. Univariate logistic regression analysis was performed to discern the association between SNPs in CDA and response to HMAs.

54 pts had available blood, marrow, or buccal samples available for analysis. 33 pts provided blood or marrow samples for gene and protein analysis prior to HMA. 22 pts (67%) were classified as R in this cohort. 35 pts had available buccal swabs for genotyping, and 28 pts (80%) were classified as R. Median OS was 21 months (mo) for all pts, 23 mo among R, and 18 mo in NR. CDA expression was significantly decreased in NPM1 wild type pts compared to NPM1 mutant pts (p=0.02) but did not differ in R compared to NR. No significant differences were identified in CDA protein expression based on NPM1 status or response. No SNPs were significantly associated with response.

Baseline CDA gene expression in bulk tumor cells was significantly lower in NPM1 wild type pts compared to NPM1 mutant pts but no different between pts responding to HMAs compared to NR. There was no clear correlation between CDA protein expression and NPM1 status or response. None of the CDA SNPs were predictive of response to HMAs. This analysis reveals feasibility of assessing CDA activity in this population. Our small sample size limits our ability to determine CDA activity as a biological predictor of response, and ongoing pt accrual will allow for further exploration of the role of CDA in HMA response.

Citation Format: Brittany Knick Ragon, Issam S. Hamadeh, C Greer Vestal, Alicia Hamilton, Mathew L. Smith, Danielle Boselli, Jing Ai, Thomas G. Knight, Michael R. Grunwald, Jonathan M. Gerber, Edward A. Copelan, Lawrence J. Druhan, Belinda R. Avalos, Nury M. Steuerwald, Jai N. Patel. Response to hypomethylating agents based on cytidine deaminase expression, genetic polymorphisms, and NPM1 mutation status in acute myeloid leukemia [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1920.

Opioid Risk Screening in an Oncology Palliative Medicine Clinic

PURPOSE

Little information exists on factors that predict opioid misuse in oncology. We adopted the Screener and Opioid Assessment for Patients With Pain-Short Form (SOAPP-SF) and toxicology testing to assess for opioid misuse risk. The primary objective was to (1) identify characteristics associated with a high-risk SOAPP-SF score and noncompliant toxicology test, and (2) determine SOAPP-SF utility to predict noncompliant toxicology tests.

METHODS

From July 1, 2017, to December 31, 2017, new patients completed the Edmonton Symptom Assessment Scale (ESAS), SOAPP-SF, and narcotic use agreement. Toxicology test results were collected at subsequent visits.

RESULTS

Of 223 distinct patients, 96% completed SOAPP-SF. Mean age was 61 ± 12.7 years, 58% were female, 68% were White, and 28% were Black. Eighty-three eligible patients (38%) completed toxicology testing. Younger age, male sex, and increased ESAS depression scores were associated with high-risk SOAPP-SF scores. Smoking habit was associated with an aberrant test. An SOAPP-SF score ≥ 3 predicted a noncompliant toxicology test.

CONCLUSION

Male sex, young age, and higher ESAS depression score were associated with a high SOAPP-SF score. Smoking habit was associated with an aberrant test. An SOAPP-SF of ≥ 3 (sensitivity, 0.74; specificity, 0.64), not ≥ 4, was predictive of an aberrant test; however, performance characteristics were decreased from those published by Inflexxion, for ≥ 4 (sensitivity, 0.86; specificity, 0.67). The specificity warrants caution in falsely labeling patients. The SOAPP-SF may aid in meeting National Comprehensive Cancer Network recommendations to screen oncology patients for opioid misuse.

Outcomes in minority patients (pt) with aggressive B cell lymphoma (BCL) if optimally managed with equal access to care and nurse navigation (NN)

