PD-L1 immunohistochemistry testing for access to pembrolizumab

Document technical information

Format docx
Size 713.5 kB
First found May 22, 2018

Document content analysis

Category Also themed
Language
English
Type
not defined
Concepts
no text concepts found

Persons

Organizations

Places

Transcript

Page 1 of 23
MEDICAL SERVICES ADVISORY COMMITTEE
PD-L1 immunohistochemistry testing for access to
pembrolizumab for the treatment of locally advanced
or metastatic non-small cell lung cancer
Protocol 1440
June 2016
MSAC website: www.msac.gov.au
Page 2 of 23
Purpose of application
This application is requesting a Medicare Benefits Schedule (MBS) listing for testing of Programmed
Death 1 Ligand (PD-L1) expression in advanced non-small cell lung cancer (NSCLC). This listing is
required to support an integrated co-dependent submission for use of the PD-L1 test to determine
eligibility for pembrolizumab in first line treatment of NSCLC.
A protocol (No. 1414) has already been ratified for PD-L1 testing to determine eligibility to
pembrolizumab in the second line setting. Table 1 outlines the differences between this protocol
(No. 1440) and Protocol 1414.
Table 1: Table outlining differences between Protocol 1414 and Protocol 1440
Population and medical condition eligible for the proposed medical service
Background on non small cell lung cancer Same as Application 1414
Role of Programmed Death-1 pathway as a therapeutic target
in cancer
Testing for PD-L1 expression
Prevalence and prognostic value of PD-L1 expression in
NSCLC
Proposed patient population New information for Application 1440
Evidence for the proposed patient population
KN001
KN024 New information for Application 1440
KN042 New information for Application 1440
Intervention – proposed medical service
Description of proposed medical service Same as Application 1414
Proposed MBS listing
Expected utilisation
Reference standard New information for Application 1440
Delivery of the proposed medical test
Where, by whom, frequency of testing Same as Application 1414
Co-dependent information
Co-dependent drug New information for Application 1440
Comparator
Test New information for Application 1440
Drug New information for Application 1440
Co-dependence New information for Application 1440
Clinical claim for proposed medical service
PD-L1 test outcomes New information for Application 1440
Drug Outcomes Same as application 1414
Risk to the patient New information for Application 1440
Type of economic evaluation New information for Application 1440
Fee for the proposed medical service
Proposed funding Same as application 1414
Direct costs of equipment/resources used with the service Same as application 1414
The proposed fee Same as application 1414
Clinical Management Algorithm - clinical place for the proposed intervention
Current treatment algorithm New information for Application 1440
Future treatment algorithm New information for Application 1440
Regulatory Information
New information for Application 1440
Decision analytic
New information for Application 1440
Healthcare Resource use
New information for Application 1440
Question for public funding
New information for Application 1440
MSAC website: www.msac.gov.au
Page 3 of 23
As the biomarker aspects of this protocol are largely unchanged compared to those ratified through
Protocol 1414, the sponsor proposes that a streamlined review process is undertaken by PASC.
Population and medical condition eligible for the proposed medical
services
Non-small cell lung cancer
Lung cancer is the 5th most commonly diagnosed cancer, with over 10,000 patients diagnosed each
year, and a prevalence of around 94 people per 100,000.1 In 2014, lung cancer was the most
common cause of cancer death, accounting for 18.9% of all cancer deaths (8,630 deaths).2 Non-small
cell lung cancer (NSCLC) accounts for approximately 66% of all lung cancer cases3. Progress has been
made in the clinical management of early stage NSCLC. However, the prognosis for advanced disease
has not improved substantially. With an overall 5-year survival rate of 13-16%, the treatment of
NSCLC remains a high unmet medical need4.
Role of the Programmed Death-1 pathway as a therapeutic target in cancer
In recent years, it has become apparent that cancers are recognized by human immune system and
that under certain circumstances the immune system can obliterate tumours. Recently, the PD-1
pathway has emerged as a major immune checkpoint by which tumours suppress lymphocyte
function. This pathway consists of PD-1, a protein expressed on activated immune cell types such as
T cells and B cells, and its ligands, PD-L1 and PD-L2 which are expressed on many tumours. Cancer
cells drive high expression levels of PD-L1 on their surface, allowing activation of the inhibitory PD-1
receptor on any T cells that infiltrate the tumour microenvironment, effectively switching those cells
off. Indeed, up-regulation of PD-L1 expression levels has been demonstrated in many different
cancer types (eg, melanoma [40%-100%], NSCLC [35%-95%], and multiple myeloma [93%]), and high
levels of PD-L1 expression have been linked to poor clinical outcomes (Hino et al, 2010, Wang et al,
2011, Dong et al, 2002, Konishi et al, 2004, Liu et al, 2007, Patel et al, 2015).
It has been proposed that immunotherapy targeting this pathway may be a potential cancer
treatment modality. Hence several molecules targeting this pathway are currently under clinical
development in NSCLC. One such molecule is pembrolizumab.
Pembrolizumab mechanism of action
Pembrolizumab is a potent and highly selective humanized monoclonal antibody (mAb) designed to
target the programmed death-1 receptor and thus directly block the interaction between PD-1 and
its ligands, PD-L1 and PD-L2. This blockade enhances functional activity of the target lymphocytes to
facilitate tumour regression and ultimately immune rejection. Pembrolizumab only potentiates
existing immune responses in the presence of antigen and does not non-specifically activate T cells.
1
Lung Foundation: Lung Disease in Austraila [accessed 7th May 2015]
in Australia: an overview 2014, AIHW, Table 7.2, Pg 49 of document, AIHW Cancer in Australia an overview [accessed 7th May
2015]
3 Lung cancer in Australia: an overview, AIHW, AIHW Lung cancer in Australia an overview Table 3.8, Pg 24 [accessed 7th May 2015]
4 Lung cancer in Australia: an overview, AIHW, AIHW Lung cancer in Australia an overview Figure 5.2, Pg 65 [accessed 7th May 2015]
2Cancer
MSAC website: www.msac.gov.au
Page 4 of 23
Evidence shows that PD-L1 expression levels correlate with increased response to pembrolizumab.
For instance, in NSCLC phase 1 clinical trial data showed a correlation between PD-L1 expression and
response to pembrolizumab, supporting the role of PD-L1 expression testing as a predictive
biomarker (Garon et al, 2015).
Testing for PD-L1 expression
PD-L1 expression in NSCLC tumour biopsies can be assessed using immunohistochemical (IHC)
testing with antibodies that bind specifically to the PD-L1 protein.
Three PD-L1 assays have been used during the pembrolizumab NSCLC clinical development program:



