Everolimus for angiomyolipoma associated with tuberous
`sclerosis complex or sporadic lymphangioleiomyomatosis
`(EXIST-2): a multicentre, randomised, double-blind,
`placebo-controlled trial
`
`John J Bissler, J Christopher Kingswood, Elżbieta Radzikowska, Bernard A Zonnenberg, Michael Frost, Elena Belousova, Matthias Sauter,
`Norio Nonomura, Susanne Brakemeier, Petrus J de Vries, Vicky H Whittemore, David Chen, Tarek Sahmoud, Gaurav Shah, Jeremie Lincy,
`David Lebwohl, Klemens Budde
`
`Summary
`Background Angiomyolipomas are slow-growing tumours associated with constitutive activation of mammalian
`target of rapamycin (mTOR), and are common in patients with tuberous sclerosis complex and sporadic
`lymphangioleiomyomatosis. The insidious growth of these tumours predisposes patients to serious complications
`including retroperitoneal haemorrhage and impaired renal function. Everolimus, a rapamycin derivative, inhibits the
`mTOR pathway by acting on the mTOR complex 1. We compared the angiomyolipoma response rate on everolimus
`with placebo in patients with tuberous sclerosis or sporadic lymphanioleiomyomatosis-associated angiomyolipomata.
`
`Methods In this double-blind, placebo-controlled, phase 3 trial, patients aged 18 years or older with at least one
`angiomyolipoma 3 cm or larger in its longest diameter (defi ned by radiological assessment) and a defi nite diagnosis
`of tuberous sclerosis or sporadic lymphangioleiomyomatosis were randomly assigned, in a 2:1 fashion with the use of
`an interactive web response system, to receive oral everolimus 10 mg per day or placebo. The primary effi cacy
`endpoint was the proportion of patients with confi rmed angiomyolipoma response of at least a 50% reduction in
`total volume of target angiomyolipomas relative to baseline. This study is registered with ClinicalTrials.gov
`number NCT00790400.
`
`Results 118 patients (median age 31·0 years; IQR 18·0–61·0) from 24 centres in 11 countries were randomly assigned
`to receive everolimus (n=79) or placebo (n=39). At the data cutoff , double-blind treatment was ongoing for 98 patients;
`two main reasons for discontination were disease progression (nine placebo patients) followed by adverse events (two
`everolimus patients; four placebo patients). The angiomyolipoma response rate was 42% (33 of 79 [95% CI 31–53%])
`for everolimus and 0% (0 of 39 [0–9%]) for placebo (response rate diff erence 42% [24–58%]; one-sided Cochran-
`Mantel-Haenszel test p<0·0001). The most common adverse events in the everolimus and placebo groups were
`stomatitis (48% [38 of 79], 8% [3 of 39], respectively), nasopharyngitis (24% [19 of 79] and 31% [12 of 39]), and acne-like
`skin lesions (22% [17 of 79] and 5% [2 of 39]).
`
`Interpretation Everolimus reduced angiomyolipoma volume with an acceptable safety profi le, suggesting it could be
`a potential treatment for angiomyolipomas associated with tuberous sclerosis.
`
`Funding Novartis Pharmaceuticals.
`
`Introduction
`Tuberous sclerosis complex is caused by decreased or
`absent expression of the genes TSC1 (hamartin) or TSC2
`(tuberin), and
`is characterised by the growth of
`hamartomas in the kidney, brain, heart, liver, and skin.1–3
`These hamartomas predispose patients to organ system
`dysfunction,
`including autism spectrum disorder,
`intellectual disability, and epilepsy.1–3 Angio myolipomas
`are associated with loss of heterozygosity at the TSC1 or
`TSC2 gene locus, occur in roughly 80% of patients with
`tuberous sclerosis, and cause the largest proportion of
`adult deaths from the disease.3–5 As angiomyolipomas
`enlarge, aneurysm and haemorrhage risk increases.6
`Enlarging renal angiomyolipomas can cause chronic
`kidney dis ease needing dialysis or eventual renal
`
`transplantation.7 Patients with angiomyolipomas often
`develop new lesions and recurrence of treated lesions.8
`Lymphangio leiomyomatosis occurs in around 30% of
`women with tuberous sclerosis, and also rarely occurs in
`patients without tuberous sclerosis complex (sporadic
`lymphangioleiomyomatosis).3,8–11