e19040

Background: Aggressive BCL is curable but previous studies have shown that minorities have inferior survival, partly due to socioeconomic barriers and poor access to care. NN programs are designed to reduce barriers to care via various methods. We present disease characteristics, treatment, and outcomes of Caucasian (C) & non-Caucasian (NC) pts with aggressive BCL at Levine Cancer Institute which has an active NN program. Methods: We collected demographic, insurance, disease characteristics, treatment, and outcomes for pts with aggressive BCL [diffuse large B cell lymphoma (DLBCL), primary mediastinal B cell lymphoma (PMBCL), or high grade B cell lymphoma (HGL)] between Jan 2016 and Jun 2019. Race (C or NC) was self-reported. NN encounters were characterized as low intensity (basic needs) or high (moderate/high needs). OS and PFS were calculated using Kaplan Meier. Demographics were compared using Fisher's Exact tests. Results: 204 pts (186 = DLBCL, 14 = PMBCL, 4 = HGL) were included (NC = 47; C = 157). NC were younger at diagnosis (median age 56 vs 62 yrs, p = 0.03) and more likely to be uninsured/Medicaid (26% vs 4%, p < 0.0001). There were no significant differences in prognostic scores (44% vs 50% R-IPI score 3-5, p = 0.5), incidence of double hit (11% vs 13%, p = 0.8), frontline rituximab/anthracycline containing chemotherapy (98% vs 96%, p = 0.9), and incidence of relapsed/refractory (R/R) disease (40% vs 37%, p = 0.7) for NC compared to C. For R/R BCL, similar % of pts underwent hematopoietic stem cell transplant (SCT) (32% NC vs 28% C, p = 0.8) or CAR-T cell therapy (16% NC vs 19% C, p = 0.9). Enrollment in clinical trials was comparable (17% NC vs 14% C, p = 0.6). The % of pts receiving NN was similar (81% NC, 87% C, p = 0.4) but NC had higher intensity NN encounters (42% vs 21%, p = 0.01). With median follow up of 35 mo, OS and PFS were comparable between both groups. The 2 yr OS was 81% for NC and 76% for C, p = 0.3; 2 yr PFS was 62% for NC and 64% for C, p = 0.8. Conclusions: We show equivalent survival between Caucasian and non-Caucasian pts with aggressive BCL. Disease biology and treatment patterns--including access to SCT, CAR-T and clinical trials--were similar in both groups. Differences in insurance coverage favored Caucasians. Similar proportion of pts in both groups received nurse navigation, but non-Caucasian pts had higher intensity navigation needs. Providing equal access to care and availability of an active nurse navigation program may overcome racial heath disparities. This study has implications for national health policy.

Limited Stage Small Cell Bladder Cancer: Outcomes of a Contemporary Cohort

BACKGROUND: Limited stage small cell bladder cancer is curable with multi-modality therapy using external beam radiotherapy or radical cystectomy. The optimal management strategy for this rare disease is still debated, yet few case series have described patients treated after 2010. OBJECTIVE: To analyze outcomes from a contemporary cohort of patients undergoing definitive treatment. METHODS: Patients diagnosed with small cell bladder cancer after January 1, 2010 were identified from an institutional database. Clinical histories were collected by chart review. Survival outcomes were analyzed in patients who received curative-intent therapy consisting of bladder radiotherapy or cystectomy. RESULTS: Thirty patients with limited stage disease that received definitive therapy were identified. Seventeen patients received primary radiotherapy, and thirteen underwent cystectomy. Median age was 70 years. Median follow up was 39.6 months (range 7.2–95.8). The median overall survival of patients undergoing radiotherapy or cystectomy were 36.8 and 30.6 months, respectively (hazard ratio 0.99, 95% confidence interval 0.35–2.85). The median metastasis free survival for patients receiving radiotherapy was not reached, and 18.9 months in the cystectomy group (hazard ratio 0.94, 95% confidence interval 0.34–2.61). The most common sites of relapse were lymph node (n = 6) and bone (n = 5). Brain metastases were less common (n = 3). CONCLUSIONS: Patients receiving cystectomy or radiotherapy had similar outcomes in this contemporary series, but definitive comparisons are limited by the cohort size and high censoring rate (53%). Survival in our cohort is improved compared with older reports, though outcomes remain poor, reiterating the need for better therapeutic options.

Pain Management Using Clinical Pharmacy Assessments With and Without Pharmacogenomics in an Oncology Palliative Medicine Clinic

PURPOSE:

Approximately 30% of patients with cancer who have pain have symptomatic improvement within 1 month using conventional pain management strategies. Engaging clinical pharmacists in palliative medicine (PM) and use of pharmacogenomic testing may improve cancer pain management.

METHODS:

Adult patients with cancer with uncontrolled pain had baseline assessments performed by PM providers using the Edmonton Symptom Assessment Scale. Pharmacotherapy was initiated or modified accordingly. A subset of patients consented to pharmacogenomic testing. The first pharmacy assessment occurred within 1 week of baseline and a second assessment was done within another week if intervention was required. Each patient’s final visit was at 1 month. Pain improvement rate (a reduction of two or more points on a 0-to-10 pain scale) from baseline to final visit was compared applying the Fisher exact test to published historical control data, and between patients with and without pharmacogenomic testing. Multivariate logistic regression identified pain improvement covariates.

RESULTS:

Of 142 patients undergoing pharmacy assessments, 53% had pain improvement compared with 30% in historical control subjects ( P < .001). Pain improvement was not different between those who received (n = 43) and did not receive (n = 99) pharmacogenomics testing (56% v 52%; P = .716). However, of 15 patients with an actionable genotype, 73% had pain improvement. Higher baseline pain (odds ratio [OR], 1.79; 95% CI, 1.43 to 2.24; P < .001), black or other race (OR, 0.42; 95% CI, 0.18 to 0.95; P = .04), and performance status 3 or 4 (OR, 0.18; 95% CI, 0.04 to 0.83; P = .03) were associated with odds of pain improvement, but pharmacogenomic testing was not ( P = .64).

