BRAFV600E-mutant metastatic NSCLC: disease overview and treatment landscape

In this review, we cover the current understanding of BRAF mutations and associated clinical characteristics in patients with metastatic NSCLC, approved and emerging treatment options, BRAF sequencing approaches, and unmet needs. The BRAFV600E mutation confers constitutive activity of the MAPK pathway, leading to enhanced growth, proliferation, and survival of tumor cells. Testing for BRAF mutations enables patients to be treated with therapies that directly target BRAFV600E and the MAPK pathway, but BRAF testing lags behind other oncogene testing in metastatic NSCLC. Additional therapies targeting BRAFV600E mutations provide options for patients with metastatic NSCLC. Emerging therapies and combinations under investigation could potentially overcome issues of resistance and target non-V600E mutations. Therefore, because targeted therapies with enhanced efficacy are on the horizon, being able to identify BRAF mutations in metastatic NSCLC may become even more important.

limited and conflicting for patients with BRAF V600 -mutant NSCLC [22][23][24] .Current targeted therapeutic approaches have limited efficacy in patients with non-V600 BRAF mutations, and most clinical trials have focused primarily on the V600E mutation since its discovery 11,17,18,25 .There remains a need to better understand the incidence, impact, and management of brain metastases; mechanisms of resistance; optimal sequencing; and other patient-(e.g., smoking history) or disease-related factors (e.g., PD-L1 expression) that influence treatment outcomes of BRAF-mutant metastatic NSCLC.This manuscript provides a review of BRAF-mutant metastatic NSCLC and the therapeutic landscape with particular emphasis on targeted therapies for the V600E mutation.

Clinical characteristics
While BRAF mutations are predominantly found in adenocarcinomas (>85%), there is no clear association of BRAF mutation status with other patient characteristics, such as age, ethnicity, and sex 6,8,9,26 .Epidemiological patterns are difficult to identify since BRAF mutations occur in a small number of patients with advanced NSCLC 5 .One study reported that BRAF V600E mutations were significantly more common in females (P < 0.001), but that finding was not consistent with other studies 6,10,26 .While data are limited, studies have also shown that the incidence of BRAF mutations is lower in Asian patients than white patients 6,26,27 .Smoking history can be associated with BRAF mutations, as well as KRAS, MET, and other mutations 9,28,29 .However, a few studies have also suggested that BRAF V600E is less associated with smoking history than other BRAF mutations 26,30,31 .Therefore, all patients with advanced NSCLC regardless of smoking history should undergo broad-based mutation testing, including BRAF 5,21,22 .Programmed death ligand 1 (PD-L1) expression and tumor mutation burden (TMB) are potentially important indicators of immunotherapy response and are frequently elevated in BRAF-mutant NSCLC 24 .The prognostic implication of BRAF mutation status is inconclusive due to small patient numbers and conflicting studies 8,26,28 .However, a few studies reported worse outcomes with chemotherapy in patients with V600E mutations versus wild type (WT) BRAF and conflicting results for V600E versus non-V600E mutations 6,[31][32][33] .In addition to the small patient number, discrepancies in patient characteristics and prognosis could be attributed to the heterogeneity of BRAF mutations in NSCLC 33,34 .
Classification of BRAF mutations BRAF mutations are heterogenous with distinct mutation classes that each yield a functionally diverse BRAF protein, resulting in different clinical features and treatment ramifications 33,35 .BRAF-mutant proteins interact with and activate the MAPK pathway in various ways and to different degrees (Fig. 1) 25,35,36 .In the MAPK pathway, the signaling cascade begins with growth factor binding to receptor tyrosine kinase (RTK) which facilitates RAS binding GTP 14,25,36 .Activated RAS promotes a cascade of activation starting with RAF family kinases (ARAF, BRAF, CRAF), which form activated RAF homo-or heterodimers.RAF kinases activate MEK, which subsequently activates extracellular signal-regulated kinase (ERK).ERK has multiple downstream targets, including transcription factors that promote cell growth, proliferation, and survival.Negative feedback loops from ERK also regulate the pathway 25 .
There are over 200 identified BRAF mutations categorized into three classes based on dimerization status, RAS-dependence, and kinase activity level 5,37 .Class I BRAF mutations occur on codon 600 (BRAF V600 ), substituting the valine to a glutamic acid, lysine, aspartic acid, arginine, or methionine (V600E, V600K, V600D, V600R, and V600M mutations, respectively) and can biochemically transform BRAF into a RASindependent constitutively active monomer 25,36 .BRAF V600E is the most prevalent class I mutation and accounts for ~30-50% of all BRAF mutations in NSCLC 6,7,33 .Class I mutant proteins have a high level of kinase activity, and BRAF V600E is ~500-fold times more active than WT BRAF, leading to increased cellular proliferation 4,36 .Class I mutations all occur at codon 600 aberrantly activating monomers, but class II and III mutations occur at various other non-600 codons and form dimers 5,25,36 .Class II BRAF mutations span from G464 to K601, which includes the activation segment and P-loop domains of BRAF.These mutations, along with fusions and in-frame    25,38 .To overcome that feedback inhibition, class III BRAF mutations are often observed with concurrent RAS activating mutations 25,33,38 .While these coexisting mutations often occur in melanoma cancers, CRC and NSCLC cancers have fewer cases 25,38 : one study in NSCLC tumors reported coexisting RAS mutations in 22% of class III mutations (n = 54), including 42% of kinasedead mutations (n = 19) 33 .This may be explained by basal RTK activation and subsequent RAS activity that is sufficient to support these class III mutants 25 .
Currently approved BRAF inhibitors effectively inhibit only class I mutant proteins and show substantially less efficacy against BRAF-mutant dimers (Fig. 1) 25 .BRAF monomer inhibitors binding to dimers can cause paradoxical transactivation of the unbound RAF WT protomer, enabling MEK/ERK signaling and subsequent disease progression 12,13,25 .Upstream inhibition may be an effective therapeutic strategy for treatment of class III mutations 25,33,38 .A better understanding of BRAF mutations, especially class II and III mutations, may enable the rational design of new targeted therapies and the development of next-generation drug combination strategies to treat BRAF-mutant cancers, including NSCLC 25,33 .

