September 21, 2017 - 12:09 pm 1 Comments
The NCCN Guidelines and Drug/Biologics Compendium have justifiably earned the Preferred position as the most influential resources in establishing the recommendations that serve as the basis for coverage policies by payers/MCOs. This preferred position has been critically important to patients, clinicians, provider institutions, and innovator biopharma companies. A major reason why the NCCN has assumed the mantle of authority has been the simple, direct, and elegant evaluative process and recommendations that have characterized the NCCN for years.
NCCN has recently released its Categories of Preference as a newly added feature of the NCCN Guidelines. The Categories of Preference will serve as a counterpart to the Categories of Evidence and Consensus and to the Evidence Blocks. It is herein acknowledged that the Categories of Preference have just been released and will likely evolve over time. For now and as written, the Categories of Preference likely will be a concern to patients seeking access to therapies based upon their individual characteristics, to providers making prescribing decisions, and to biopharma companies investing hundreds of millions of dollars or more to meet the needs of patients through innovation.
Briefly some concerns are described below:
Overall, the release of the Categories of Preference is not a positive for the patient, provider or biopharma communities;
The appearance and components of the Categories of Preference duplicate and seem to seek to atone for the failure of the Evidence Blocks to achieve utilization and influence across Stakeholders.
The definition of the category of “other recommended intervention” includes four negative components. As such, NCCN has turned negative with a formal negative recommendation category. Much like negative recommendations from the other four non-NCCN Compendia, one negative recommendation can trump positive recommendations from the other 4 Compendia. This one Category (“other recommended intervention”) may be used by payers to support denials or restrictive UM.
The aforementioned point cannot be taken lightly. NCCN Institutional memory should recall that originally the Category 2B was defined as “lower level evidence and non-uniform consensus”. This “non-uniform consensus” phrase resulted in CMS being silent on NCCN 2B recommendations and some plans like Anthem indicating that that payer would not cover NCCN Category 2B recommendations. Thus, negative definitions such as that under “other recommended intervention” can be and have been used by payers to deny.
The issue of how the stated institutional preference would help determine the Category of Preference is confusing. What happens if four institutions have one preference, 5 another preference, 6 yet another etc.?
The adjectives “preferred” and “optimal” are both used (interchangeably?) in the document including for the goal of “Provide guidance to users of the Guidelines on which recommendation(s) is considered optimal”. “Optimal” and “preferred” do not have the same meaning.
The NCCN answer about “how will NCCN know what each therapy will cost” is a tribute to circumlocution. The question is just not answered and this is a major issue.
The statement that a reason to develop the Categories of Preference is that “restrictive pathways are being developed in response to payer demands” seems to represent a tilt to the payer side; will the tilt be accentuated in the near future. This is concerning given the NCCN Guidelines traditionally have been the bastion of support for patients and providers.
The FAQ piece has inconsistencies and this seems to proceed from wanting to have it both ways.
The “Useful in Certain Circumstances” categories duplicates what is already in the Guidelines as the algorithms already drive to use of agents in specific subpopulations.
The interpretation of the Categories of Consensus, Evidence, and Preference along with the Evidence Blocks may lead to Babelian inconsistencies across the Guidelines and Compendium.
by Bill McGivney, PhD, President, McGivney Global Advisors
By Jay Grisolano, PhD and Stephanie Ritz, PhD
Approximately a quarter of breast cancer patients are classified as HER2-positive, according to Kantar Health’s 2016 Treatment Architecture data. 1 Effective targeted agents for the HER2 receptor, such as Herceptin (trastuzumab; Roche) and Perjeta (pertuzumab; Roche), have been developed for this particular patient segment and have greatly improved clinical outcomes for these patients. Despite these major advances, the need for improved early stage treatments still exists to keep disease from reaching an incurable stage. Currently, up to one in three early stage HER2+ patients treated with Herceptin in combination with chemotherapy eventually recur.2 In an effort to address this unmet need, Roche is conducting the Phase 3 APHINITY trial (NCT01358877) evaluating the combination of Herceptin, Perjeta, and chemotherapy in the adjuvant setting, hoping to improve upon these outcomes.