A Prototype Research Assay (PRA).
A Clinical Trial Assay (CTA).
The PD-L1 22C3 pharmDx Market Ready Assay (MRA).
All of these assays tests use the same antibody (mouse anti-human monoclonal antibody clone
22C3). However the associated kit reagents are slightly different
The Clinical Trial Assay and the Market Ready Assay were both developed by Dako, the company
with whom MSD are partnering for development of the companion diagnostic.
Prevalence and prognostic value of PD-L1 expression in NSCLC
As PD-L1 is a relatively new biomarker, there is limited data on the prevalence and prognostic role of
PD-L1 expression in NSCLC. Whilst earlier studies have given rise to mixed results, two recent metaanalyses have shown that positive PD-L1 expression is correlated with poor prognosis in NSCLC
patients (Wang et al, 2015; Zhou et al, 2015).
Zhou et al, 2015 also found that PD-L1 expression is not related to gender, histology type, smoking
status, tumor stage, or the absence or presence of lymph node metastasis. The relationship
between PD-L1 expression and other biomarkers such as KRAS, EGRF and ALK has yet to be
determined. Whilst some studies have associated PD-L1 positive status with the presence of KRAS
and EGFR mutation, this finding has been inconsistent (Ji et al, 2015).
In terms of PD-L1 prevalence, early screening data from multinational clinical trials (including
Australia) that MSD is undertaking has found that approximately 61% of advanced NSCLC patients
screened are PD-L1 positive (≥1% PD-L1 expression) and that approximately 23% of patients are
strongly positive (≥50% PD-L1 expression) (Garon et al, 2015). MSD is committed to providing an
overview of the prevalence and prognostic data for PD-L1 in NSCLC as part of co-dependent
submission.
Proposed patient population
As proposed in Application 1414, the patient population which would benefit from PD-L1 testing are
metastatic (stage IV) NSCLC patients ( squamous, non squamous and not otherwise specified) who
have had no prior treatment. The outcome of this test will determine whether the patients are
eligible for subsequent treatment with pembrolizumab.
In the co-dependent technology submission MSD will present data on intra-block and intra-case
heterogeneity for PD-L1 expression in NSCLC. MSD also commits to reviewing the literature for
MSAC website: www.msac.gov.au
Page 5 of 23
additional publicly available evidence on tissue heterogeneity with respect to PD-L1 expression in
NSCLC.
Clinical evidence for the predictive role of PD_L1
Keynote 001
The role of PD-L1 testing in predicting patient response to pembrolizumab in locally
advanced/metastatic NSCLC was identified in Keynote 001 (KN001), an adaptive phase 1 trial (Garon
et al, 2015).
The objectives of KN001 were to assess the efficacy and safety of pembrolizumab in patients with
advanced NSCLC, and to define and validate an expression level of PD-L1 that is associated with the
likelihood of clinical benefit. Key characteristics of the KN001 trial are outlined in Table 1.
Table 2: Trial design for Keynote 001
Trial
Patient population
Study design
Sample Size and Endpoints
Keynote
001
 Part C: NSCLC of any
histology
 Part F: NSCLC with PDL1 protein expression
Open label phase 1
 10 mg/kg Q3W
Pembrolizumab
 10 mg/kg Q2W
Pembrolizumab
 2 mg/kg Q3W
Pembrolizumab
Part C N=38
Part F N=457
Primary endpoint
 Response rate as per RECIST
1.1
 No. of pts experiencing
adverse events
 No. of pts experiencing doselimiting toxicities
 Mix of treatment naïve
and progressive
disease following 1 or
two treatments
Note: a full explanation of the design and results can be found in Garon et al, 2015.
Early results (Part C) of KN001 showed that pembrolizumab had clinical activity in subjects with
NSCLC (Gandhi et al, 2014). Moreover, a greater clinical benefit from pembrolizumab treatment
appeared to be associated with a higher level of PD-L1 expression.
On the basis of these results, amendments were made to the KN001 trial protocol to further explore
this relationship (Part F). In particular, part F focussed on defining and validating an expression level
of PD-L1 associated with a greater likelihood of clinical benefit.
Biomarker analysis in KN001
All three PD-L1 assays (PRA, CTA and MRA), using the 22C3 antibody, were used in the KN001 trial:

The PRA was used to screen patients for eligibility to KN001 Part C and Part F. It is no longer
in use.
MSAC website: www.msac.gov.au
Page 6 of 23


The CTA was used for biomarker cut point determination and assessment of PD-L1
expression during biomarker validation.
The MRA was used for retrospective scoring of the Biomarker Validation subjects as part of
the efficacy analysis in KN001 Part F.
Biomarker analysis to determine patient eligibility to KN001
All patients enrolled in the KN001 trial were to have been deemed positive for PD-L1 expression
(≥1%) using the Prototype Research Assay. Testing was to be performed on a contemporaneous
biopsy sample if possible. This meant that either the sample needed to be collected within 60 days of
the first dose of pembrolizumab or the sample needed to be collected in the time between the last
dose of the previous systemic anticancer therapy and the first dose of pembrolizumab. Archival
tissue was analysed when contemporaneous tissue were not available.
PD-L1 expression cut point selection and scoring system
Overall, 182 patients from KN001 were assigned to a group to define a PD-L1 cut off.
Key points of this assessment are:



129 patients had measureable disease (RECIST criteria) and samples that could be evaluated
for PD-L1 expression
PD-L1 expression was assessed using the Clinical Trial Assay
Contemporaneous biopsy specimens (≤60 days old) were predominantly used, although
archival tissue was analysed when contemporaneous tissue were not available (n=25 archival
samples)
Receiver operating characteristic (ROC) analysis was employed to develop a PD-L1 expression scoring
system and to define potential PD-L1 cut points which were associated with an enhanced response
to pembrolizumab.
ROC analysis was performed on the following immunohistochemistry (IHC) scoring methods:




Proportion score (PS): defined as the percentage of cells with membranous PD-L1 staining at
any intensity
Proportion score 2+ or 3+ (P2S): defined as the percentage of cells with membranous PDL1
staining at moderate (2+) or strong (3+) intensity
Proportion score 3+ (P3S): defined as the percentage of cells with membranous PD-L1
staining at strong intensity (3+)
Modified H-score (HS): which provides a numerical value that accounts for the proportion of
cells staining for PD-L1 at each of the 3 intensities.
The results of the ROC analyses are presented in Figure 1.
MSAC website: www.msac.gov.au
Page 7 of 23
Figure 1: Receiver operating characteristic analysis based on investigator-assessed immune–related response
criteria (irRC) and membranous PD-L1 expression.
Source: Figure S.3 (p.10) of Supplement to Garon et al (2015)
The open circle on the PS curve represents the point at which Youden’s J statistic (Youden’s Index) is
maximised for the ROC curve assessing PD-L1 expression defined as the proportion of cells with
membranous PD-L1 staining at any intensity. This point corresponds to a cut point of membranous
PD-L1 expression of any intensity in 45-50% of tumour cells.
No major differences were observed in ROC area under the curve for the potential scoring methods,
regardless of the approach used (Figure 1). The positive predictive value of the Clinical Trial Assay
was not improved by incorporating PD-L1 expression on inflammatory T cells.
Hence, membranous PD-L1 expression in at least 50% of tumor cells (proportion score, ≥50%) was
selected as the PD-L1 strong vs weak cut point on the basis of the ease of use and ROC analysis
Biomarker validation of PD-L1 expression
Following biomarker cut point selection, an analysis of the anti-tumour activity of pembrolizumab
according to PD-L1 expression level was performed on a subset of patients enrolled in KN001. This
group included 313 patients (223 previously treated; 90 previously untreated), and PD-L1 status was
measureable in 230 patients.
MSAC website: www.msac.gov.au
Page 8 of 23
Key points of this assessment were:




All PD-L1 testing was performed using the Clinical Trial Assay.
Scoring was also done retrospectively using the Market Ready Assay and results were
identical to the Clinical Trial Assay results
When archival tissue was used, slides must have been sectioned within 6 months of
performing PD-L1 testing due to antigen degradation.
PD-L1 scoring was reported as based on following categories:
o Percentage of neoplastic cells with PD-L1 staining of <1% (PS <1%)
o Percentage of neoplastic cells with PD-L1 staining between 1-49% (PS 1 - 49%)
o Percentage of neoplastic cells with PD-L1 staining ≥ 50% (PS ≥50%)
The results of this analysis showed that the response rate to pembrolizumab was increased in
patients with higher levels of PD-L1 expression (Garon et al, 2015).
Keynote -010
The clinical utility of the PD-L1 test was confirmed in the Keynote 010 trial. This trial compared the
efficacy of pembrolizumab to docetaxel in PD-L1+ patients who had failed platinum-based
chemotherapy. It showed that overall survival benefit was greater in patients who expressed higher
levels of PD-L1 (≥50%) compared to those expressing lower levels (≥1-49%) (Herbst et al, 2015).
PD-L1 tumour proportion score
≥50%
1-49%
n/N
204/442
317/591
HR (95% CI)
0.53 (0.40-0.70)
0.76 (0.60-0.96)
Keynote -024
Data from KN024 will represent the pivotal evidence presented in MSD’s co-dependent submission
to support listing of pembrolizumab as a 1st line therapy in patients with NSCLC.
Keynote 024 is a prospective randomised controlled trial designed to assess the efficacy and safety of
pembrolizumab monotherapy compared to investigator’s choice of platinum-based chemotherapies
in subjects previously untreated for Stage IV non-small cell lung cancer, whose tumours strongly
express PD-L1 (≥50% positive) and are EGFR wild type and ALK transloction negative.
PD-L1 expression was determined using the Market Ready Assay .
Key characteristics of the KN024 trial are outlined in Table 3.
Table 3: Trial design for KN024
Trial
Patient population
Study design
Sample Size and Endpoints
Keynote
024

Multi-centre, multi-country,
Phase II/III
Randomized (1:1)
 200 mg Q3W
pembrolizumab
 Platinum-based
chemotherapy Investigators
choice of
cisplatin/carboplatin +
pemetrexed (non-squamous
N=300
Primary endpoint:
 PFS
 Key safety
Other endpoints
 OS
 ORR
 EORTC QLQ-C30 and EORTC
QLQ LC-13,
PD-L1 strong positive
Stage IV NSCLC (≥50%)
 No prior therapy
 EGFR wild-type and
ALK negative
MSAC website: www.msac.gov.au
Page 9 of 23
only with or without
pemetrexed maintenance);
cisplatin/carboplatin
+gemcitabine; paclitaxel +
gemcitabine (with or
without pemetrexed
maintenance)
 EQ-5D 3L
Control patients offered
pembrolizumab following
progression
Abbreviations: Q3W = every 3 weeks; OS= overall survival; PFS = progression free survival; SQ = squamous; NSQ = nonsquamous; IRC =
independent Review Committee
Keynote -042
KN042 is another study investigating use of pembrolizumab as a 1st line therapy in NSCLC patients
whose tumours are PD-L1 positive (≥1% positive), EGFR wild type and ALK translocation negative.
Evidence from this study could inform a co-dependent submission in the future.
Table 4: Trial design for KN042
Table
1Table 1
Keynote
042
Patient population
Study design
Sample Size and Endpoints
 PD-L1+ (≥1%) Stage
Multi-centre, multi-country,
Phase II/III
Randomized (1:1)
N=1240
Primary endpoint:
 200 mg Q3W pembrolizumab
Other endpoints:
 Platinum-based chemotherapy
 PFS
IV NSCLC
 No prior therapy
investigators choice of
carboplatin + paclitaxel with or
without pemetrexed
maintenance for nonsquamous
patients; carboplatin +
pemetrexed with or without
pemetrexed maintenance for
nonsquamous patients.
 OS
 Key safety
Abbreviations: Q3W = every 3 weeks; OS= overall survival; PFS = progression free survival; SQ = squamous; NSQ = nonsquamous; IRC =
independent Review Committee
Prerequisites to biomarker analysis in KN024 and KN042
Key features of the biopsy material used to assess for PD-L1 status were:
•