BRAF testing
Guidelines recommend that all patients with advanced non-squamous NSCLC undergo broad-based molecular testing to identify molecular drivers-including but not limited to BRAF V600 mutations 21,22 .Recommended and approved molecular testing assays include polymerase chain reaction (PCR) and next-generation sequencing (NGS) [39][40][41] .PCR offers rapid turnaround, reproducibility, specificity, sensitivity, and lower costs, but it is a single-gene assay typically limited to detection of V600E mutation 39 .Panelbased NGS has gained popularity for the ability to simultaneously test multiple genes, including BRAF, for both V600E and non-V600E mutations, which is more cost-effective than sequential single-gene assays and uses relatively little tumor tissue 39,40 .Availability of sufficient tumor tissue is a major constraint when testing for the numerous actionable mutations in NSCLC, so approaches that conserve tissue while providing a full molecular profile are being evaluated 42 .Liquid biopsy, a noninvasive and more rapid alternative to tissue biopsies that collects blood-based biomarkers-including circulating tumor DNA (ctDNA), can be used to detect genomic alterations 5,43 .Immunohistochemistry (IHC) is a highly sensitive and specific diagnostic test that uses the monoclonal antibody VE1 to detect BRAF V600E -mutant proteins 39,44 .While there are no current standard recommendations for IHC in BRAF V600E mutation detection, guideline recommendations for other oncogenes suggest that IHC be confirmed with a molecular test 22,39 .BRAF testing rates lag behind other driver oncogene testing rates, likely due to limited tissue availability and the fact that other actionable biomarkers (e.g., EGFR, ALK, PD-L1) are prioritized for testing when a sequential selective approach is used 45 .Given the demonstrated efficacy of BRAF V600E inhibitors in NSCLC, there is a need to improve the rate of BRAF testing so the results can be applied to clinical decision-making.
While BRAF inhibitor monotherapy is initially effective, acquired resistance enables reactivation of the MAPK pathway, limiting the clinical utility of monotherapy 14,61 .In addition, BRAF monomer inhibitors can cause paradoxical activation of the MAPK pathway in BRAF WT cells, which has been associated with hyperproliferative cutaneous events, including squamous cell carcinoma (SCC) and keratoacanthoma [12][13][14]57 . BRA monomer inhibitors were combined with downstream MEK inhibitors to overcome resistance and paradoxical activation of the MAPK pathway, which increased efficacy and tolerability, resulting in several combination therapies being approved for unresectable metastatic BRAF V600 -mutant melanoma (e.g., dabrafenib plus trametinib, vemurafenib plus cobimetinib, encorafenib plus binimetinib) 14,[62][63][64] .The FDA granted accelerated approval of dabrafenib plus trametinib treatment for previously treated unresectable or metastatic solid tumors with the BRAF V600E protein in patients with no alternative treatment options 65 .This approval was supported by the ROAR study, which enrolled 206 patients into eight cohorts, each a different BRAF V600E -mutant rare cancer.ORR ranged from 0% for gastrointestinal stromal tumor (n = 1) to 89% for hairy cell leukemia (n = 55); ORR was ≥33% for the remaining cohorts.