CONCLUSION:

Including pharmacists in PM improves pain management effectiveness. Although pharmacogenomics did not statistically improve pain, a subset of patients with actionable genotypes may have benefited, warranting larger and randomized studies.

Pharmacogenetic (PGx) guided cancer pain management in an oncology palliative medicine (PM) clinic

117

Background: About 30% of cancer patients presenting with pain have symptomatic improvement using conventional strategies within one month. PGx may help personalize opioid selection and improve cancer pain management. Methods: This is a pragmatic pilot trial investigating the feasibility and application of PGx testing to improve pain management in adults with uncontrolled cancer pain referred to an oncology PM clinic. PM providers assessed patients using Edmonton Symptom Assessment Scale at baseline and opioid therapy was initiated or modified. A buccal swab was obtained for genotyping single nucleotide polymorphisms in: COMT, CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP3A4, CYP3A5, and OPRM1. The first assessment occurred within one week of baseline and a second within another week if intervention was required. PGx results were available before the first assessment and utilized, if applicable, throughout the one-month study period. Pain improvement rate (≥ 2-point reduction on a 0-10 scale) from baseline to final visit, was compared to historical control data by a one-sided exact binomial test of proportions. Results: Of 75 undergoing PGx testing, 52 were evaluable for the primary endpoint (54% female, 81% white, 17% black, median age 63, 75% stage 3 or 4 disease, median personalized pain goal 3 [0-6]). 56% had pain improvement compared to 30% in historical controls (p < 0.001). At final assessment, 35% met their personalized pain goal. Of 26 (50%) requiring opioid adjustments, 18 (69%) had an actionable genotype with a 61% pain improvement rate. The two most common genes for opioid adjustment were CYP2D6 (16/18; 89%) and COMT (8/18; 44%). The most common PGx-guided modification involved switching from a CYP2D6-metabolized drug (hydrocodone, oxycodone, tramadol) to a non-CYP2D6-metabolized drug (fentanyl, hydromorphone, methadone, morphine). Conclusions: PGx implementation in an oncology PM clinic was feasible and improved pain management. Half of those requiring opioid adjustments had an actionable genotype, with the largest impact from CYP2D6 polymorphisms. Future studies should focus on preemptive PGx testing to guide initial drug selection and confirm clinical utility in a randomized trial. Clinical trial information: NCT02542397.

Anxiety, depression, pain, and social support in a large representative cancer population

76

Background: Pain is one of the most common cancer symptoms. Individuals in pain often experience psychological distress in the form of anxiety and depression. Social support is an important resource utilized by patients to cope with cancer. Aims: (1) Identify clinicodemographic factors influencing cancer pain; (2) examine social support as a moderator of the relation between anxiety, depression and cancer pain. Methods: Participants included stage I-IV cancer patients (N = 11,815) who completed a routine tablet-based psychosocial distress screening at a large academic hybrid, multi-site, community-based cancer institute (Jan 2017- Jan 2019). Participants were matched to the Cancer Registry (N = 7,333); clinicodemographic factors were incorporated into lasso regression models. Models identified pain predictors from self-reported anxiety, depression and social support. Analyses examined if the effect of anxiety and depression on pain differed by levels of social support. Results: Median age was 59 (RNG, 18-101), 61% female and 77% white. Tumor site (GI, Gyn, head/neck), advanced disease, black race, and lower income were independently associated with severe pain. Anxiety (β = 0.48, p < .001) and depression (β = 0.69, p < 0.001) were related to pain intensity after accounting for clinicodemographic factors. The effect of depression on pain differed by level of social support (p = 0.009). The effect of anxiety on pain differed in patients reporting transportation issues (p = 0.035). Conclusions: This is the largest study to date examining cancer pain intensity, psychological factors of anxiety and depression, and social support. Our data suggests that patient characteristics of race, income, tumor site, and disease staging independently predict pain intensity. Anxiety and depression are significant factors of pain intensity; these associations remain after accounting for patient characteristics. Social support buffers the negative impact of anxiety/depression on pain. Clinicians who treat cancer pain should be attuned to modifiable psychological factors which can greatly influence a patient’s pain experience. Findings emphasize the need for interdisciplinary multimodal approaches for cancer pain.