•
•
Biopsy of a tumour lesion must have been undertaken either at the time of or after the
diagnosis of metastatic disease has been made AND from a site not previously irradiated.
Core needle or excisional biopsies, or resected tissue were required. Any cytology specimens
including fine needle aspirate (FNA), endobronchial ultrasound (EBUS) or cell blocks were not
acceptable
Biopsy specimens could come from either a primary tumour or a metastatic lesion.
Sufficient biopsy material was required for biomarker analysis. A biopsy length of
0.3-0.5 cm was recommended. The PD-L1 test required a sample of 4-5 microns.
MSAC website: www.msac.gov.au
Page 10 of 23
Scoring of the PD-L1 test
Based on the analyses undertaken in KN001, the following scoring approach was used in KN024:
•
•
•
•
•
Scoring was undertaken using a proportional approach.
From a histology sample, at least 100 evaluable tumour cells are scored.
Positive staining is defined as complete or partial staining of the cell membrane. Any
intensity of staining is acceptable.
The proportion of cells that are positive is the outcome of the scoring.
Negative staining is defined as no staining of plasma membrane.
•
Granular staining in the cytoplasm is not considered to be
positive staining.Intervention – proposed medical service
Description of proposed medical service
The Market Ready Assay (PD-L1 22C3 pharmDx assay) will be made commercially available in
Australia. TGA registration of the Market Ready Assay, including any applicable registered
trademark, is being undertaken by Dako. Registration is pending but is scheduled to be completed
prior to consideration of the co-dependent technology submission by MSAC.
Proposed MBS listing
In light of the co-dependency issues between PD-L1 testing on NSCLC tumours and treatment with
pembrolizumab, MSD has received advice from the Department that a new MBS item number should
be used as a placeholder through the assessment process. This arrangement provides MSAC with the
flexibility to recommend a new MBS item number be created specifically for PD-L1 testing associated
with access to pembrolizumab, should they deem it necessary.
Category 6 – Pathology Services
MBS item number
Immunohistochemical examination of biopsy material by immunoperoxidase or other labelled antibody
techniques using a PD-L1 antibody to determine if the requirements relating to programmed cell death ligand 1
(PD-L1) status for access pembrolizumab under the Pharmaceutical Benefits Scheme (PBS) are fulfilled
Fee: To be determined Benefit: To be determined
This proposed MBS listing is the same as that nominated in Protocol 1414, i.e. the same MBS item
number would be used to test for PD-L1 expression, regardless of whether this is to determine
eligibility to pembrolizumab in the first or later lines of therapy.
Expected utilisation
An estimate of the size of the testing population is provided below. The proposed incidence of
NSCLC is comparable to that determined by the Assessment group for ALK testing. Given that the
preferred clinical place in therapy of the PD-L1 test is at time of initial diagnosis (see Clinical
Management Algorithm section), the co-dependent submission subject of this protocol is not
expected to result in a change in utilisation of the test, relative to that nominated in Protocol 1414.
Table 5: Incidence of NSCLC
MSAC website: www.msac.gov.au
Page 11 of 23
No. of patients including all lung cancers (2014)
Incidence of all NSCLC
Eligible patient pool for PD-L1 testing
11,5801
66% (based on 2002-2007)2
7,643
1Cancer
in Australia: an overview 2014, AIHW, Table 3.2 Pg 17 of document, Cancer in Australia: an overview 2014, AIHW [accessed May 5
2015]
2Lung cancer in Australia: an overview, AIHW, Lung cancer in Australia: an overview, AIHW Pg 24
31250-ALK-Final DAP-Accessible, Pg 8 of document
Reference standard
Currently, there is no established evidentiary standard for PD-L1 testing.
As the Market Ready was used to determine eligibility for enrolment in the KN024 and KN042 trials,
MSD nominates the Market Ready Assay as the evidentiary standard for PD-L1 expression testing
associated with pembrolizumab treatment.
Delivery of proposed medical test
Where service would be delivered
As IHC is a common procedure in histopathology laboratories and as PD-L1 expression is anticipated
to be identified frequently (in approx. 61% of cases for ≥1% PD-L1 expression; 23% for ≥50% PD-L1
expression (Garon et al, 2015), it is proposed that PD-L1 IHC testing be eligible to be carried out in
any pathology laboratory holding the appropriate accreditation to claim pathology services through
the MBS.
In practice, it is anticipated that the majority of PD-L1 testing would occur in pathology laboratories
associated with a public hospital. Whilst many patients for whom PD-L1 testing is done would be
outpatients (MBS pays testing costs), some patients may also be inpatients (state government pays
testing costs).
Consistent with introduction of diagnostic tests associated with access to other targeted therapies,
pathologist training and quality assurance programs would be expected to be developed with
respect to delivery of diagnostic tests for access to treatments targeting the PD-1 pathway on the
PBS.
By whom
A certified pathologist would be responsible for conducting the test and reporting the results.
Frequency of testing
Patients would require only 1 PD-L1 test through the course of their disease. The test should be
undertaken prior to commencement of pembrolizumab to enable identification of those patients
most likely to benefit from treatment. Potential options regarding the clinical place in therapy of the
PD-L1 test are outlined in the section entitled Clinical Management Algorithm.
There is no known role for PD-L1 testing in monitoring a patient’s response to pembrolizumab
treatment.
Co-dependent information
MSAC website: www.msac.gov.au
Page 12 of 23
Co-dependent drug
Pembrolizumab is the co-dependent pharmaceutical medicine.
In line with the clinical trial data from KN024, the initial application will propose reimbursement of
pembrolizumab for treatment naïve Stage IV NSCLC patients whose tumours are EGFR wild type and
ALK translocation negative, and have a high expression of PD-L1 (≥50% staining).
A further submission could be lodged in the future in line with the clinical trial data from KN042. This
submission would propose to change the reimbursement criteria for pembrolizumab to allow use in
treatment naïve Stage IV NSCLC patients whose tumours are EGFR wild type and ALK translocation
negative, and express PD-L1 regardless of level of expression (≥1% staining).
Comparator
Test
Given that the evidentiary standard and the test to be used in clinical practice are the same, unlike
Protocol 1414, no comparison of PD-L1 assays (e.g. Clinical Trial Assay and Market Ready assay) is
needed for the purposes of this submission.
As a supplementary comparison, it is proposed that an assessment of comparative assay
performance of the Market ready assay and any alternate PD-L1 test(s) reported in the public
domain be presented for consideration by MSAC. This assessment will also consider alternative cut
points used for alternative PD-L1 tests.
Drug
In treatment naïve patients with metastatic NSCLC whose tumours are EGFR wild type, ALK
translocation negative, platinum-based therapy such as carboplatin + gemcitabine is the most
frequently used treatment option. Pemetrexed maintenance may also be used in nonsquamous
NSCLC patients who respond to chemotherapy.
Co-dependence
It is proposed that the MSAC submission presents efficacy, safety and cost effectiveness comparisons
of PD-L1 testing and pembrolizumab with