BRAF and MEK inhibitors in metastatic NSCLC
Clinical trials with BRAF inhibitor monotherapy in patients with BRAF V600mutant NSCLC are summarized in Table 1 10,11,55,66 .In several trials, vemurafenib monotherapy was an effective treatment for treatment-naive (ORR: 37.5%; median PFS: 12.9 months) and previously treated patients (ORR: 37.0-44.8%;median PFS: 5.2-6.1 months) with BRAF V600 -mutant NSCLC 11,66 .In one study, serious AEs occurred in 63% of patients, most commonly cutaneous SCC (15%) and keratoacanthoma (15%) 66 .Dabrafenib monotherapy was effective for previously treated patients (ORR: 33%; median PFS: 5.5 months) with BRAF V600E -mutant metastatic NSCLC; however, data were limited for treatment-naive patients because of a decision to prioritize the combination with trametinib with the expectation of improved response rates 10 .Pyrexia was the most common any-grade AE (36%), including grade 3 occurrences in 2% of patients.Pyrexia led to dose interruption or reduction in 11% and 4% of patients, respectively.Serious AEs occurred in 42% of patients, including pyrexia in 6% of patients.
Primary and acquired drug resistance to BRAF-targeted therapy Baseline concurrent mutations prior to receipt of targeted therapy have been identified in 22-30% of patients with BRAF V600E -mutant NSCLC and may be a cause of primary resistance 17,61,69 .Common concurrent mutations included alterations in the TP53 and SETD2 genes and the PI3K (e.g., PIK3CA E545K, PTEN R14K) and MAPK (e.g., KRAS G12C) pathways 17,18,61,69 .The incidence of concurrent TP53 or RAS gene family mutations was higher with BRAF class II or III mutations than class I mutations 31,33,70   the presence of a concurrent mutation in TP53, PIK3CA, KRAS, or PTEN was associated with poorer outcomes 17,61,69 .In a study evaluating dabrafenib plus trametinib for BRAF V600E -mutant metastatic NSCLC, patients with a concurrent mutation in the PI3K pathway (n = 4) had a median OS of 5.4 months compared with 22.7 months in patients without an identified concurrent mutation (n = 34) 17 .In the PHAROS trial, which evaluated encorafenib plus binimetinib for BRAF V600E -mutant metastatic NSCLC, concurrent mutations were identified, with the most common being SETD2 and TP53 (43%, each); however, there was no association between concurrent mutations and outcome 18 .As most of this data comes from studies with small numbers of patients, further research is necessary to understand the impact of concurrent mutations in patients with BRAF V600E -mutant NSCLC 17,18,69 .
The mechanisms of acquired resistance to BRAF inhibitors, alone or combined with MEK inhibitors, are poorly understood, and there is no standardized sequential treatment strategy 21,61 .While data are limited, acquired resistance appears to occur through bypassing or reactivating the MAPK pathway 14,61 .Bypassing the MAPK pathway and activating a parallel pathway (e.g., PI3K/AKT) can lead to disease progression.A PTEN frameshift mutation that could potentially activate the PI3K pathway was identified in a patient with BRAF V600E -mutant NSCLC that progressed on dabrafenib 71 .In preclinical BRAF V600E -mutant lung cancer models, the presence of a cooperating mutation silencing RBMS3, a regulator of the WNT pathway, promoted resistance to dabrafenib plus trametinib 72 .
Reactivation of the MAPK pathway can occur in BRAF-dependent or -independent manners.Resistance to targeted therapies often occurs due to secondary mutations or epigenetic changes in the target gene, and an aberrant splice variant of BRAF was identified in BRAF V600E NSCLC cells that acquired resistance 73,74 .However, secondary BRAF mutations may be rare, as none were discovered in several resistance studies with BRAFmutant NSCLC and melanoma 61,73,75 .BRAF-independent reactivation of the MAPK pathway includes alterations either upstream or downstream of BRAF 14 .Mutations in RAS (NRAS, KRAS) were discovered in a few studies 61,71 .Strong evidence came from a study that compared ctDNA sequencing at diagnosis and disease progression for 35 patients with BRAFmutant NSCLC who progressed on either BRAF inhibitor monotherapy or dabrafenib plus trametinib 61 .RAS mutations (KRAS G12V; KRAS Q61R; NRAS Q61R) were present at disease progression and not diagnosis, which suggests mutation occurred during treatment.Resistance studies of patients with BRAF V600E -mutant melanoma suggested that upregulation of RAS or overexpression of ARAF and CRAF could alleviate BRAF-dependence in tumor cells 73,75 .Downstream mutations in MEK1 were also identified in patients with BRAF V600E -mutant NSCLC that progressed on dabrafenib plus trametinib 71 .Further understanding of acquired resistance mechanisms is critical to inform optimal sequencing and providing insight for evaluation of combination approaches or next-generation target therapies.