The triplet combination of Herceptin, Perjeta, and chemotherapy was initially approved in the HER2+ metastatic setting for first-line patients based on data from the Phase III CLEOPATRA trial, showing addition of Herceptin and Perjeta to chemotherapy improved median progression-free survival (PFS) and median overall survival (OS).3 In addition, the triplet has already demonstrated efficacy in early stage disease. The Phase II NEOSPHERE study (NCT00545688) compared Herceptin plus docetaxel with or without Perjeta in the neoadjuvant setting. Results reported at ASCO 2015 and published in Lancet Oncology showed a significantly improved complete response without an increase in cardiotoxicity (Herceptin / docetaxel: pCR 21.5%; Herceptin / docetaxel / Perjeta: pCR 39.3%, p=0.0063). Additionally, the risk of disease progression or recurrence was reduced by 31% and 40%.4 Based on these data, the triplet neoadjuvant regimen was granted approval in September 2013.
Now Roche is pushing to move their triplet combination into the adjuvant setting with APHINITY, results of which were presented today at the 2017 Annual Meeting of the American Society of Clinical Oncology (ASCO).5 APHINITY is an international, double-blind, placebo-controlled Phase III trial evaluating the efficacy of Herceptin plus chemotherapy with or without Perjeta in the adjuvant setting. The study enrolled 4,805 patients of lymph-node-positive and -negative status with confirmed HER2 positivity, as defined by IHC3+ or FISH-/CISH-positive. Patients received 6 to 8 cycles of chemotherapy with Herceptin (8mg/kg) and Perjeta (840 mg)/placebo, followed by Herceptin (6 mg/kg) and Perjeta (420 mg)/placebo alone every 3 weeks for one year (52 weeks) of treatment. At the time of presentation, the primary endpoint of invasive disease-free survival (IDFS) in the overall population was 92.3% in the triplet arm versus 90.6% seen in the control arm (p=0.045) at four years; however, subgroup analysis suggested that clinical benefit for IDFS appeared to be limited to node-positive and hormone receptor-negative cohorts. The disease-free interval and recurrence-free interval was modestly improved (DFI: 93.4% v. 92.3%, p=0.033; RFI: 95.2% vs. 94.3%, p=0.043); however, the addition of Perjeta did not improve the distant recurrence-free interval. No difference was noted in median overall survival at first interim analysis (97.7% v. 97.7%, p=0.467), but only one-quarter of data points needed for final analysis had been collected.
The most common Grade 3/4 adverse events associated with the triplet arm included neutropenia in 16.3% of patients (versus 15.7% in the control arm), febrile neutropenia in 12.1% of patients (versus 11.1% in the control arm, anemia in 6.9% of patients (versus 4.7% in the control arm), and diarrhea in 9.8% of patients (versus 3.7% in the control arm), which was predominately observed during the administration of chemotherapy. Cardiac toxicity was low and similar between the two arms.
Based on these data, it will be interesting to see how physicians adopt the triplet regimen in the adjuvant setting if it gains accelerated approval. Physician attendees vocalized great concerns during today’s session regarding the financial burden of adding Perjeta for a modest 1.7% IDFS benefit, despite its favorable tolerability profile. It may be essential for Roche to find specific subsets of patients who benefit most from the triplet in order for physicians to seriously consider integrating the addition of Pejeta into their practice. Another concern is whether Perjeta retreatment of a patient who recurs with metastatic disease will still be effective. It was mentioned that trials are being initiated to evaluate retreatment with Perjeta. Given the modest efficacy of APHINITY and the current absence of data supporting retreatment with the triplet beyond progression, physicians may to choose to save Perjeta until the metastatic setting. These issues may make it difficult to compete with Puma’s HER2 tyrosine kinase inhibitor, neratinib, which already has a PDUFA date set for July. Data supporting Puma’s application for approval is based on the Phase III ExteNET (NCT02400476) study, which showed a 2% benefit in 2-year DFS (2-year rate: 93.9% vs. 91.6%, p=0.0009) with a relatively well-tolerated toxicity profile.6 Subgroup analysis showed that neratinib may be more effective in ER-positive patients, which could help Puma find its niche in this space should neratinib receive approval.
Nevertheless, APHINITY did meet its primary endpoint showing that addition of Perjeta to Herceptin and chemotherapy as adjuvant treatment achieved a statistically significant improvement in IDFS. These data suggest that the triplet may become an option for adjuvant treatment of HER2+ early breast cancer.
