Impact of a G-CSF Policy to Reduce Low-Value Care on Guideline Adherence and Mortality

PURPOSE

Practice guidelines recommend the prophylactic use of granulocyte colony-stimulating factors (G-CSFs) in patients with high risk of febrile neutropenia, but evidence suggests that G-CSFs are frequently overused. The objectives of this study were (1) to determine the prevalence and prescribing patterns of G-CSF and (2) to evaluate the impact of a program initiative on G-CSF prescribing patterns, adherence to guidelines, and mortality.

METHODS

In this retrospective cohort study, data were used from the electronic health records of patients with metastatic colorectal cancer who received care at a multicenter oncology practice network during two time periods: July 01, 2013, to December 31, 2014, and July 01, 2017, to December 31, 2017. Beginning 2016, a site-wide program initiative that involved educational materials, appropriate nonuse recommendations, and prior authorization was introduced in the oncology practice network with an aim of reducing G-CSF overutilization. Descriptive statistics, t tests, and chi-squared tests were employed to analyze program impact.

RESULTS

There were 3,426 chemotherapy regimens corresponding to 2,968 patients. There were a total of 387 (11.3%) G-CSF-treated patients and 3,095 G-CSF administrations during the study period. G-CSF use was significantly lower in the postperiod, compared with the preperiod (P < .0001). Adherence to guidelines was significantly higher in the postperiod, compared with the preperiod (P < .0001). Mortality rates did not significantly differ between the two time periods.

CONCLUSION

This study demonstrates that policy initiatives have the potential to positively affect G-CSF prescription patterns and promote guideline adherence. These findings could help prescribers adopt a cost-effective approach in patients with metastatic colorectal cancer, leading to enhanced clinical practice and value-based care.

Cost differential by site of service for cancer patients receiving chemotherapy

OBJECTIVES

To compare the costs of: 1) chemotherapy treatment across clinical, demographic, and geographic variables; and 2) various cancer care-related cost categories between patients receiving chemotherapy in a community oncology versus a hospital outpatient setting.

STUDY DESIGN

Data from the calendar years 2008 to 2010 from the Truven Health Analytics MarketScan Commercial Claims and Encounters Database were analyzed. During 2010, the data set contained approximately 45 million unique commercially insured patients with 70,984 cancer patients receiving chemotherapy. These patients were assigned to cohorts depending on whether they received chemotherapy at a community oncology or hospital outpatient setting.

METHODS

Cost data for 9 common cancer types were extracted from the database and analyzed on a per member per month basis to normalize costs; costs included amounts paid by the payer and patient payment. Community oncology and hospital outpatient setting chemotherapy treatment costs were categorized and examined according to cancer diagnosis, patient demographics, and geographic location.

RESULTS

Patients receiving chemotherapy treatment in the community oncology clinic had a 20% to 39% lower mean per member per month cost of care, depending on diagnosis, compared with those receiving chemotherapy in the hospital outpatient setting. This cost differential was consistent across cancer type, geographic location, patient age, and number of chemotherapy sessions. Various cost categories examined were also higher for those treated in the hospital outpatient setting.

CONCLUSIONS

The cost of care for patients receiving chemotherapy was consistently lower in the community oncology clinic compared with the hospital outpatient setting, controlling for the clinical, demographic, and geographic variables analyzed.

Abstract P2-11-16: Cardiac Morbidity After Adjuvant Chemotherapy (CT) for Early Breast Cancer in the Community Setting

Cardiac morbidity after exposure to CT is a known and previously described risk. Anthracycline exposure can be complicated by acute or chronic cardiac toxicity. Trastuzumab exposure is largely associated with acute and reversible cardiac morbidity.

Methods: We retrospectively queried the electronic health record (EHR) from our network of community oncology practices, iKnowMed, for patients (pts) diagnosed with Stage I-III breast cancer (BC) from 2007–2010 with at least 5 visits and follow-up (f/u) through 2012. We stratified this group by CT utilization (yes/no), regimen type, age, and characterized the incidence of cardiac disease or initiation of cardiac medication through the f/u period to determine the association of cardiac disease or treatment with CT utilization. Cardiac diseases analyzed included congestive heart failure, valvular and ischemic heart disease, arrythmias, and hypertension. Cardiac medications included beta blockers, angiotensin-converting-enzyme inhibitors, angiotensin receptor II blockers, loop and thiazide diuretics. Hazard ratios by prespecified risk parameters were then analyzed by multivariate analysis for all pts who did not have cardiac disease preceding their diagnosis of BC.