Immunotherapy
Compared with studies with BRAF and MEK inhibitors, data concerning efficacy and safety of immunotherapy in patients with BRAF V600E -mutant NSCLC are limited; studies have not prospectively enrolled patients with BRAF V600 mutations, and immunotherapy is not specifically approved for patients with this mutation [76][77][78][79] .Evidence for efficacy of immunotherapy is derived from conflicting, small, retrospective studies (Table 2) 23,24,77 .In a multi-cohort retrospective study of immunotherapy in patients with oncogene-driven advanced lung cancer, patients (n = 10) with BRAFmutant NSCLC in one cohort had a significantly longer median PFS (7.4 months) than patients with other common oncogene drivers (versus KRAS 2.8 months; HR 0.36, 95% CI, 0.14-0.88;P = 0.026) 24 .In the other cohort, PFS was longer in patients with V600E mutations (n = 30; 9.8 months) versus non-V600E mutations (n = 35; 5.4 months).However, in another retrospective study, similar PFS (2.1-3.4 months) was reported across oncogenes with immunotherapy 77 .In that study, two trends emerged in the BRAF cohort (n = 43).PFS was longer in patients who previously or currently smoked (4.1 months) versus had never smoked (1.9 months) and with non-V600E mutations (4.1 months) versus V600E mutations (1.8 months).According to guidelines of the European Society for Medical Oncology, patients with BRAF V600 -mutant metastatic NSCLC that progresses on BRAF plus MEK inhibitor should receive immunotherapy with optional chemotherapy (in patients with smoking history) or chemotherapy with optional immunotherapy (in patients without smoking history) 22 .

Chemotherapy
Prior to the development of targeted therapy for BRAF V600E -mutant metastatic NSCLC, platinum-based chemotherapy was the standard of care 68,78 .However, retrospective studies reported that patients with BRAF V600Emutant NSCLC had poorer outcomes with platinum-based chemotherapy than those with BRAF WT NSCLC (Table 2) 6,32 .While those retrospective studies also reported a shorter PFS in patients with V600E mutations (4.1-5.2 months) versus non-V600E mutations (6.4-8.9 months), another retrospective study reported that carboplatin-pemetrexed in patients with treatment-naive BRAF-mutant metastatic NSCLC resulted in longer PFS in patients with class I mutations (6.2 months) versus class II or III mutations (3.3 months and 4.9 months, respectively) 33 .Additionally, in a multiinstitutional prospective lung cancer screening project, median PFS with platinum-containing chemotherapy was longer in patients with class I mutations (11.5 months) than in those with class III mutations (5.3 months) 31 .Several trials demonstrated that BRAF monotherapy or BRAF plus MEK inhibitor therapy was effective in patients who had progressed on chemotherapy 10,15,18,66 .Chemotherapy remains a second-line recommendation for patients with a BRAF V600 mutation 21,22 .