`Hamartin and tuberin form a complex with Tre2-Bub2-
`Cdc16 (TBC) 1 domain family, member 7 (TBC1D7),
`which is required for the proper regulation of Rheb and
`mammalian target of rapamycin complex 1 (mTORC1)
`by cellular growth conditions.12 The hamartin-tuberin
`complex inhibits mTORC1, and loss of the complex
`leads to constitutive mTORC1 activation, aberrant sig nal-
`ling, and tumour growth.13 The ability of sirolimus,
`an mTOR inhibitor, to improve manifestations of
`
`Articles
`
`Lancet 2013; 381: 817–24
`Published Online
`January 11, 2013
`http://dx.doi.org/10.1016/
`S0140-6736(12)61767-X
`See Comment page 783
`Cincinnati Children’s Hospital
`Medical Center, Cincinnati,
`OH, USA (Prof J J Bissler MD);
`Royal Sussex County Hospital,
`Brighton, UK
`(J C Kingswood FRCP); National
`Tuberculosis and Lung
`Diseases Research Institute,
`Warsaw, Poland
`(E Radzikowska MD);
`Universitair Medisch Centrum,
`Utrecht, Netherlands
`(B A Zonnenberg MD);
`Minnesota Epilepsy Group,
`St Paul, MN, USA (M Frost MD);
`Moscow Research Institute of
`Pediatrics & Pediatric Surgery,
`Moscow, Russia
`(Prof E Belousova MD);
`Medizinische Klinik und
`Poliklinik IV, Klinikum der
`Universität München, Munich,
`Germany (M Sauter MD); Osaka
`University Hospital, Osaka,
`Japan (Prof N Nonomura MD);
`Charité Universitätsmedizin,
`Berlin, Germany
`(S Brakemeier MD,
`Prof K Budde MD); Division of
`Child & Adolescent Psychiatry,
`University of Cape Town,
`South Africa
`(Prof P J de Vries PhD); National
`Institute of Neurological
`Disorders and Stroke, National
`Institutes of Health,
`Bethesda, MD, USA
`(V H Whittemore PhD);
`Novartis Pharmaceuticals
`Corporation, Florham Park,
`NJ, USA (D Chen PhD,
`T Sahmoud MD, G Shah MD,
`D Lebwohl MD); and Novartis
`Pharmaceuticals Corporation,
`Basel, Switzerland (J Lincy MS)
`
`www.thelancet.com Vol 381 March 9, 2013
`
`817
`
`NOVARTIS EXHIBIT 2060
`Par v Novartis, IPR 2016-00084
`Page 1 of 8
`
`

`
`Articles
`
`Correspondence to:
`Prof John J Bissler, Cincinnati
`Children’s Hospital Medical
`Center, 240 Albert Sabin Way,
`Cincinnati 45229-3039, OH, USA
`john.bissler@cchmc.org
`
`See Online for appendix
`
`tuberous sclerosis, including angiomyolipoma, subepen-
`dymal giant-cell astrocytoma, and
`lymphangio leio-
`myomatosis, has been shown in a small number of
`studies that were reported after the initiation of our
`trial.14–16 Furthermore, angio myolipomas are thought to
`arise from one progenitor cell called the perivascular
`epithelioid cell (PEC).8 The PEC gives rise to a family of
`neoplasms termed perivascular epithelioid cell neo-
`plasms or PEComas.5,8 Since angio myolipomas are an
`example of a PEComa, they have the potential to produce
`vascular collagen (ie, collagen type IV).
`Everolimus is a rapamycin derivative that inhibits the
`mTOR pathway by acting on mTORC1. A phase 1–2
`clinical trial with everolimus was done in a small
`population of patients (n=38, median age 32 years,
`28 were men) with tuberous sclerosis, lymphangio-
`leiomyomatosis, or both (ClinicalTrials.gov identifi er
`NCT00457964) at doses of 5 mg or 10 mg once daily, or
`30 mg, 50 mg, or 70 mg once weekly. A mean reduction
`in the sum of the volumes of target angiomyolipoma
`lesions (defi ned as lesions with a maximum diameter of
`at least 1 cm) at 12 months of 47% was reported
`(p<0·0001). The response was similar between daily and
`weekly dosing arms. Adverse events were primarily
`characterised by mouth ulcers and infections, and were
`manageable and consistent with the known safety profi le
`of everolimus in patients with tuberous sclerosis.
`Here we report the fi rst prospective, international,
`randomised, double-blind, placebo-controlled, phase 3
`study to assess everolimus effi cacy and safety in patients
`with angiomyolipoma associated with tuberous sclerosis
`or sporadic lymphangioleiomyomatosis. Additionally,
`the eff ect of everolimus on mediators of tumour
`vascularisation, vascular endothelial growth factor D
`(VEGF-D), and collagen type IV, are presented.