Results: We identified 20,900 pts with a median f/u of 3.2 yrs (1.4–5.4). 11,295 (54%) pts received adjuvant CT and 9,605 (46%) did not. Median age at diagnosis in the CT-treated arm and non CT-treated arm was 54 and 64 yrs, respectively (p &lt; 0.0001). Among both the non-CT and CT-treated group, the baseline prevalence of cardiac disease was 14%. Among the CT-treated group, 3475 pts or 31% (95% CI, 30 %−32%) had or developed cardiac disease within the study period. In the non-CT group, 3790 pts or 39% (95% CI, 38%−40%) had or developed cardiac disease with the study period (p &lt; 0.01). Receiving CT conveyed a lower risk of cardiac morbidity overall, HR 0.86 (p &lt; 0.01). Incidence of cardiac disease was higher among pts who were in the non-CT treated arm (39%) than among the various CT-treated arms: anthracycline and trastuzumab (30%), anthracycline without trastuzumab (26%), non-anthracycline with trastuzumab (33%), and non-anthracycline without trastuzumab (34%). Incidence of cardiac disease increased proportionally over time in all age groups as expected in both cohorts.

Creating a process to standardize regimen order sets within an electronic health record

PURPOSE

US Oncology uses regimen order sets in clinical practice to treat patients. However, the process to assure accuracy and upkeep of these order sets has not been described. The purpose of this project was to evaluate the regimens housed in the electronic health record, iKnowMed, to determine their appropriateness and accuracy.

MATERIALS AND METHODS

US Oncology conducted an audit of its standardized regimen library. A utilization review compared chemotherapy regimens in the library and consolidated order sets on the basis of past utilization. Next, internal and external clinical pharmacists were contracted to verify the accuracy, dose, duration, and cycle length of regimens. References cited in the regimen library were evaluated. New or updated references or clinical practice standards were added or modified when necessary. US Oncology corporate pharmacists reviewed the recommendations and discussed findings with an oversight committee. Final proposals were voted on before being incorporated into iKnowMed. An internal database tracking system tool for all reviewed recommendations was created to track and communicate needed changes to the electronic health record.

RESULTS

Out of 511 regimen order sets, 51 were recommended for removal or consolidation. Of the remaining 460 regimen order sets, all had some administrative changes. Specifically, 75% had title changes, 14% had cycle-related changes, 31% had reference updates, and 13% had dosing updates.

CONCLUSION

Electronic health records systems, such as iKnowMed, can provide standardized order sets for a large oncology network. However, the regimens need to be evaluated routinely using standardized procedures to ensure they are accurate and current.

Epidermal growth factor receptor inhibitor-associated cutaneous toxicities: an evolving paradigm in clinical management

Epidermal growth factor receptor inhibitors (EGFRIs) have demonstrated improved overall survival in patients with non-small cell lung cancer, pancreatic cancer, and colorectal cancer; however, their use is associated with dermatologic reactions of varying severity. The similar spectrum of events observed with monoclonal antibodies and tyrosine kinase inhibitors suggests such toxicities are a class effect. While such reactions do not necessarily require any alteration in EGFRI treatment, being best addressed through symptomatic treatment, there is limited evidence on which to base such therapies. In October 2006, at an international and interdisciplinary EGFRI dermatologic toxicity forum, the underlying mechanisms of these toxicities were discussed and commonly used therapeutic interventions were evaluated. Our aim was to reach a current consensus on management strategies. A three-tiered, EGFRI-focused toxicity grading system is suggested for the purposes of therapeutic decision making, and as a framework on which to build a stepwise approach to intervention. This approach to successful management is specifically tailored to accurately categorize dermatologic toxicity associated with EGFRIs, and can be easily applied by all health care professionals. The goal is to maximize quality of life in patients who are being treated with these agents--many of whom will be on these drugs for several months or even years.