Immunochemotherapy
The combination of immunotherapy and chemotherapy is approved for first-line treatment of patients with metastatic NSCLC but not specifically for patients with BRAF V600E mutations 80 .In a phase 3 trial of treatment-naive patients with metastatic non-squamous NSCLC, the combination of pembrolizumab, pemetrexed, and a platinum resulted in significantly longer median PFS compared with chemotherapy alone (8.8 versus 4.9 months; HR = 0.52; P < 0.001) 81 .In a small retrospective study in China, immunotherapy plus chemotherapy (n = 9) in treatment-naive patients with BRAF V600E -mutant advanced NSCLC resulted in a significantly longer median PFS compared with chemotherapy or targeted therapy (n = 20; 18.5 versus 4.1 months; P = 0.0098) 82 .This efficacy benefit with immunochemotherapy was not observed in later lines.Another retrospective study showed similar efficacy with immunotherapy-based treatments in patients with advanced NSCLC with or without BRAF mutations; median PFS was 8.4 months in both patient populations 83 .In the BRAF cohort, median PFS was similar for V600E and non-600E mutations (10.0 versus 8.0 months).Median PFS was longer in the first line than in subsequent treatment lines in patients with WT (12.8 versus 5.6 months) and BRAF-mutant (11.2 versus 4.0 months) NSCLC.These studies suggest that immunotherapy-based treatments are an option for patients with BRAF V600E -mutant advanced NSCLC 82,83 .
A meta-analysis of triplet therapies compared with doublet therapy or monotherapy for melanoma revealed that triplet therapy significantly improved PFS and OS but was associated with increased frequency of immune-related AEs, including hypothyroidism, arthralgia, myalgia, alanine aminotransferase elevations, aspartate aminotransferase elevations, asthenia, and pyrexia 90 .Triplet therapy did not increase the overall incidence of AEs or grade ≥3 AEs.The increased incidence of AEs should be considered when determining the optimal combination of immunotherapy and targeted therapy 88,90 .

Next-generation BRAF inhibitors
Next-generation BRAF inhibitors target dimerization since it is an essential component of activation for WT and many mutant BRAF kinases, plays a role in resistance mechanisms to BRAF inhibitors, and is associated with AEs 91 .These drugs were developed following two main strategies aimed at inhibiting mutant RAF while preventing paradoxical activation and common acquired resistance mechanisms 92 .Type II pan-RAF inhibitors bind the active conformation of RAF monomers and dimers 92,93 .Despite the name of pan-RAF, at least three of these agents (naporafenib [LXH254], tovorafenib [TAK-580], belvarafenib) demonstrated poor inhibition of ARAF and potent inhibition of WT and mutant versions of CRAF and BRAF 93 .Paradox breakers (e.g., PLX8394, PLX7904) are more specific BRAF inhibitors that alter the dimer interface and subsequently prevent BRAF-homodimer and BRAF:CRAF heterodimer formations 92,94 .Data from preclinical and clinical studies suggest that BRAF non-V600 mutations could be targeted with these new inhibitors, and several are being investigated in ongoing trials to better define their efficacy and safety (Table 3) 34,95,96 .).The treatment breakdown for specifically the BRAF-mutant cohort was not reported.
Better understanding of the epidemiology, risk, impact, and treatment of BMs in patients with BRAF-mutant metastatic NSCLC remains an unmet need.
More robust efficacy analysis of patients with BMs has been conducted for BRAF-mutant metastatic melanoma treated with BRAF/MEK inhibitors.In a phase 3 study of patients with BRAF V600 -mutant unresectable or metastatic melanoma with BMs (n = 275), systemic outcomes with dabrafenib plus trametinib were ORR of 41.5% and median PFS of 5.68 months (95% CI, 5.29-6.87) 99.Treatment-naive patients had a significantly longer median PFS than previously treated patients (7.23 versus 4.96 months; P = 0.0061).An open-label, multi-cohort, phase 2 study evaluated the efficacy of dabrafenib plus trametinib in patients with BRAF V600 -mutant melanoma with BMs (n = 125), including the largest cohort (n = 76) with asymptomatic BRAF V600E -mutant BMs and no prior brain-directed therapy 100 .For the largest cohort, intracranial response rate (IC ORR) was 58% (95% CI, 46-69%), median PFS was 5.6 months (95% CI, 5.3-7.4), and median OS was 10.8 months (95% CI, 8.7-19.6).The safety profile was consistent with those in previous dabrafenib plus trametinib studies.In a multicenter, retrospective case series investigation, encorafenib plus binimetinib combination treatment of 24 patients with BRAF-mutant metastatic melanoma BMs resulted in IC ORR of 33% and ORR of 21% 101 .The safety profile was consistent with what was reported in patients with melanoma without BMs.