No PD-L1 testing and management with platinum-based therapy
No PD-L1 testing and management with pembrolizumab
Clinical claim for the proposed medical service
The hypothesis being tested in the KN024 and KN042 clinical trials is that PD-L1 testing followed by
pembrolizumab as a first line treatment in patients with metastatic NSCLC whose tumours are EGFR
wild type and ALK translocation negative and express PD-L1 (≥1% and ≥50%) is associated with
improved health outcomes. It will be driven by two factors:
1.
Acceptable safety and analytical performance of PD-L1 test (to be assessed by MSAC).
MSAC website: www.msac.gov.au
Page 13 of 23
2.
Superior effectiveness with acceptable safety of treating PD-L1 positive patients with
pembrolizumab relative to standard of care (to be assessed by PBAC).
3.
Acceptable clinical utility of the PD-L1 test
Expected health outcomes relating to the medical service
PD-L1 Test Outcomes
Outcome measures suitable to assess the analytic performance of PD-L1 IHC testing include:





Sensitivity
Specificity
Positive Predictive Value
Negative Predictive Value
Receiver Operating Characteristic (ROC)
Measures of comparative performance of PD-L1 testing methods:


Rates of re-testing
Concordance with other commercially available PD-L1 antibodies
Measure of Clinical Utility of PD-L1 test

Health outcomes with pembrolizumab in NSCLC population whose tumours express PD-L1
compared to health outcomes with pembrolizumab in a population who have not had a PDL1 test
Other considerations