Managing Dermatologic Toxicities of Epidermal Growth Factor Receptor Inhibitors

There is considerable evidence that epidermal growth factor receptor (EGFR) plays an important role in non-small-cell lung cancer tumor growth and proliferation. Clinical experience with EGFR inhibitors, such as erlotinib, gefitinib, and cetuximab, has suggested that there are subgroups of patients with non-small-cell lung cancer that demonstrate dramatic responses to these agents. Researchers have sought to determine whether molecular or clinical characteristics correlate with therapeutic outcomes. Rash, the most commonly reported adverse effect of anti-EGFR therapy, has also been examined as a potential marker of response. Several trials evaluating anti-EGFR therapies have reported a positive correlation between rash and response and even rash and survival. If rash is truly a predictor of outcome, it becomes imperative that clinicians identify effective methods for managing this toxicity to avoid unwanted dose reduction, therapy interruption, delay, or discontinuation in patients experiencing a therapeutic benefit. Unfortunately, because of a lack of well-defined rash etiology, variability of use, and interpretation of rash grading scales, as well as a lack of clinical trials evaluating approaches to rash management, we are left without systematic, evidence-based guidelines for treatment. Preliminary results of a prospective study evaluating a rash treatment algorithm developed at the University of Texas M. D. Anderson Cancer Center have been positive, and there is universal agreement that initiation of more prospective trials to evaluate EGFR-inhibitor rash management is needed.

Management of Rash and Other Toxicities in Patients Treated with Epidermal Growth Factor Receptor—Targeted Agents

Inhibition of the epidermal growth factor receptor (EGFR) represents one of the most important avenues for research and development in the field of cancer therapy. The EGFR is a member of the ErbB receptor tyrosine kinase (TK) family, which also includes ErbB-2 (HER2/neu), ErbB-3 (HER3), and ErbB-4 (HER4). Current EGFR therapies available for use include monoclonal antibodies, such as cetuximab, and small-molecule EGFR TK inhibition by agents such as erlotinib. Side effects of these agents include dermatologic manifestations without the bone marrow suppressive properties of chemotherapy. Understanding of rash and how it relates to EGFR inhibitor toxicity and, perhaps more importantly, EGFR inhibitor response must be more clearly defined with clinical trials. The optimum management of rash in patients receiving anti-EGFR therapy remains somewhat controversial; this is secondary to imprecise classification of rash as well as the lack of clinical trials to determine the most appropriate treatment algorithm for these patients. We propose a treatment strategy to help aggressively treat dermatologic side effects allowing patients to continue receiving therapy without dose interruption or drug discontinuation.

Genetic contributions to generalized arousal of brain and behavior

We have identified a generalized arousal component in the behavior of mice. Analyzed by mathematical/statistical approaches across experiments, investigators, and mouse populations, it accounts for about 1/3 of the variance in arousal-related measures. Knockout of the gene coding for the classical estrogen receptor (ER-alpha), a ligand-activated transcription factor, greatly reduced arousal responses. In contrast, disrupting the gene for a likely gene duplication product, ER-beta, did not have these effects. A combination of mathematical and genetic approaches to arousal in an experimentally tractable mammal opens up analysis of a CNS function of considerable theoretical and practical significance.

Effects of the phytoestrogen coumestrol on locomotor and fear-related behaviors in female mice

Effects of the estrogenic plant isoflavonoid coumestrol (COUM) on locomotor activity and fear-related behaviors in both the absence and the presence of estradiol benzoate (EB) were examined in adult ovariectomized Swiss-Webster mice. In a running wheel paradigm conducted over 10 days, with treatment beginning 12 days prior to testing, daily subcutaneous (sc) injections of 10 microg of COUM did not influence locomotor activity, whereas even the low dose of EB supplied by sc Silastic implant (2.5 microg lasting 5 weeks) produced a significant increase in locomotor activity over animals receiving vehicle alone. In animals receiving both COUM and EB, locomotor activity was significantly diminished compared to the activity observed in animals receiving EB alone. The same animals were also tested in three behavioral paradigms having known activity and fear components (open field, dark/light transition, elevated plus-maze). COUM did not produce significant effects in these assays. However, in an associative fear learning paradigm (fear conditioning), COUM produced significantly less freezing ("fearfulness") in the conditioned fear task than with animals receiving vehicle or a low dose of EB, which did not differ from each other. Unlike EB, which has comparable affinities for estrogen receptor (ER) alpha and ER beta, COUM has a higher affinity for ER beta than for ER alpha. COUM might disrupt estrogen-enhanced locomotor activity by altering the dynamic by which EB acts on the brain through these two ER isoforms. The conditioned fear result suggests that COUM may produce additional effects through alternative pathways.