Conclusion
BRAF V600E is an actionable mutation for metastatic NSCLC with safe and effective treatment options 8,16,18 .BRAF and MEK inhibitor combination therapies have demonstrated rapid and durable responses in the majority of patients 15,16,18 .The safety profiles of these combinations are well described, and appropriate therapy management principles should be employed for responding patients.Disease progression is inevitable, as only ~50% of patients with metastatic NSCLC receive a second-line treatment 102,103 .Therefore, it is prudent to use the most efficacious agents in the first-line setting.Further studies are necessary to determine optimal sequencing methods, understand resistance mechanisms, determine efficacy of treatments for BMs, and develop targeted therapies for non-V600E mutations.

Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Fig. 1 |
Fig. 1 | BRAF mutation classes and mechanism of actions for BRAF/MEK inhibitors.Class I and II mutations are RAS-independent, constitutively active monomers (class I) or dimers (class II).Class III mutations are RAS-dependent

Table 1 |
Selected trials of BRAF and MEK inhibitors 18,16verse event, CR complete response, DOR duration of response, NA not available, NE not estimable or not reached, ORR objective response rate, OS overall survival, PFS progression-free survival, PR partial response, SAE serious AE, SCC squamous cell carcinoma, TRAE treatment-related AE. a Dose interruption, dose reduction, and permanent treatment discontinuation due to AEs. b Efficacy and safety were analyzed for 19 patients.However, one patient dropped out prior to assessment of response and was included in the denominator for efficacy but as having no response.cORRwasanalyzedfor96 patients.dUpdated5-yearanalysisreportedORR of 63.9%, median PFS of 10.8 months, and median OS of 17.3 months 17 .eUpdated5-yearanalysisreportedORR of 68.4%, median PFS of 10.2 months, and median OS of 18.2 months 17 .fMediantimetoresponsewas reported as 6 weeks in the paper.Weeks to months was calculated with 1:0.23 conversion.gOnepreviouslytreatedpatient had both V600D and V600E mutations in their tumor.hMedianduration of follow-up for this study was for PFS.https://doi.org/10.1038/s41698-024-00552-7ReviewarticlenpjPrecisionOncology| (2024)8:90a a Adverse events (AEs) shown for dabrafenib plus trametinib occurred in at least 10% of patients in combined data from interim analysis of treatment-naive and previously treated patients15,16.AEs led to dose reduction, dose interruption, and permanent treatment discontinuation in 37%, 67%, and 16% of patients.aChestpain includes musculoskeletal chest pain.bTreatment-relatedAEs (TRAEs) shown for encorafenib plus binimetinib occurred in at least 10% of patients18.TRAEs led to dose reduction, dose interruption, and permanent treatment discontinuation in 24%, 44%, and 15% of patients.Comparisons of safety profiles should be done cautiously since data are from different trials and reported as all-causality AEs for one treatment combination and as TRAEs for the other combination.ALP alkaline phosphatase, ALT alanine aminotransferase, AST aspartate aminotransferase, CNS central nervous system, CPK creatine phosphokinase.npj Precision Oncology | (2024)8:90

Table 2 |
Efficacy data for chemotherapy and immunotherapy trials that included patients with BRAF-mutant NSCLC NA not available, NE not estimable, ORR objective response rate, OS overall survival, PD-1 programmed cell death protein 1, PD-L1 programmed death ligand 1, PFS progression-free survival.All patients received platinum-based doublet chemotherapy (most commonly, carboplatin-pemetrexed with or without bevacizumab, but several patients received cisplatin instead of carboplatin and docetaxel or etoposide instead of pemetrexed).
a All patients received platinum-based doublet combination chemotherapy.b Median OS was calculated for n = 12 in each group.c d Includes nivolumab (n = 11), pembrolizumab (n = 10), and atezolizumab (n = 1).e ORR for non-V600E was calculated out of 9 patients.