`
`Methods
`Patients
`Eligible patients aged 18 years or older had at least one
`angiomyolipoma 3 cm or larger in its longest diameter,
`and a defi nite diagnosis of tuberous sclerosis per con-
`sensus criteria17,18 or sporadic lymphangioleio myo ma-
`tosis (biopsy-proven or chest CT scan).19 Patients were
`excluded if their angiomyolipoma required surgery at
`randomisation, or if they had angiomyolipoma-related
`bleeding or em bolisation during the 6 months before
`randomisation. Patients with lymphangio leiomyomatosis
`were excluded if their carbon monoxide diff usion
`capacity (DLco) was 35% or less, oxygen saturation was
`below normal at rest, or oxygen saturation was 88% or
`less on a 6-min walking test with up to 6 L/min of oxygen.
`All patients (or legal representatives if patients had
`developmental delays) provided written informed consent
`according to local guidelines before enrolment. Inde pen-
`dent ethics committees, local ethics review boards, or both,
`approved the protocol, which was executed according to
`International Conference on Harmonisation Good Clinical
`
`Practice guidelines. An independent data monitor ing
`committee did safety reviews every 6 months, and their
`responsibilities included minimising the exposure of
`patients to an unsafe therapy or dose, making recom-
`mendations for changes in trial processes where ap pro-
`priate, advising on the need for dose adjustments because
`of safety issues, and endorsing continuation of the trial. A
`study steering committee supervised study conduct.
`
`Study design and treatment
`Patients were randomly assigned in a 2:1 fashion to
`receive either everolimus or placebo, stratifi ed by enzyme-
`inducing antiepileptic drug use at random isation and by
`the presence of sporadic lymphangio leiomyomatosis.
`Everolimus 10 mg per day was administered orally, with
`dose modifi cations allowed on the basis of safety fi ndings.
`Concomitant use of strong inhibitors or inducers of
`cytochrome P450 3A4 or p-glycoprotein (PgP) was to be
`avoided during the study; use of antiproliferative agents
`other than study drug was prohibited.
`Patients received blinded study treatment until angio-
`myolipoma progression, occurrence of unaccept able
`toxicity, or patient withdrawal for any other reason.
`Angiomyolipoma progression was defi ned as one or
`more of: increase from the nadir of 25% or more in
`angiomyolipoma volume (sum of volumes of all target
`angiomyolipomas identifi ed at baseline) to greater than
`baseline; appearance of a new angiomyolipoma at least
`1 cm in longest diameter; increase from nadir of 20% or
`more volume of either kidney to greater than baseline; or
`angiomyolipoma-related bleeding grade 2 or more as
`defi ned by the Common Terminology Criteria for Adverse
`Events, version 3.0.20 Patients with angio myolipoma
`progression were unblinded, and placebo patients were
`off ered open-label everolimus. The core phase of the trial,
`which lasted until the last randomised patient had been
`treated for 6 months, was analysed, and only data until
`the database lock on June 30, 2011 (before the start of the
`open-label phase), were considered in this analysis.
`Kidney CT or MRI (same modality used throughout
`the study for each patient) was done at baseline, 12, 24,
`and 48 weeks, and annually after start of study treatment
`and assessed with a blinded central radiology review.
`Skin lesions resulting from tuberous sclerosis complex
`include hypomelanotic macules, the shagreen patch,
`periungual or subungual fi bromas, and facial angio-
`fi bromas, forehead plaques, or both, and were assessed
`every 12 weeks using the seven-point grading scale
`Physician’s Global Assessment of Clinical Condition
`(appendix).21,22 Adverse events were monitored through-
`out the study and graded according to the Common
`Terminology Criteria for Adverse Events v3.0 via patient-
`reported or caregiver-reported responses as well as
`investigator assessment.20
`The primary effi cacy endpoint was the proportion of
`patients with a confi rmed angiomyolipoma response,
`defi ned as a reduction in angiomyolipoma volume (sum
`
`818
`
`www.thelancet.com Vol 381 March 9, 2013
`
`NOVARTIS EXHIBIT 2060
`Par v Novartis, IPR 2016-00084
`Page 2 of 8
`
`

`
`Articles
`
`of volumes of all target angiomyolipomas identifi ed
`at baseline) of 50% or more relative to baseline and
`absence of angiomyolipoma progression. Initial response
`required confi rmation by another scan. Key secondary
`endpoints were time to angiomyolipoma progression
`and skin lesion response rate. Additional endpoints
`included time to angiomyolipoma response, duration of
`angiomyolipoma response, duration of skin
`lesion
`response, pharmacokinetics of everolimus (including
`exposure and exposure-effi cacy relation), plasma angio-
`genic molecules (including change from baseline and
`correlation with the angiomyolipoma volume), change
`from baseline in pulmonary function in lymphangio-
`leiomyomatosis and sporadic lymphangioleiomyoma-
`tosis patients, and safety.