Rates of re-biopsy
Anticipated test turnaround time
The estimated number of patients being tested
The number of patients tested per case of PD-L1 positive (≥1% and ≥50%) result detected
The number of patients tested per case of PD-L1 positive (≥1% and ≥50%) result treated with
pembrolizumab
The cost of testing per case of PD-L1 positive (≥1% and ≥50%) NSCLC detected
The cost of testing per case of PD-L1 positive (≥1% and ≥50%) NSCLC treated with
pembrolizumab.
Drug Outcomes
Measures of clinical efficacy for pembrolizumab include:
Primary outcome:



Overall survival
Progression free survival
Safety and tolerability
Secondary outcomes
MSAC website: www.msac.gov.au
Page 14 of 23






Objective tumour response rates (complete response or partial response according to RECIST
and irRC criteria)
Quality of life
Disease control rate (response rate + rate of stable disease)
Duration of response
Rate of disease progression
Time to progression
Risks to patient
PD-L1 testing is performed on tissue slices taken from a biopsy specimen obtained as part of
standard diagnostic work-up and thus, in itself, does not incur any risks to patient.
The main risk to patient would occur if a re-biopsy is required in order to obtain tissue to perform
the IHC test. Re-biopsies can result in complications such as pneumothorax and haemorrhage.
These complications are considered to occur in 14% of cases5. A re-biopsy would be required if
insufficient tissue is retrieved from the initial biopsy to undertake the desired biomarker tests.
However, it is unlikely that the re-biopsy would be required specifically to undertake PD-L1 testing
alone as IHC only uses a small amount of tissue (one 4 micron section, compared to approximately
50 microns for EGFR testing). Instead the re-biopsy would be required to undertake all biomarker
tests relevant to the patient. Hence, introduction of this test is not expected to result in an increase
in re-biopsy rates.
Type of economic evaluation
As KN024 and KN042 are designed as superiority trials, it is anticipated that a cost-utility evaluation
will be presented.
Fee for the proposed medical service
Proposed funding
It is proposed that PD-L1 testing should be a “pathologist determinable test”, in line with all other
IHC tests.
Direct costs of equipment/resources used with service
IHC testing is a well established technique in all major pathology labs. Laboratories already have the
platform infrastructure and reagents to perform PD-L1 IHC testing. The PD-L1 antibody is the only
additional resource required.
5
Pg 7 1161-FinalPSD-Aug2013 Gefitinib first line testing for mutations of epidermal growth factor receptor (EGFR) in patients
with metastatic non-small cell lung cancer (NSCLC)
MSAC website: www.msac.gov.au
Page 15 of 23
The proposed fee
The final fee request has yet to be determined. It is expected to be consistent with other fees for
immunohistochemistry and will be based on consideration of the capital and the labour components
required for pathologists to undertake PD-L1 testing and interpret and report the results.
Clinical Management Algorithm - clinical place for the proposed
intervention
Current treatment algorithm
The current treatment algorithm is outlined in Figure 2. After histological confirmation of NSCLC,
biomarker tests are conducted (for EGFR and ALK) on all nonsquamous and NOS (not otherwise
specified) patients. If the patient has Stage IV NSCLC, these tests determine first line treatment. If
the tumour is EGFR mutant or ALK translocation positive, patients are treated with targeted therapy
first (erlotinib/gefitinib for EGFR and crizotinib for ALK). All other patients (non-squamous patients
who are EGFR wild type or ALK translocation negative and those with squamous histology) will be
treated with a platinum doublet (e.g. carboplatin + gemcitabine) as the initial therapy. Pemetrexed
is used as a first line maintenance therapy for some non-squamous patients without progressive
disease.
As the co-dependent technology submission associated with Application 1414 proposes that the PDL1 test be undertaken as part of the initial diagnostic workup of a Stage IV NSCLC patient, there will
be no change required to the placement of the test with Application 1440, should this be approved.
Future treatment algorithms
The sponsor proposes that PD-L1 testing is undertaken in both squamous and nonsquamous
tumours on recently cut (within 6 months) sections from initial biopsy of Stage IV patients. It should
be performed at the time of other biomarker assessments, i.e. PD-L1 IHC could be done alongside
ALK IHC and other diagnostic IHC tests and in parallel to EGFR testing (Figure 3) and undertaken in