Supportive care medications (SCMs) and pharmacogenomics (PGx) relevance in 6,985 cancer patients (pts) undergoing distress screening

11592

Background: SCMs are prescribed based on symptom burden, but response is variable, possibly due to PGx. We investigated the association between symptom burden, SCM prescribing, and frequency of SCMs with PGx evidence. Methods: Cancer pts ≥ 18 years old and completing electronic distress screening within 90 days of intake between 1/1/2017-12/31/2017 were included. Anxiety was measured using Generalized Anxiety Disorder 2-item (0-6) and depression using Patient Health Questionnaire-2 (0-6). Fatigue, nausea, neuropathy, pain and sleep were measured on a 0-10 scale. SCM prescribing within 90 days of intake was documented. Logistic regression compared symptom scores and SCM prescribing. Receiver Operating Characteristics analysis estimated sensitivity/specificity. Optimal symptom thresholds were selected according to Youden’s J statistic. SCMs with PGx evidence level A or B (according to Clinical Pharmacogenetics Implementation Consortium) were summarized. Results: Of 6985 pts, 65% were female, 75% Caucasian, 20% African American and median age was 60. 49% reported ≥ 1 severe symptom, which correlated with SCM prescribing (p < 0.001). 3208 (46%) were prescribed SCM(s), mainly for pain (69%) or nausea (46%). Of these, 2759 (86%) received ≥ 1 SCM with PGx evidence and 2695 (84%) received a SCM metabolized by CYP2D6 - hydrocodone (47%), ondansetron (41%), and oxycodone (28%). Based on reported CYP2D6 allele frequencies conferring altered metabolism (~20%), 539 of the 2695 pts may have altered drug response. Threshold scores for each symptom are summarized in the table. Fatigue and nausea were not associated with SCM prescribing. Conclusions: Symptom burden is high in cancer pts and correlates with SCM prescribing. Many SCMs have PGx evidence, suggesting preemptive testing, particularly for CYP2D6, may have broad applicability in this population.[Table: see text]

Minimally Invasive Ivor Lewis Esophagectomy Without Patient Repositioning

INTRODUCTION

The standard technique for Ivor Lewis minimally invasive esophagectomy involves a two-stage approach necessitating repositioning mid-procedure.

TECHNIQUE

We describe our technique for a one-stage hand-assisted minimally invasive esophagectomy that allows sequential access to the chest and abdomen within the same surgical field, eliminating the need for repositioning. The patient is positioned in a "corkscrew" configuration with the abdomen supine and the chest rotated to the left to allow access to the right chest. The abdomen and chest are prepped into a single operative field. This technique allows sequential access to the abdomen for gastric mobilization, chest for division of the esophagus, abdomen for construction of the gastric conduit, and chest for intrathoracic anastomosis.

CONCLUSION

This approach enables extracorporeal construction of the conduit, which helps ensure a clear distal margin on the specimen and facilitates conduit length by placing the stomach on stretch during stapling.

Nintedanib in metastatic appendiceal carcinoma

TPS723

Background: Appendiceal carcinomas are rare with an incidence of about 0.12 cases per 1,000,000 people per year. There is limited, mostly retrospective data in the treatment of metastatic appendiceal carcinomas. Generally, fluoropyrimidine-based therapy is used in the first line, adapting regimens for metastatic colorectal cancer. However, beyond progression, no treatments have shown clear activity. In appendiceal cancer, high vascular endothelial growth factor receptor (VEGFR)2 expression has been correlated with poor survival. Moreover, malignant ascites has been demonstrated to have elevated levels of VEGF. Nintedanib is an oral tyrosine kinase inhibitor of VEGFR which demonstrated activity in lung and ovarian cancer in clinical trials, and has undergone investigation in heavily pretreated metastatic colorectal cancer. Given the analogies between appendiceal and colorectal cancer and potentially ovarian cancer, and the limited information about the optimal treatment of metastatic appendiceal carcinomas, further investigation with nintedanib is warranted. Methods: This is a single arm, open label, investigator initiated, two-stage phase II trial (NCT 03287947) in metastatic appendiceal cancer patients after failure (defined as progression on or within 6 months or intolerance) of initial fluoropyrimidine-based therapy and at least one measurable site of disease. The trial started enrolling patients in June 2018, and up to 39 subjects will be enrolled. They will be treated with 200 mg of oral nintedanib twice daily and undergo disease evaluation every two months. The primary objective of this study is to evaluate the disease control rate (DCR), the composite of objective response and stable disease per RECIST 1.1. Secondary objectives include evaluation of safety and toxicity, objective response rate (ORR), 6-month progression free survival (PFS) and overall survival (OS). DCR, ORR & 6-month PFS will be estimated with the corresponding 95% Clopper-Pearson confidence interval. PFS & OS will be estimated using Kaplan-Meier techniques. Exploratory objectives include evaluation of serum VEGF, ascites VEGF, hypertension and paracentesis frequency in subjects with ascites at study entry. Clinical trial information: NCT 03287947.