`Other prespecifi ed exploratory endpoints in the trial
`(which will be presented in another paper) included
`change in other lesions associated with tuberous sclerosis
`complex (namely subependymal giant-cell astrocytomas,
`tubers, and subependymal nodules), change from base-
`line in neuropsychological assessments and cognitive
`function, correlation between volume and
`longest
`diameter of angiomyolipomata, change from baseline in
`seizure severity, mutational analysis of TSC1 and TSC2
`genes correlated with angiomyolipoma response rate and
`time to angiomyolipoma progression, relation between
`everolimus concentration and safety, and incidence of
`angiomyolipoma-related surgery.
`At each visit, everolimus trough levels were measured
`in a central laboratory by high-performance liquid
`chroma tography tandem mass spectrometry (WuXi App
`Tec, Shanghai, China). Plasma samples were collected
`predose at baseline and week 24 from all trial partici pants
`to measure VEGF-D
`(Quantikine, R&D Systems,
`Minneapolis, MN, USA) and collagen IV (Kamiya
`Biomedical, Seattle, WA, USA) levels by ELISA to assess
`their prognostic value for disease outcome or predictive
`value of treatment response. All samples were analysed in
`duplicate wells and mean concentrations were reported.
`
`118 patients randomised
`
`79 allocated to everolimus
`
`39 allocated to placebo
`
`72 patients ongoing in
` double-blind period
` 7 discontinued intervention
`
` 2 adverse events
`
` 1 abnormal laboratory value
`
` 1 patient withdrew consent
`
` 1 administrative problem
`
` 1 death
`
` 1 protocol deviation
`
`26 patients ongoing in
` double-blind period
` 13 discontinued intervention
`
` 9 disease progression
`
` 4 adverse events
`
`79 analysed for efficacy
`79 analysed for safety
`
`39 analysed for efficacy
`39 analysed for safety
`
`Figure 1: CONSORT fl ow diagram of patient disposition in the double-blind
`period up to the cutoff date (June 30, 2011)
`
`decided to group the sporadic lymph angioleiomyomatosis
`stratum with patients with tuberous sclerosis not using
`an enzyme-inducing antiepileptic drug; the modifi ed
`stratifi cation became use versus non-use of an enzyme-
`inducing antiepileptic drug. Multiplicity was controlled via
`a predefi ned fi xed sequence testing pro cedure with a
`hierarchy of angiomyolipoma response rate, time to
`angiomyolipoma progression, and skin lesion response
`rate. The exposure-effi cacy analysis was fi tting a linear
`mixed model to log-transformed tumour size with log-
`baseline tumour size and log Cmin as covariates. The
`association of collagen IV and VEGF-D with angio-
`myolipoma volume was investigated by Spearman rank
`correlation (post-hoc analysis for the statistical test) to
`establish the relation between bio marker concentration
`and percentage change from baseline to week 24 in
`angiomyolipoma volume while adjusting for the baseline
`levels of the two variables. Statistical analyses were done
`with SAS software (SAS Institute).
`
`Statistical analysis
`The planned sample size (N=99) provided 93% power to
`detect a 20% diff erence in angiomyolipoma response
`rates between treatments. Effi cacy analyses were done on
`all randomised patients, and safety analyses were done
`on all patients who received at least one dose of study
`drug and had at least one post-baseline assessment.
`Patients not able to be assessed (either by drop out or
`other reasons) were considered as non-responders.
`Treatment groups were compared by use of an exact
`stratifi ed Cochran-Mantel-Haenszel test for angio myo-
`lipoma and skin lesion response rates, and a one-sided
`stratifi ed log-rank test for time to angiomyolipoma pro-
`gression (all at the one-sided 2·5% signifi cance level).
`Stratifi cation was modifi ed for statistical testing because
`only fi ve patients had sporadic lymphangio leiomyoma-
`tosis. Therefore, before database lock and unblinding, we
`
`Role of the funding source
`Academic investigators and the sponsor designed the
`study, and the sponsor analysed the data (monitored and
`stored by PAREXEL). All authors had full access to the
`data and attest to the accuracy and completeness of the
`reported data and that the study conformed to the protocol
`and statistical analysis plan. The corresponding author
`had full access to all the data in the study and had fi nal
`responsibility for the decision to submit for publication.