By Liseth Parra, Ph.D., and Stephanie Hawthorne, Ph.D.
A very exciting therapeutic area that has been constantly highlighted throughout ASCO this year is the enormous potential for targeting genomic instability. While genomic stability is a major force of tumor growth, it provides a vulnerable point of tumorigenesis that can be used as an actionable target in clinical oncology. BRCA1 and BRCA2 are responsible for activating DNA damage response pathways as a result of DNA double-strand breaks (DSB) and thus play an important role in maintaining the genetic stability of cells.
Hereditary (germline) mutations in one copy of either the BRCA1 or BRCA2 gene (gBRCA1/2) are associated with a high risk of developing primarily breast and ovarian cancer and represent one of the greatest unmet needs in gynecologic cancers. These mutations account for about 5% to 10% of all breast cancers and about 15% of all ovarian cancers1 and are particularly vulnerable to poly(ADP-ribose) polymerase (PARP) inhibition. PARP proteins normally function in the repair of DSB, and it’s presumed that their inhibition leads to the breakdown of the DNA machinery involved in DSB repair that cannot take place in BRCA1/2 deficient cells, a concept referred to as synthetic lethality.
Within the past two years, three PARP inhibitors gained regulatory approval for the treatment of advanced ovarian cancer – Lynparza® (olaparib, AstraZeneca), Rubraca® (rucaparib, Clovis Oncology), and Zejula™ (niraparib, Tesaro). Additionally, four PARP inhibitors – Lynparza, talazoparib (Pfizer / Medivation), veliparib (ABT-888, AbbVie), and Zejula – are in Phase III development for locally advanced or metastatic BRCA1/2 mutated breast cancer, starting a race to see which agent will be first to market in this indication.
AstraZeneca, as an output of its 2005 acquisition of the British biotechnology company KuDOS, became the first pharmaceutical company to launch a PARP inhibitor, with Lynparza obtaining accelerated approved as a monotherapy for the treatment of gBRCA ovarian cancer patients previously treated with three or more lines of chemotherapy. Following this initial launch, AstraZeneca has an extensive life-cycle management development plan in-place for Lynparza, with its eyes on breast cancer as the next indication for approval.
In the Plenary session at the 2017 ASCO conference, the first randomized results to support its development in breast cancer were presented1,2. The Phase III OLYMPIAD trial (NCT02000622) evaluated Lynparza in comparison to “physician’s choice” chemotherapy (capecitabine, vinorelbine, eribulin) in 302 metastatic HER2- breast cancer patients with deleterious or suspected gBRCA mutations.
Patients could have had up to two prior lines of chemotherapy and must have received prior anthracycline and taxane, and hormone receptor-positive patients must have received at least one prior line of hormone therapy unless considered unsuitable; ultimately, one-third of enrolled patients were treated on-study as first-line therapy for metastatic disease, one-quarter as third-line, and the remainder as second-line.
Late breaking results for this trial presented at ASCO demonstrated for the first time positive data for a PARP inhibitor in metastatic breast cancer with a statistically-significant improvement in progression-free survival (PFS) compared to standard chemotherapy and a meaningful improvement in health-related quality of life (HRQoL) in these patients.
The data showed a statistically significant PFS gain of nearly 3 months in patients treated with Lynparza (300 mg BID; n=205) versus chemotherapy (n=97) (7.0 vs 4.2 months, respectively; HR 0.58; p=0.0009). Although immature, overall survival (OS) was not statistically different between the two arms (19.3 versus 19.6 months, HR 0.90; p=0.5665). The ORR in the two arms was 60% versus 29%, respectively, with more complete responses in patients treated with Lynparza (9%) versus chemotherapy (2%). Although exploratory, data suggest similar level of benefit regardless of patient’s prior exposure to platinum or other chemotherapy, but suggest greater benefit in patients with triple-negative breast cancer (PFS HR 0.43) than in patients with hormone receptor-positive disease (PFS HR 0.82).
Fewer patients in the Lynparza arm than in the chemotherapy arm experienced Grade 3/4 adverse events (AE; 36.6% vs. 50.5%) and slightly fewer AE-related treatment discontinuations (4.9% vs. 7.7%). The most common Grade 3/4 adverse events in the Lynparza and chemotherapy arms were anemia (16% vs. 4%) and neutropenia (9% vs. 26%). As a secondary endpoint, the results showed that Lynparza led to a modest but clinically meaningful improvement in HRQOL compared to chemotherapy (estimated difference in EORTC QLQ-C30 global HRQOL score of 7.5 points, p=0.0035) and significantly longer time to deterioration of HRQOL (median not reached vs. 15.3 months; HR 0.44, p=0.0043).