Examination of the Screener and Opioid Assessment for Patients with Pain-Short Form (SOAPP-SF) in an oncology palliative medicine clinic

196

Background: The National Comprehensive Cancer Network states opioids can be used to treat cancer pain and prescribers should identify patients at risk for opioid misuse; research in this area is limited. In the non-cancer population, SOAPP-SF is a validated tool to predict aberrant drug behavior; a score of ≥ 4 (out of 20) is considered high risk. We performed a retrospective observational study to determine the utility of the SOAPP in identifying opioid misuse in the oncology population as measured by a non-compliant toxicology screen. Methods: Consecutive consults seen during a 6-month period completed the 5-question SOAPP-SF and Edmonton Symptom Assessment System (ESAS) form. Toxicology screens assessed non-compliance (i.e., absence of prescribed medications and/or presence of non-prescribed or illegal substances). Logistic regression models estimated the associations of composite and individual SOAPP-SF scores and ESAS symptom scores with non-compliant screens. Threshold analysis were conducted to identify an optimal SOAPP-SF cutoff. Results: Of 192 consults, 64 patients providing SOAPP-SF score and toxicology screen were evaluable. Mean age was 59 ± 9.8 years: 56% were female, 34% and 62% were African American and Caucasian respectively. Median SOAPP-SF score was 2 (range: [0, 12]). Non-compliant screens were observed in 31% of patients. The area under the curve (AUC) was 0.65. The validated SOAPP-SF cutoff score of ≥ 4 was associated with a sensitivity and specificity of 0.43 and 0.79, respectively (p = 0.082). Sensitivity (0.76) and specificity (0.72) were maximized at a cutoff score of ≥ 3 (p < 0.001). When evaluated individually, the SOAPP-SF question about smoking habit was associated with a non-compliant screen (p = 0.020). Increased ESAS pain scores were associated with SOAPP-SF score ≥ 3 (p = 0.013). Conclusions: SOAPP-SF can identify oncology patients at risk for opioid misuse. Preliminary analyses suggest a more appropriate threshold of identification is a score of ≥ 3 not ≥ 4. Future work will increase numbers of evaluable patients and examine other factors associated with opioid misuse.

Rationale for Mastectomy after Neoadjuvant Chemotherapy

Neoadjuvant chemotherapy (NAC) reduces tumor size, facilitating the use of breast conservation surgery (BCS). However, mastectomy remains the surgical outcome for certain women. The goal of this study was to determine the rationale for mastectomy after NAC, particularly in women eligible for BCS. Retrospective data were reviewed on patients who received NAC between February 2006 and August 2010 at our institution. Demographics and tumor characteristics were compared between patients who received BCS and mastectomy after NAC. Of 149 patients meeting inclusion criteria, 102 (68%) underwent BCS and 47 (32%) underwent mastectomy. Patient preference was the most common rationale for mastectomy ( n = 19; 40%), followed by extent of disease ( n = 13; 28%), presence of a breast cancer susceptibility gene (BRCA) mutation ( n = 9; 19%), persistent positive margins ( n = 5; 11%), and wound complications ( n = 1; 2%). Of the 47 patients who underwent mastectomy, 37 (79%) were eligible for BCS after NAC. Larger pathologic tumor size (2.05 vs 1.25 cm, P = 0.04) and lobular histology [invasive lobular carcinomas, n = 12/17 (70%) vs invasive ductal carcinomas, n = 36/133 (27%); P < 0.01] were associated with increased rate of mastectomy. After NAC, patient preference, extent of disease, and the presence of a BRCA mutation account for the vast majority of mastectomies. Interestingly, most of these patients were shown to be candidates for breast conservation. This highlights the importance of educating patients about their surgical choice and the lack of evidence, showing a benefit to more extensive surgery.

Optimal Lymphadenectomy in Small Bowel Neuroendocrine Tumors: Analysis of the NCDB

BACKGROUND

Current National Comprehensive Cancer Network guidelines for resectable small bowel neuroendocrine tumors (NETs) recommend regional lymphadenectomy. However, no consensus exists on the optimal nodal harvest.

METHODS

The National Cancer Database was queried for patients with resectable small bowel NETs (1998-2013). Patients with metastatic disease and missing lymph node harvest data were excluded. We performed logistic regression of factors determining nodal positivity and multivariable survival analyses.

RESULTS

Of 11,852 patients, 81.8% underwent lymphadenectomy. 79.3% were node positive (N+) and 46.9% of patients had tumors < 1 cm. Independent predictors of N+ were large tumor size, ileal location, and neuroendocrine carcinoma histology. Logistic regression found no difference between observed and expected proportions of N+ patients with lymphadenectomy greater than or equal to eight nodes. Lower metastatic node ratio predicted improved survival on multivariable analysis and is associated with high-volume institutions.

CONCLUSION

Small bowel NETs have high rates of nodal metastasis, even in patients with small tumors, and many patients do not undergo lymphadenectomy despite the clear benefit. Lymphadenectomy of eight nodes is optimal to identify N+ patients. Additionally, minimizing metastatic node ratio with complete regional lymphadenectomy is associated with improved survival in these patients.