`
`Results
`Between May 8, 2009, and Dec 30, 2010, 118 patients
`across 24 centres in 11 countries were randomly assigned
`to receive everolimus (n=79) or placebo (n=39). At data
`cutoff (June 30, 2011), 83% (98 of 118) of patients were
`receiving double-blind study treatment, whereas 17%
`(20 of 118) had discontinued, mainly because of disease
`
`www.thelancet.com Vol 381 March 9, 2013
`
`819
`
`NOVARTIS EXHIBIT 2060
`Par v Novartis, IPR 2016-00084
`Page 3 of 8
`
`

`
`Articles
`
`progression (placebo group only; fi gure 1). Median dose
`intensity was 10 mg per day for both treatment groups;
`mean dose intensity was 8·6 mg per day in the everolimus
`
`Everolimus
`(n=79)
`
`Placebo
`(n=39)
`
`32 (18–61)
`
`35 (44)
`44 (56)
`
`27 (34)
`52 (66)
`
`71 (90)
`7 (9)
`1 (1)
`77 (97)
`2 (3)
`
`22 (28)
`77 (97)
`43 (54)
`
`Age in years, median (range)
`Age
`<30 years
`≥30 years
`Sex
`Men
`Women
`Race
`White
`Asian
`Other
`Diagnosis of tuberous sclerosis complex*
`Diagnosis of sporadic
`lymphangioleiomyomatosis
`Diagnosis of lymphangioleiomyomatosis
`≥1 skin lesion†
`Presence of subependymal giant-cell
`astrocytoma‡
`Previous angiomyolipoma therapy
`31 (39)
`Surgery/invasive procedure
`19 (24)
`Renal embolisation
`14 (18)
`Nephrectomy
`0
`Medication
`Longest diameter of largest angiomyolipoma lesion
`≥8 cm
`22 (28)
`≥4 cm and <8 cm
`45 (57)
`≥3 cm and <4 cm
`6 (8)
`<3 cm
`5 (6)
`Unknown§
`0
`Not applicable¶
`1 (1)
`Bilateral angiomyolipoma
`65 (83)
`Number of target angiomyolipoma lesions
`1 (1)
`0||
`32 (41)
`1–5
`46 (58)
`6–10
`Sum of volumes of target angiomyolipoma lesions
`Number of patients with one or more
`78
`target angiomyolipoma, n
`Median (range), cm³
`
`29 (18–58)
`
`20 (51)
`19 (49)
`
`13 (33)
`26 (67)
`
`34 (87)
`4 (10)
`1 (3)
`36 (92)
`3 (8)
`
`7 (18)
`37 (95)
`14 (36)
`
`15 (38)
`9 (23)
`8 (21)
`0
`
`12 (31)
`19 (49)
`4 (10)
`2 (5)
`1 (3)
`1 (3)
`27 (71)
`
`1 (3)
`15 (38)
`23 (59)
`
`37
`
`group and 9·6 mg per day in the placebo group. Median
`exposure duration was 38 weeks for everolimus and
`34 weeks for placebo. Coadministration of strong and
`moderate cytochrome 3A inhibitors, PgP inhibitors, CYP
`3A inducers, and PgP inducers was reported in 47 (59%)
`everolimus patients and 23 (59%) placebo patients.
`Baseline demographic and disease characteristics
`were generally balanced between treatment groups;
`more everolimus patients had subependymal giant-cell
`astrocytomas (table 1). Overall, median age was 31 years
`(range 18·0–61·0), 78% (92 of 118) of patients had
`angiomyolipomas in both kidneys, and 29% (34 of 118) of
`patients had an angiomyolipoma at least 8 cm in longest
`diameter. Almost 40% (46 of 118) of patients had a
`previous intervention, including 19% (22 of 118) with
`prior nephrectomy.
`Angiomyolipoma response rate was 42% (33 of 79
`[95% CI 31–53%]) for everolimus compared with 0%
`(0 of 39 [0–9%]) for placebo (diff erence 42% [24–58%];
`p<0·0001). Median time to angiomyolipoma response for
`everolimus was 2·9 months. The treatment eff ect was
`consistent across all subgroups evaluated, including
`modifi ed strata, sex, age, and race (fi gure 2). At week 24,
`55% (39 of 71) of everolimus patients had at least a 50%
`reduction from baseline in sum of volumes of target
`angiomyolipoma lesions compared with 0% (0 of 33) of
`placebo patients, and 80% (57 of 71) of everolimus patients
`had at least a 30% reduction compared with 3% (1 of 33) of
`placebo patients. Most everolimus patients had large
`decreases in best percentage change from baseline in sum
`of volumes of target angio myolipoma lesions (fi gure 2,
`appendix). All angio myolipoma responses were ongoing
`for between 10 and 85 weeks at the time of the data cutoff .