These results represent the first positive Phase III trial for a PARP inhibitor in breast cancer and the discussant, Dr. Allison Kurian, had an overall positive outlook stating that these results” are practice-changing.” However, the magnitude of the benefit (3 months difference) and lack of OS benefit could raise questions by others in the field about the clinical meaningfulness of these results.
It is encouraging that the incidence of adverse events was lower, fewer treatment discontinuations were needed, and HRQOL was improved with Lynparza, but the magnitude of each of these improvements seems to run on the edge of meaningful from the a clinical perspective. When the clinical benefits run along that edge, commercial considerations have a greater likelihood of taking hold, and some physicians (and payers) may question whether the benefit with Lynparza compared to generic chemotherapy justifies the costs. Despite these concerns, Lynparza stands a good chance of gaining regulatory approval, so these debates may ultimately play out in the clinic and everyday practice.
While it may launch as a monotherapy, ultimately the greater movement forward with PARP inhibitors lies in alternative development strategies. Correlative biomarkers for PARP inhibition (beyond gBRCA1/2) may help to further identify patients that will benefit the most from Lynparza. Additionally, the double-hit approach of single-agent PARP inhibitors in gBRCA mutated disease may not be robust enough, and we could ultimately find more effective strategies with a targeted combination approach – this includes combinations with chemotherapy or radiotherapy (which can induce DSBs)or with other mechanisms of action.
The closest drug taking this approach is veliparib, which is being studied in the Phase III BROCADE 3 trial (NCT02163694), which is enrolling HER2- breast cancer patients harboring deleterious BRCA1/2 or gBRCA1/2 mutations with less than two lines of prior therapy, who will be randomized to treatment with carboplatin and paclitaxel in combination with veliparib or placebo. Other PARP inhibitors in development in breast cancer are being developed as monotherapies - talazoparib in the Phase III EMBRACA trial (NCT01945775 and Zejula in the Phase III BRAVO trial (NCT01905592). The OlympiAD results, Lynparza is now officially ahead of the competition and could emerge as a potential agent of change in gBRCA breast cancer management, creating a new standard of care for the gBRCA1/2-mutated patient population. With so many competitors hot on its heels, it will be critical for Lynparza to get a solid push off of the starting block in order to put as much distance between itself and the others in this Olympic race in the breast cancer market.
By Jay Grisolano, PhD and Emily Benesh, PhD
Today was an exciting day for physicians and patients experiencing newly-diagnosed metastatic prostate cancer. At the ASCO 2017 meeting, data were presented on LATITUDE, a randomized Phase III trial comparing Zytiga (abiraterone acetate, Janssen)/prednisone plus androgen deprivation therapy (ADT) versus placebos plus ADT in newly-diagnosed high-risk metastatic hormone-naive prostate cancer (mHSPC), concurrent with a publication in the New England Journal of Medicine1, which will transform the prostate cancer treatment landscape.
The prostate cancer competitive landscape is one of the most highly dynamic and complex in all of oncology. In both early stage high-risk and treatment-naïve metastatic disease, hormone therapies are the mainstay systemic treatments for newly-diagnosed patients. In fact, ADT has been standard of care for newly diagnosed mHSPC patients for decades. Despite the durable efficacies and minimal side effects of hormonal agents, some patients fail treatment and become castrate-resistant.
Zytiga improved the treatment options for metastatic castrate resistant prostate cancer (mCRPC) patients when it was initially approved by the FDA as a second-line therapy in 2011. The label was expanded in late 2012 to include front-line treatment in the mCRPC space. Zytiga is a second-generation hormone therapy that inhibits androgen-production in the microenvironment by blocking CYP17A1 protein production in the testes, adrenal glands, and the tumor itself2.