Rationale for Mastectomy after Neoadjuvant Chemotherapy

Neoadjuvant chemotherapy (NAC) reduces tumor size, facilitating the use of breast conservation surgery (BCS). However, mastectomy remains the surgical outcome for certain women. The goal of this study was to determine the rationale for mastectomy after NAC, particularly in women eligible for BCS. Retrospective data were reviewed on patients who received NAC between February 2006 and August 2010 at our institution. Demographics and tumor characteristics were compared between patients who received BCS and mastectomy after NAC. Of 149 patients meeting inclusion criteria, 102 (68%) underwent BCS and 47 (32%) underwent mastectomy. Patient preference was the most common rationale for mastectomy (n = 19; 40%), followed by extent of disease (n = 13; 28%), presence of a breast cancer susceptibility gene (BRCA) mutation (n = 9; 19%), persistent positive margins (n = 5; 11%), and wound complications (n = 1; 2%). Of the 47 patients who underwent mastectomy, 37 (79%) were eligible for BCS after NAC. Larger pathologic tumor size (2.05 vs 1.25 cm, P = 0.04) and lobular histology [invasive lobular carcinomas, n = 12/17 (70%) vs invasive ductal carcinomas, n = 36/133 (27%); P < 0.01] were associated with increased rate of mastectomy. After NAC, patient preference, extent of disease, and the presence of a BRCA mutation account for the vast majority of mastectomies. Interestingly, most of these patients were shown to be candidates for breast conservation. This highlights the importance of educating patients about their surgical choice and the lack of evidence, showing a benefit to more extensive surgery.

Pathologic complete response rate after neoadjuvant chemoradiation in patients with locally advanced rectal cancer affects survival in patients with prolonged radiation-surgery interval

3608

Background: The current standard of care in locally advanced rectal cancer is neoadjuvant chemoradiation and R0 resection. An optimal radiation-surgery interval (RSI) has not been established. A small institutional dataset showed RSI > 49 days improved pathologic complete response (pCR) rates and disease free survival. However, in a national dataset, RSI greater than 60 days was associated with increased rates of positive margins and impaired overall survival. Because pCR is associated with improved survival, we used a national database to evaluate the relationship between RSI, pCR and survival after neoadjuvant therapy for rectal cancer. Methods: The NCDB was queried for cases 2004-2013 of AJCC stage II or III rectal adenocarcinoma that underwent neoadjuvant radiation followed by radical resection. We excluded patients with missing and outlier RSI. pCR was defined as ypT0N0M0. Chi-square, univariate, multivariable Cox model, and Cochran-Armitage time trend analyses were performed. Results: 23475 patients were identified. 7901 (33.7%) had RSI ≥60 days. pCR occurred in 1766 (11.3%) of the < 60 group and 1174 (14.9%) of the ≥60 group (p < 0.001). RSI ≥60 days has increased over time, from 22.1% in 2004 to 45.4% in 2013 (p < 0.001), as have pCR rates, from 8.4% in 2004 to 14.2% in 2013 (p < 0.001). Multivariable Cox model of the total cohort showed that RSI ≥60 days (HR = 1.11, 95% CI = 1.04-1.19) and residual disease (HR = 2.04, 95% CI = 1.78-2.34) were associated with increased mortality. Subgroup analysis of patients with pCR showed RSI ≥60 days was not associated with worse survival (HR = 1.07, 95% CI = 0.82-1.41). However, analysis of patients with residual disease showed RSI ≥60 days was associated with worse survival (HR = 1.13, 95% CI = 1.06-1.21). Conclusions: In a large national database, RSI ≥60 days worsens survival in patients who have residual disease after neoadjuvant therapy for locally advanced rectal cancer, while there is no difference in those with pCR. Emphasis should be placed on identifying patients who are unlikely to have pCR and to prioritize resection in these patients within 60 days of completion of chemoradiation.

Utilization of primary chemoradiotherapy for anal squamous cell carcinoma in the elderly: An analysis of the NCDB

726

Background: Definitive chemoradiotherapy (CHEMORT) is the treatment of choice for anal squamous cell carcinoma (SCC), while surgery is typically reserved for salvage therapy. Patients (pts) who are frail due to advanced age or medical comorbidities often have difficulty completing therapy. Methods: The NCDB was queried for pts with anal SCC (2004-2012). Pts < 50 years, and those with in situ or metastatic disease, or with incomplete CHEMORT treatment data were excluded. The primary outcome was completion of CHEMORT. Secondary outcome was requirement of salvage surgical therapy. Statistical analyses include Chi-square, univariate and multivariable logistic regression. Results: N = 11918. 5907 (49.5%) did not complete recommended CHEMORT. 9862 (82.8%) received CHEMO, 6011 (61.0%) of whom completed RT with dosage > 45Gy. Factors significantly associated with failure to complete therapy on multivariable analysis include: older age at diagnosis, higher Charlson-Deyo score, earlier year of diagnosis, male gender, and earlier clinical T and N stages (Table 1). 41.7% of pts who did not complete CHEMORT required salvage surgical therapy, versus 25.1% of pts completing CHEMORT (OR: 2.14 95% CI [1.97, 2.31], p < 0.01). Conclusions: Approximately half of pts older than 50 years of age with anal SCC failed to complete definitive CHEMORT. This study highlights the negative impact of frailty on the ability of pts to receive optimal therapy, resulting in more operative interventions. Medical optimization of older pts with more comorbidities in order to improve utilization of CHEMORT is one possible area of improvement in the management of anal SCC. [Table: see text]