`Everolimus was superior to placebo in time to
`angiomyolipoma progression (hazard ratio 0·08 [95% CI
`0·02–0·37], p<0·0001; fi gure 3). Angiomyolipoma pro-
`gressions were noted in three everolimus patients (4%)
`and eight placebo patients (21%). No patient achieving an
`angiomyolipoma response had progressed at the data
`cutoff date. Estimated progression-free rates (95% CI)
`for everolimus and placebo, respectively, were 98%
`(89–100%) and 83% (65–93%) at 6 months, and 92%
`(65–98%) and 25% (1–64%) at 12 months. Median time to
`angiomyolipoma progression was 11·4 months for
`placebo and was not reached for everolimus.
`Skin lesions associated with tuberous sclerosis were
`present at baseline in 114 patients. Everolimus had a
`signifi cantly higher skin lesion response rate than
`placebo (26% [20 of 77; 95% CI 17–37%] vs 0% [0 of 37;
`95% CI 0–10%]; p=0·0002). At the data cutoff , skin lesion
`responses were ongoing in the 20 everolimus patients
`who had a skin lesion response (range 10–84 weeks).
`Adverse events were consistent with the known
`everolimus safety profi le. Stomatitis, nasopharyngitis,
`acne-like skin lesions, headache, cough, and hyper-
`cholesterolaemia were the most common adverse events
`with everolimus therapy (each reported in 20% of
`
`85 (9–1612) 120 (3–4520)
`
`Data are n (%) unless otherwise stated. *All patients diagnosed with tuberous
`sclerosis complex had two or more major features.17,18 †Based on patients having
`skin lesion photos at baseline or assessment post-baseline, not based on the
`modifi ed Gomez criteria. ‡Based on the major feature of subependymal
`giant-cell astrocytoma in the modifi ed Gomez criteria17,18 being ticked yes.
`§Longest diameter of the largest angiomyolipoma lesion is unknown when at
`least one target lesion larger than 1 cm is confi rmed but no precise diameter
`could be measured. ¶Lesions marked as not applicable are those where there is
`not at least one target lesion. ||Lesions identifi ed as not meeting target status
`were determined by central radiology whereas eligibility criteria were based on
`the local radiologist.
`
`Table 1: Baseline patient demographic and disease characteristics
`
`820
`
`www.thelancet.com Vol 381 March 9, 2013
`
`NOVARTIS EXHIBIT 2060
`Par v Novartis, IPR 2016-00084
`Page 4 of 8
`
`

`
`Articles
`
`Difference in
`response rate
`(95% CI)
`
`Everolimus
`response
`n/N (%)
`
`Placebo
`response
`n/N (%)
`
`42 (24 to 58)
`
`33/79 (42)
`
`0/39 (0)
`
`46 (–2 to 82)
`41 (20 to 59)
`
`6/13 (46)
`27/66 (41)
`
`0/7 (0)
`0/32 (0)
`
`63 (34 to 86)
`31 (6 to 53)
`
`17/27 (63)
`16/52 (31)
`
`46 (19 to 68)
`39 (12 to 62)
`
`16/35 (46)
`17/44 (39)
`
`0/13 (0)
`0/26 (0)
`
`0/20 (0)
`0/19 (0)
`
`42 (22 to 60)
`38 (–19 to 79)
`
`30/71 (42)
`3/8 (38)
`
`0/34 (0)
` 0/5 (0)
`
`A
`
`All patients (N=118)
`Modified strata
`EIAED use (N=20)
`EIAED non-use (N=98)
`Sex
`Men (N=40)
`Women (N= 78)
`Age
`<30 years (N=55)
`≥30 years (N=63)
`Race
`White (N=105)
`Non-white (N=13)
`
`0
`–20
`Favours placebo
`
`20
`
`40
`
`80
`60
`Favours everolimus
`
`Everolimus (N=66)
`
`Placebo (N=32)
`
`*
`
`Everolimus
`number (%)
`
`Placebo
`number (%)
`
`B
`100
`75
`50
`25
`0
`–25
`–50
`–75
`–100
`
`angiomyolipoma lesions)
`(sum of volumes of target
`Best % change from baseline
`
`Decrease in best percentage change from baseline
`Increase in best percentage change from baseline
`
`63 (95)
`3 (5)
`
`19 (59)
`13 (41)
`
`Figure 2: Magnitude of treatment eff ect
`(A) Angiomyolipoma response rates by subgroup. A forest plot shows the eff ect of study treatment on the
`angiomyolipoma response rate across diff erent subgroups (modifi ed strata, sex, age, and race). The area of each
`diamond is proportional to the number of patients in a particular subgroup. The diff erence in response rates is
`everolimus minus placebo, and 95% CI were obtained from the exact unconditional confi dence limits. Patients
`with an overall angiomyolipoma response as not evaluable were considered as non-responders. (B) Best
`percentage change from baseline in the sum of volumes of target angiomyolipoma lesions, per central radiology
`review. EIAED=enzyme-inducing antiepileptic drug. Patients were excluded if the overall angiomyolipoma
`response was not evaluable. Each bar represents one patient. *Subsequent scans for this patient did not
`demonstrate tumour shrinkage but instead revealed no change from baseline or slight tumour growth.