Since their initial approvals, Zytiga and its second-generation hormone therapy competitor, Xtandi (enzalutamide, Astellas/Medivation; approved in April 2012), an androgen receptor antagonist, have eaten up mCRPC market-share. According to Kantar Health’s CancerMPact, Treatment Architecture module, in 2016, Zytiga and Xtandi captured over 70% share of the United States, EU5, and Japanese first-line mCRPC markets for both asymptomatic and symptomatic diseases2; Zytiga sees utilization in over 100 other countries across the world3.
Having demonstrated efficacy in the mCRPC setting, Janssen now aims to move Zytiga even earlier in the treatment paradigm: into the metastatic, hormone-naïve setting. No other next-generation hormone therapies or cutting-edge systemic therapies have been approved for use in this space2.
In 2016, nearly 70% of mHSPC patients in the U.S. were treated with traditional hormone therapies [Luteinizing Hormone Releasing Hormone (LHRH) agonists or anti-androgens]2. Importantly, roughly 20% of patients received off-label second-generation hormone therapy (namely, Zytiga or Xtandi), either as a monotherapy or in combination with traditional hormone treatments2. This suggests that awareness about the utility of second-generation hormone therapies is high and that uptake of these agents in the hormone-sensitive metastatic prostate cancer space will be rapid upon approval.
Janssen has strong previous data supporting the use of second-generation hormone therapies in the metastatic hormone-sensitive space. In a Phase II study, Zytiga and prednisone were added to a first-generation LHRH agonist in 37 patients with high-risk localized prostate cancer4. High-risk was defined as patients with Stage T1c/T2 disease and a Gleason score of at least eight, or patients with at least Stage T2b/T2c disease with a Gleason score of at least 7 and a PSA level > 10 ng/mL. Patients were randomized 2:1 to receive either: Zytiga/prednisone/LHRH agonist or LHRH agonist alone. The percentage of patients with preoperative prostate-specific antigen (PSA) less than 0.1 ng/mL was significantly increased in the Zytiga/prednisone/LHRH agonist arm compared to the control arm (68% versus 0%, p<0.0001). In addition, there was a numerical increase and decrease, respectively, in the percent of patients with near complete cytoreductions (< 6 mm scattered cells; 24% versus 8%, p=0.10) and lymph node infiltration (25% versus 50%, p=0.10) in the Zytiga/prednisone/LHRH agonist arm compared to controls. The safety profile was similar to what has been observed in the mCRPC setting. These results showed that the addition of Zytiga and prednisone to standard hormonal therapy can add significant efficacy even in a hormone-sensitive population.
In February 2013, Janssen initiated a randomized, double-blind, global (including Japan) Phase III trial (LATITUDE, CR100900; NCT01715285) of Zytiga in newly-diagnosed, high-risk metastatic patients who are hormone-naïve. Patients must have been diagnosed within three months of randomization and have not yet received any treatment for their prostate cancer [with the exception of fewer than three months of ADT received since diagnosis]. Patients were considered high-risk if they met at least two of the following requirements: Gleason score of at least 8, presence of at least three lesions on a bone scan, or presence of measurable visceral metastasis.
Patients were randomized 1:1 to receive either: Zytiga/prednisone/ADT (either LHRH agonists or surgical castration) or placebo/ADT. Co-primary endpoints were overall survival (OS) and radiographic progression-free survival (rPFS); secondary endpoints included time to next skeletal-related event, time to prostate-specific androgen progression, time to next therapy, time to initiation of chemotherapy, changes in mRNA and miRNA levels, and safety. The trial intended to accrue 1,200 patients and have three analyses: interim one – after 50% of total projected events (426 events), interim two – after 65% of events (554 events) and a final analysis – after a total of 852 events. The trial would be considered positive if P values for overall survival were lower than 0.011, 0.022 and 0.049 for interim one, two and the final analysis, respectively.
Data from the first interim analysis of the LATITUDE trial were reported in a plenary session on June 4, 2017 at the 53rd annual meeting of the American Society for Clinical Oncology (ASCO) held in Chicago, Illinois5. The trial had accrued 1,199 patients. At the first interim analysis 406 events had occurred: 169 occurred in the Zytiga arm, while 237 events occurred in the placebo controls. The median follow-up time was 30.4 months.
As the discussant, Eric Small, MD, stated, “LATITUDE was a profoundly positive study.”