Comparing pre-operative stereotactic radiosurgery (SRS) to post-operative whole brain radiation therapy (WBRT) for resectable brain metastases: a multi-institutional analysis

Pre-operative stereotactic radiosurgery (pre-SRS) has been shown as a viable treatment option for resectable brain metastases (BM). The aim of this study is to compare oncologic outcomes and toxicities for pre-SRS and post-operative WBRT (post-WBRT) for resectable BM. We reviewed records of consecutive patients who underwent resection of BM and either pre-SRS or post-WBRT between 2005 and 2013 at two institutions. Overall survival (OS) was calculated using the Kaplan-Meier method. Cumulative incidence was used for intracranial outcomes. Multivariate analysis (MVA) was performed using the Cox and Fine and Gray models, respectively. Overall, 102 patients underwent surgical resection of BM; 66 patients with 71 lesions received pre-SRS while 36 patients with 42 cavities received post-WBRT. Baseline characteristics were similar except for the pre-SRS cohort having more single lesions (65.2% vs. 38.9%, p = 0.001) and smaller median lesion volume (8.3 cc vs. 15.3 cc, p = 0.006). 1-year OS was similar between cohorts (58% vs. 56%, respectively) (p = 0.43). Intracranial outcomes were also similar (2-year outcomes, pre-SRS vs. post-WBRT): local recurrence: 24.5% vs. 25% (p = 0.81), distant brain failure (DBF): 53.2% vs. 45% (p = 0.66), and leptomeningeal disease (LMD) recurrence: 3.5% vs. 9.0% (p = 0.66). On MVA, radiation cohort was not independently associated with OS or any intracranial outcome. Crude rates of symptomatic radiation necrosis were 5.6 and 0%, respectively. OS and intracranial outcomes were similar for patients treated with pre-SRS or post-WBRT for resected BM. Pre-SRS is a viable alternative to post-WBRT for resected BM. Further confirmatory studies with neuro-cognitive outcomes comparing these two treatment paradigms are needed.

Pathologic complete response rates after neoadjuvant treatment in rectal cancer: An analysis of the NCDB

713

Background: Pathologic complete response (pCR) of rectal cancer following neoadjuvant therapy is associated with decreased local recurrence and increased overall survival. The present study utilizes a large national dataset to identify predictors of pCR in rectal cancer. Methods: The NCDB was queried for patients with non-metastatic rectal cancer (2004-2011) who underwent neoadjuvant therapy (regional radiation dose 4500 cGy, boost dose 540 cGy) followed by surgical resection. Generalized linear mixed models were used to analyze the probability of pCR by hospital volume with adjustments for demographic, socioeconomic, staging, and tumor characteristics. Hospitals were separated into groups based on the number of resections performed per year <2, 2-5, and 5+. To account for clustering of cases at individual hospitals, a random effect was used at the hospital level and covariates were included as fixed effects. Results: 7,859 patients met inclusion criteria from 951 participating hospitals. Generalized linear mixed models demonstrated that the odds of achieving pCR was independently associated with more recent diagnosis, female gender, private insurance, smaller tumor size, lower grade, lower clinical T-classification, increasing interval between the end of radiation and surgery, and treatment at higher volume institutions (Table). Conclusions: The incidence of pCR was associated with favorable tumor factors (size, grade, T classification), demographics (insurance status) as well as treatment factors (time between radiation and surgery and institutional volume). With the data available, it is not clear what is driving the higher rates of pCR at high volume institutions. Research specifically targeted at understanding processes which are associated with pCR in high volume institutions is needed so that similar results can be achieved across the spectrum of facilities caring for patients in this population. [Table: see text]

Impact of Tumor Size on Probability of Pathologic Complete Response After Neoadjuvant Chemotherapy

BACKGROUND

The prospective Neoadjuvant Breast Symphony Trial (NBRST) study found that MammaPrint/BluePrint functional molecular subtype is superior to conventional immunohistochemistry/fluorescence in situ hybridization subtyping for predicting pathologic complete response (pCR) to neoadjuvant chemotherapy. The purpose of this substudy was to determine if the rate of pCR is affected by tumor size.

METHODS

The NBRST study includes breast cancer patients who received neoadjuvant chemotherapy. MammaPrint/BluePrint subtyping classified patients into four molecular subgroups: Luminal A, Luminal B, HER2 (human epidermal growth factor receptor 2), and Basal type. Probability of pCR (ypT0/isN0) as a function of tumor size and molecular subgroup was evaluated.