`
`Censoring times
`Everolimus
`Placebo
`
`Hazard ratio 0·08 (95% CI 0·02–0·37); p<0·0001
`
`12
`Months
`
`14
`
`16
`
`18
`
`20
`
`22
`
`24
`
`00
`
`31
`
`41
`
`41
`
`51
`
`51
`
`71
`
`0
`
`79
`39
`
`2
`
`66
`33
`
`4
`
`63
`29
`
`6
`
`25
`8
`
`8
`
`25
`8
`
`10
`
`21
`8
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Probability of angiomyolipoma
`
`progression free (%)
`
`Number at risk
`Everolimus
`Placebo
`
`Figure 3: Kaplan-Meier plot showing time to angiomyolipoma progression, as assessed by central review
`The median time to angiomyolipoma progression was not reached for patients in the everolimus treatment arm
`and was 11·4 months in the placebo treatment group, as of data cutoff . The hazard ratio and 95% CI were
`obtained from the Cox model, stratifi ed by the modifi ed stratifi cation factor (use vs non-use of enzyme-inducing
`antiepileptic drug). The p value was obtained from the one-sided log-rank test, stratifi ed by use vs non-use of
`enzyme-inducing antiepileptic drug.
`
`patients or more) and were primarily grade 1−2 (table 2).
`Infections (most frequently urinary tract and upper
`respiratory tract infections) occurred in 65% (51 of 79) of
`patients on everolimus and 72% (28 of 39) on placebo;
`there were no grade 4 infections. Adverse events leading
`to discontinuation occurred in 4% (three) of evero-
`limus patients and 10% (four) of placebo patients. In
`everolimus patients, these adverse events included
`grade  2 blood phosphorous decrease, one patient with
`concurrent grade 3 hypersensitivity, grade 3 angio-
`oedema, and grade 3 bronchospasm, and convulsion
`deemed not related to study drug, which resulted in
`death due to status epilepticus. The 28-year-old male
`patient had a medical history of intractable seizures, and
`the investigator did not consider the death treatment
`related. Dose reduction or interruption because of an
`adverse event occurred in 48% (38 of 79) of everolimus
`and 21% (eight of 39) of placebo patients.
`One case of grade 2 non-infectious pneumonitis was
`reported in the everolimus group, which resolved within
`14 days after dose reduction. Lung function in patients
`with lymphangioleiomyomatosis and sporadic lymph-
`angioleiomyomatosis on everolimus showed slightly less
`deterioration during the study than did patients in the
`placebo group; median percentage change from baseline
`to week 24 for DLCO was −3% in the everolimus group
`and −8% in the placebo group, and for forced expiratory
`volume in 1 second was −1% for everolimus and −4% for
`placebo. Interpretation of this exploratory endpoint was
`limited because of the short duration of treatment
`exposure and the low number of patients.
`The median values at baseline and week 24 for DLCO in
`the everolimus group were 6·56 and 6·00, respectively,
`and in the placebo group were 6·16 and 5·84, respectively.
`The median values at baseline and week 24 for forced
`expiratory volume in 1 second in the everolimus group
`were 2·90 and 2·70, respectively, and in the placebo
`group were 2·70 and 2·70, respectively.
`Amenorrhoea (ranging from 1 to 66 weeks) was
`reported in seven of 52 women receiving everolimus
`(13%), including two grade 3 cases, compared with one of
`26 women on placebo (4%). Of the everolimus patients
`who continued on study treatment, four cases resolved
`and three are ongoing. None of these patients had dose
`reductions or had to discontinue treatment.
`Renal events were less common in the everolimus
`group than in the placebo group (5% [ four of 79] vs 15%
`[six of 39]) and included proteinuria (everolimus 4%
`[three of 79] vs placebo 8% [3 of 39]), increased blood
`creatinine (everolimus 1% [one of 79] vs placebo 8%
`[three of 39]), and transient acute renal failure (everolimus
`2·5% [two of 79] vs placebo 0% [0 of 39], all grade 1 or 2.