Estrogens, brain and behavior: studies in fundamental neurobiology and observations related to women's health

Mechanisms and consequences of the effects of estrogen on the brain have been studied both at the fundamental level and with therapeutic applications in mind. Estrogenic hormones binding in particular neurons in a limbic-hypothalamic system and their effects on the electrophysiology and molecular biology of medial hypothalamic neurons were central in establishing the first circuit for a mammalian behavior, the female-typical mating behavior, lordosis. Notably, the ability of estradiol to facilitate transcription from six genes whose products are important for lordosis behavior proved that hormones can turn on genes in specific neurons at specific times, with sensible behavioral consequences. The use of a gene knockout for estrogen receptor alpha (ERalpha) revealed that homozygous mutant females simply would not do lordosis behavior and instead were extremely aggressive, thus identifying a specific gene as essential for a mammalian social behavior. In dramatic contrast, ERbeta knockout females can exhibit normal lordosis behavior. With the understanding, in considerable mechanistic detail, of how the behavior is produced, now we are also studying brain mechanisms for the biologically adaptive influences which constrain reproductive behavior. With respect to cold temperatures and other environmental or metabolic circumstances which are not consistent with successful reproduction, we are interested in thyroid hormone effects in the brain. Competitive relations between two types of transcription factors - thyroid hormone receptors and estrogen receptors have the potential of subserving the blocking effects of inappropriate environmental circumstances on female reproductive behaviors. TRs can compete with ERalpha both for DNA binding to consensus and physiological EREs and for nuclear coactivators. In the presence of both TRs and ERs, in transfection studies, thyroid hormone coadministration can reduce estrogen-stimulated transcription. These competitive relations apparently have behavioral consequences, as thyroid hormones will reduce lordosis, and a TRbeta gene knockout will increase it. In sum, we not only know several genes that participate in the selective control of this sex behavior, but also, for two genes, we know the causal routes. Estrogenic hormones are also the foci of widespread attention for their potential therapeutic effects improving, for example, certain aspects of mood and cognition. The former has an efficient animal analog, demonstrated by the positive effects of estrogen in the Porsolt forced swim test. The latter almost certainly depends upon trophic actions of estrogen on several fundamental features of nerve cell survival and growth. The hypothesis is raised that the synaptic effects of estrogens are secondary to the trophic actions of this type of hormone in the nucleus and nerve cell body.

Estrogenic potential of certain pyrethroid compounds in the MCF-7 human breast carcinoma cell line

Estrogens, whether natural or synthetic, clearly influence reproductive development, senescence, and carcinogenesis. Pyrethroid insecticides are now the most widely used agents for indoor pest control, providing potential for human exposure. Using the MCF-7 human breast carcinoma cell line, we studied the estrogenic potential of several synthetic pyrethroid compounds in vitro using pS2 mRNA levels as the end point. We tested sumithrin, fenvalerate, d-trans allethrin, and permethrin. Nanomolar concentrations of either sumithrin or fenvalerate were sufficient to increase pS2 expression slightly above basal levels. At micromolar concentrations, these two pyrethroid compounds induced pS2 expression to levels comparable to those elicited by 10 nM 17ss-estradiol (fivefold). The estrogenic activity of sumithrin was abolished with co-treatment with an antiestrogen (ICI 164,384), whereas estrogenic activity of fenvalerate was not significantly diminished with antiestrogen co-treatment. In addition, both sumithrin and fenvalerate were able to induce cell proliferation of MCF-7 cells in a dose-response fashion. Neither permethrin nor d-trans allethrin affected pS2 expression. Permethrin had a noticeable effect on cell proliferation at 100 microM, whereas d-trans allethrin slightly induced MCF-7 cell proliferation at 10 microM, but was toxic at higher concentrations. Overall, our studies imply that each pyrethroid compound is unique in its ability to influence several cellular pathways. These findings suggest that pyrethroids should be considered to be hormone disruptors, and their potential to affect endocrine function in humans and wildlife should be investigated.