The Zytiga/prednisone/ADT arm had a 38% reduction in risk of death when compared to the control (OS: not reached vs. 34.7 months, HR: 0.62, 95% CI: 0.51-0.76; P<0.0001). The 3-year OS rate was 66% in the Zytiga arm versus 49% in the placebo arm. Overall survival benefit persisted across all of the pre-specified subgroups (e.g., ECOG status and visceral disease).
Additionally, patients in the Zytiga arm experienced a 53% reduction in the risk of radiographic progression (rPFS: 33.0 vs. 14.8 months, HR: 0.47, 95% CI: 0.39-0.55; P<0.001). All secondary endpoints were statistically improved in the Zytiga arm. Zytiga patients had an astounding 70% reduced risk of time to PSA progression (33.2 vs. 7.4 months, HR: 0.30, 95% CI: 0.26-0.35; P<0.0001). As mentioned by the discussant, given the concern of patients regarding PSA levels, this finding is expected to greatly improve quality of life for this population. Patients in the Zytiga arm receive fewer subsequent, post-study, life-prolonging therapies, suggesting that OS benefit was due to Zytiga and not subsequent treatment. Adverse events in the Zytiga arm were manageable and consistent with studies in mCRPC patients. Elevated rates of hypertension, hypokalemia, ALT and AST increases, and cardiac disorders were observed with Zytiga use. Taken together, these data were sufficiently positive to support early termination of the trial, un-blinding of participants and cross-over of placebo receiving patients to the Zytiga arm.
On May 26, 2017 Janssen announced that they filed for approval of Zytiga for mHSPC to the Ministry of Health Labor and Welfare in Japan3. Given the unmet need of development of resistance to ADT in mHSPC patients, the previous enthusiastic global uptake of Zytiga in metastatic CRPC, and the current off-label use of second-generation hormones in hormone-sensitive disease, it is expected that Zytiga will have rapid uptake upon approval for mHSPC patients.
The approval of Zytiga will have practice-altering effects on the metastatic prostate cancer landscape; Zytiga will likely become the new standard of care for high-risk mHSPC patients. Furthermore, data presented yesterday from the Phase III STAMPEDE trial suggest that Zytiga will soon move into early-stage/low-risk disease, as well. In a population that was 48% low-risk (N0M0 or N+M0) overall survival was significantly improved (HR: 0.63, 95% CI: 0.52-0.76; P=0.00000015)6. Thus, the reach of Zytiga is expected to continue expanding in earlier settings of the prostate cancer treatment landscape.
It is important to note that data reported from the Phase III CHAARTED study showed that addition of docetaxel to ADT in castration-naïve metastatic prostate cancer significantly improved survival outcomes in the overall population of patients (51.2 versus 34.4 months, HR: 0.73, P<0.0001) as well as in those with high-volume disease7, providing evidence for use of docetaxel in this setting Questions remain about how physicians will use Zytiga versus docetaxel in metastatic prostate cancer patients. Cross-trial comparisons between LATITUDE and CHAARTED show nearly identical improvements in risk of death for patients receiving Zytiga versus docetaxel with ADT (HR of 0.62 versus 0.63, respectively). Physicians may prefer the improved toxicity profile of Zytiga over docetaxel, but may also consider cost in treatment choice. Notably, duration of therapy with Zytiga has a median of 33 months in LATITUDE, while docetaxel was given for 6 cycles, or 4.5 months, in CHAARTED, potentially impacting cost as well as convenience to the patient. Another concern is that if Zytiga is used in an earlier setting, what will be used for subsequent treatment if the patient becomes mCRPC. Only time will tell regarding whether physicians will re-treat with the same general mechanism of action, the sequencing of the agents, and whether a space will open for another novel mechanism of action for use with progressed patients.
While Zytiga is poised to be first-to-market, several other second-generation agents have ongoing Phase III trials for use in first-line hormone-sensitive metastatic prostate cancer. Phase III trials testing combinations of androgen-deprivation therapy plus either apalutamide (ARN-509/JNJ-927, Aragon Pharmaceuticals/Janssen; TITAN/NCT02489318 initiated November, 2015), Xtandi (NCT02677896 initiated March 2016), and darolutamide (ODM-201, Bayer; ARASENS/NCT02799602 initiated June 2016) have the promise of bringing several competitors to the market in the near future. Competition in this space is likely to be intense, but Zytiga may have an advantage by being first-to-market.
In any case, today was a bright day for the field of prostate cancer and for prostate cancer patients.