`Mean everolimus
`trough
`levels ranged between
`7·63 (SD 4·32) ng/mL (week 2) and 9·37 (SD 8·83) ng/mL
`at week 24 exhibiting large inter-individual variability
`(56–94%), with lower exposure in patients using enzyme-
`inducing antiepileptic drugs
`(5·10
`[SD 3·02] vs
`
`www.thelancet.com Vol 381 March 9, 2013
`
`821
`
`NOVARTIS EXHIBIT 2060
`Par v Novartis, IPR 2016-00084
`Page 5 of 8
`
`

`
`Articles
`
`Everolimus (n=79)
`
`Placebo (n=39)
`
`All grades Grade 3 Grade 4*
`
`All grades Grade 3 Grade 4*
`
`Stomatitis
`Nasopharyngitis
`Acne-like skin lesions
`Headache
`Cough
`Hypercholesterolaemia
`Aphthous stomatitis
`Fatigue
`Mouth ulceration
`Nausea
`Urinary tract infection
`Vomiting
`Anaemia
`Arthralgia
`Diarrhoea
`Abdominal pain
`Blood lactate dehydrogenase increased
`Hypophosphataemia
`Eczema
`Leucopenia
`Oropharyngeal pain
`Upper respiratory tract infection
`
`38 (48)
`19 (24)
`17 (22)
`17 (22)
`16 (20)
`16 (20)
`15 (19)
`14 (18)
`13 (16)
`13 (16)
`12 (15)
`12 (15)
`10 (13)
`10 (13)
`10 (13)
`9 (11)
`9 (11)
`9 (11)
`8 (10)
`8 (10)
`8 (10)
`8 (10)
`
`1 (1)
`0
`0
`0
`0
`0
`2 (3)
`1 (1)
`2 (3)
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`3 (8)
`12 (31)
`2 (5)
`7 (18)
`5 (13)
`1 (3)
`4 (10)
`7 (18)
`2 (5)
`5 (13)
`6 (15)
`2 (5)
`1 (3)
`2 (5)
`2 (5)
`3 (8)
`2 (5)
`0
`3 (8)
`3 (8)
`4 (10)
`2 (5)
`
`0
`0
`0
`1 (3)
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1 (3)
`0
`0
`0
`0
`0
`0
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`Data are n (%). A patient with multiple occurrences of an adverse event is counted only once in that adverse event
`category. *Four grade 4 adverse events were reported in the everolimus group: two were laboratory abnormalities
`reported by the central laboratory (blood uric acid increased and neutropenia), one was a convulsion, and one was a
`hypertensive crisis. One grade 4 adverse event occurred in the placebo group (volvulus).
`
`Table 2: Adverse events of any cause experienced by 10% or more patients in the everolimus treatment
`group, by grade
`
`10·41 [SD 9·47] ng/mL). A large intersubject variability in
`response was also seen in the relation between percentage
`change from baseline in angiomyolipoma lesion volume
`and time-normalised minimum (trough) concentration
`(Cmin). An analysis fi tting a linear mixed model to log-
`transformed tumour size with log-baseline tumour size
`and log Cmin as covariates indicated a 10% tumour size
`reduction from baseline for a 2-fold Cmin increase (95% CI
`−16 to −4%). There was no correlation between response-
`exposure and subgroups of angiomyolipoma volume at
`baseline (<100 cm³, 100–200 cm³, and >200 cm³).
`Plasma VEGF-D and collagen IV levels decreased from
`baseline through week 24 only in the everolimus treat-
`ment group. Median VEGF-D concentration decreased
`62% in everolimus-treated patients from 1307·0 to 503·0
`compared with a 6% increase from 1762·5 to 1840·5 in
`placebo patients, and median collagen IV concentration
`decreased by 42% in everolimus patients from 216·7 to
`130·4 compared with no change for placebo, indicating
`that decreases in VEGF-D and collagen IV concentrations
`were due to everolimus administration.
`Plasma VEGF-D level correlated with angiomyolipoma
`lesion volume at baseline (appendix) and week 24 for the
`everolimus group (Spearman correlation coeffi cients
`
`0·55 for both, p<0·0001) and the placebo group
`(Spearman correlation coeffi cients 0·41 and 0·35,
`respect ively, p=0·0191, baseline and p=0·0461 for
`week 24). Similarly, plasma collagen IV level correlated
`with angiomyolipoma lesion volume at baseline (appen-
`dix) in the everolimus arm (Spearman correlation
`coeffi cient 0·49, p<0·0001). Moreover, reductions in
`angiomyo lipoma volume and VEGF-D levels were also
`correlated (Spearman partial correlation coeffi cient 0·43,
`p<0·0007) in the 62 everolimus patients for whom both
`VEGF-D concentrations and angiomyolipom