Estrogenic and antiprogestagenic activities of pyrethroid insecticides

Many pesticides possess hormonal activity and have thus been classified as endocrine disruptors. Pyrethroids are commonly used insecticides worldwide, but little has been done to characterize their hormone agonist/antagonist potential. We tested four frequently encountered pyrethroids, fenvalerate, sumithrin, d-trans allethrin, and permethrin, for estrogen and progesterone agonist/antagonist activities using the Ishikawa Var-I human endometrial cancer cell line and the T47D human breast cancer cell line. Both cell lines produce alkaline phosphatase as an indicator of hormonal activity. Fenvalerate and sumithrin demonstrated significant estrogenicity; at concentrations of 10 microM, these compounds achieved maximal activities comparable to that of 10 nM 17alpha-ethynylestradiol in Ishikawa Var-I cells. None of the four compounds showed statistically significant estrogen antagonist activity or acted as progestins. However, fenvalerate and d-trans allethrin significantly antagonized the action of progesterone in T47D cells. Through these hormonal pathways, exposure to certain pyrethroids may contribute to reproductive dysfunction, developmental impairment, and cancer.

Nomograms for determining the probability of axillary node involvement in women with breast cancer

We have previously reported that a history of pregnancy is independently associated with axillary node involvement in breast cancer patients. We have now studied additional women with breast cancer and have used our data and the logistic model to produce nomograms for determining the risk of axillary node involvement, based on tumor size, age, and number of pregnancies. There was an increase in the incidence of axillary node involvement in women with a history of pregnancy. To exclude the confounding effect that tumor size or age might have on node involvement, logistic regression was performed. Pregnancy, tumor size, and age were the three independent variables. History of pregnancy had a significant effect on node involvement (P = 0.036) that was independent of tumor size and age. Nomograms were constructed from these data. Surgeons do not perform an axillary dissection in every breast cancer patient. If the axilla is clinically negative and the tumor small, the surgeon, medical oncologist, and radiation oncologist may decide that a dissection need not be done. The nomograms in this article may allow for a more methodical choice of patients for axillary dissection. Moreover, a radiation oncologist might use the nomograms to help decide whether to irradiate an undissected axilla.

In vitro bioassays of non-steroidal phytoestrogens

Some of the isoflavonoids present in human diet as well as in urine are expected to exert biologic effects as they have been reported to bind to estrogen receptors and to be estrogenic in other species. This report describes the in vitro assessment of estrogenic effects of isoflavonoids using human endometrial cells and tissue. The relative estrogenic potencies (EC50 values) of estradiol, 3 dietary isoflavonoids (coumestrol, genistein and daidzein) and one of their metabolites (equol), were estimated by using a recently developed multiwell plate in vitro bioassay based on the estrogen-specific enhancement of alkaline phosphatase (AlkP) activity in human endometrial adenocarcinoma cells of the Ishikawa-Var I line. The maximal AlkP activity elicited by the isoflavonoids tested was as high as that achieved with estradiol and their effects were suppressed by the antiestrogens 4-hydroxytamoxifen and ICI 164,384. These results indicate that estradiol and the isoflavonoids exert their effects on AlkP by similar interactions with the estrogen receptor, with potencies depending on binding affinities. The estrogenic effect of equol was confirmed by another in vitro bioassay, based on the estrogen-stimulated enhancement of prostaglandin F2 alpha output by fragments of human secretory endometrium.

Identification of amino acid changes affecting yeast uroporphyrinogen decarboxylase activity by sequence analysis of hem12 mutant alleles

The molecular basis of the uroporphyrinogen decarboxylase defect in eleven yeast 'uroporphyric' mutants was investigated. Uroporphyrinogen decarboxylase, an enzyme of the haem-biosynthetic pathway, catalyses the decarboxylation of uroporphyrinogen to coproporphyrinogen and is encoded by the HEM12 gene in the yeast Saccharomyces cerevisiae. The mutations were identified by sequencing the mutant hem12 alleles amplified in vitro from genomic DNA extracted from the mutant strains. Four mutations leading to the absence of enzyme protein were found: one mutation caused the substitution of the translation initiator Met to Ile, a two-base deletion created a frameshift at codon 247 and two nonsense mutations were found at codons 50 and 263. Four different point mutations were identified in seven 'leaky' mutants with residual modified uroporphyrinogen decarboxylase activity; each of three mutations was found in two independently isolated mutants. The nucleotide transitions resulted in the amino acid substitutions Ser-59 to Phe, Thr-62 to Ile, Leu-107 to Ser, or Ser-215 to Asn, all located in or near highly conserved regions. The results suggest that there is a single active centre in uroporphyrinogen decarboxylase, the geometry of which is affected in the mutant enzymes.