RESULTS

A total of 608 patients were evaluable with overall pCR rates of 28.5 %. Luminal A and B patients had significantly lower rates of pCR (6.1 and 8.7 %, respectively) than either basal (37.1 %) or HER2 (55.0 %) patients (p < 0.001). The probability of pCR significantly decreased with tumor size >5 cm [p = 0.022, odds ratio (OR) 0.58, 95 % confidence interval (CI) 0.36, 0.93]. This relationship was statistically significant in the Basal (p = 0.026, OR 0.46, 95 % CI 0.23, 0.91) and HER2 (p = 0.039, OR 0.36, 95 % CI 0.14, 0.95) subgroups. In multivariate logistic regression analyses, the dichotomized tumor size variable was not significant in any of the molecular subgroups.

DISCUSSION

Even though tumor size would intuitively be a clinical determinant of pCR, the current analysis showed that the adjusted OR for tumor size was not statistically significant in any of the molecular subgroups. Factors significantly associated with pCR were PR status, grade, lymph node status, and BluePrint molecular subtyping, which had the strongest correlation.

Impact of surgical care on survival in esophageal cancer

142

Background: Survival after multimodality treatment of localized esophageal cancer depends upon complex interactions between the patient, tumor biology, and treatment factors. The National Cancer Database (NCDB) was used to analyze prognostic factors to identify areas for treament optimization. Methods: 8,072 patients with localized esophageal cancer treated with neoadjuvant therapy undergoing surgical resection between 2004 and 2006 were identified from the NCDB. Covariates were analyzed for association with survival using univariate and multivariate Cox models. Results: A multiviariate Cox proportional hazards model was constructed, with the following significant factors predictive of survival. (See Table.) Survival varied markedly based upon the annual surgical volume of esophageal resection performed at the hospital. For hospitals performing 5 or fewer esophageal resections per year (15% of cases), 5-year survival was 40.0%, compared with 48.6% for hospitals performing 20 or greater (26% of cases). Hospital length of stay after surgery also profoundly affected survival. For patients with a post-operative length of stay of less than 14 days, 5-year survival was 40% and median survival 39.1 months. Median survival was 28 months, 19 months, and 15 months in patients with a hospital length of stay of 14-21 days, 21-28 days, and greater than 28 days, respectively. Conclusions: Data from the NCDB confirms the association between perioperative events and long-term survival after resection for esophageal cancer. Given the wide variance in outcomes based upon perioperative treatment factors, future improvements in outcomes are unlikely to be dramatically influenced by optimization of chemotherapy and radiation therapy. Improvement in outcomes of the treatment of esophageal cancer will likely require understanding how the perioperative period influences long-term survival, which should drive priorities for research and treatment improvement. [Table: see text]

Optimal timing of surgical resection after radiation therapy in locally advanced rectal adenocarcinoma: An analysis of the National Cancer Database (NCDB)

510

Background: In treatment of rectal adenocarcinoma, an increased time delay (TD) of 6-12 weeks from the end of radiation therapy to surgery may increase the rate of complete pathologic response (pCR), but the optimal TD with respect to survival has not been established. This study evaluates the impact of TD on overall mortality. Methods: The NCDB was queried for patients with adenocarcinoma of the rectum and no evidence of metastasis at diagnosis, who underwent preoperative chemoradiation followed by radical surgical resection. Standard statistical methods were employed for descriptive statistics and Cox model development. Results: The study included 6805 patients, predominantly Caucasian (87.2%) and males (63.9%) who generally were treated with low anterior resection (57.3%), colonanal reanastomosis (8.4%), or abdominoperineal resection (28.4%), and had median survival of 66.6 months. The effects of age, surgical margins (-/+), comorbidity index, time to discharge after surgery, TMN pathologic staging, surgical volume, and patient income significantly impacted mortality after radiation and surgery (p<0.05 for all values). There was a significant relationship between TD and pCR (p=.0002). At TD less than 30 days, 4.0% of patients achieved pCR, while 9.3% of patients have achieved pCR by 75 days. In TD of greater than 75 days, the rate of pCR decreased. Overall, 6.8% of patients (n=461) achieved pCR. Using a refined cox model, a TD of more than 60 days was associated with 20% greater risk of mortality (95% CI 1.068 – 1.367). This effect became more pronounced with increasing TD; a TD of greater than 75 days was associated with 28% (95% CI 1.06-1.55) increased risk of mortality, while patients with TD less than 60 days saw a survival benefit. Conclusions: Though an interval up to 75 days between radiation and surgery may achieve higher rates of complete pathologic response, delay of more than 60 days from radiation to surgical resection and subsequent systemic chemotherapy decreases overall survival in patients with rectal cancer.