patients with Stage IV NSCLC . From a practical perspective in this scenario, sections for all testing
would be cut at the same time. IHC testing would be performed on the first lot of sections with the
residual sections sent away for EGFR testing. This scenario has support from pathologists and
oncologists as the most efficient and useful place for testing. Patients whose tumours express PD-L1
and are EGFR wild type and ALK translocation negative will be treated with pembrolizumab instead
of platinum-based therapy in accordance with the PBS criteria accepted by the PBAC.
MSAC website: www.msac.gov.au
Page 16 of 23
Figure 2: Current treatment algorithm
Patient suspected of NSCLC undergoes biopsy
Confirmation of NSCLC diagnosis with histology/cytology/ Testing for EGFR1, ALK1
EGFR wildtype/ ALK neg
Platinum-based chemotherapy
(optional pemetrexed maintenance if non squamous)
1
MSAC website: www.msac.gov.au
non squamous or NOS histologies only
Page 17 of 23
Figure 3: Treatment algorithm showing PD-L1 testing and subsequent treatment with pembrolizumab
Patient suspected of NSCLC undergoes biopsy
Confirmation of NSCLC diagnosis with histology/cytology/ Testing for EGFR1, ALK1 and PD-L12
EGFR wildtype/ ALK neg/PD-L1 negative
EGFR wildtype/ ALK neg/PD-L1
Platinum-based chemotherapy
(optional pemetrexed maintenance if non squamous)
Pembrolizumab
1
non squamous or NOS histologies only
histologies if stage IV
2 All
MSAC website: www.msac.gov.au
positive
Page 18 of 23
Regulatory Information
Regarding the PD-L1 testing, the regulatory process will be managed by Dako. Regulatory approval
of the PD-L1 test is expected prior to MSAC consideration of this co-dependent technology
submission.
In the future, MSD anticipates filing a TGA submission for pembrolizumab in treatment naïve
patients with NSCLC whose tumours express PD-L1 and are EGFR wildtype and ALK translocation
negative
Decision analytic
An assessment of the cost-effectiveness of introducing PD-L1 testing to determine patient eligibility
to pembrolizumab should take into account the parameters outlined in Table 6,
Table 7 and Table 8
MSAC website: www.msac.gov.au
Page 19 of 23
Table 6: Summary of PICO to define research question
PICO
Comments
Patients
Intervention
Comparator
Outcomes
Patients with EGFR wild type and ALK negative non-small cell lung cancer
who have had no prior treatment, and whose tumours express PD-L1 (≥50%
and ≥1%)
Test
Immunohistochemistry testing for PD-L1 to determine if the proposed PBS
requirements relating to access to pembrolizumab are fulfilled
Drug
Pembrolizumab
Co-dependence
Access to pembrolizumab in patients who fulfil the PBS requirements with
regards to PD-L1 expression status determined by PD-L1 IHC testing.
Test
No PD-L1 testing.
A supplementary comparison will be presented between the evidentiary
standard (Market Ready Assay) and any alternative PD-L1 tests for which
there is data in the public domain.
Drug
Main comparator: Carboplatin/cisplatin+gemcitabine, followed by
pemetrexed maintenance in appropriate patients
Co-dependence
No PD-L1 testing and management with carboplatin/cisplatin+gemcitabine,
followed by pemetrexed maintenance in appropriate patients
No PD-L1 testing and management with pembrolizumab
Test
Outcome measures suitable to assess the analytic performance of PD-L1 IHC
testing include:
 Sensitivity
 Specificity
 Positive Predictive Value
 Negative Predictive Value
 Receiver Operating Characteristic (ROC)
Measures of comparative performance of PD-L1 testing methods:
 Concordance with other commercially available PD-L1 antibodies
 Rates of re-testing
Measures of clinical utility
 Health outcomes with pembrolizumab in NSCLC patients whose
tumours express PD-L1 compared to health outcomes with
pembrolizumab in all patients
Other considerations
 Rates of re-biopsy
 Anticipated test turnaround time.
 The estimated number of patients being tested
 The number of patients tested per case of PD-L1 positive (≥50% and
≥1%) result detected
 The number of patients tested per case of PD-L1 positive (≥50% and
≥1%) result treated with pembrolizumab
 The cost of testing per case of PD-L1 positive(≥50% and ≥1%) NSCLC
MSAC website: www.msac.gov.au
Page 20 of 23
PICO
Comments
detected
The cost of testing per case of PD-L1 positive (≥50% and ≥1%) NSCLC
treated with pembrolizumab.
Drug Outcomes
Measures of clinical efficacy for pembrolizumab include:
Primary outcome:
 Overall survival
 Progression free survival
Secondary outcomes
 Objective tumour response rates (complete response or partial
response according to RECIST and irRC criteria)
 Quality of life
 Disease control rate (response rate + rate of stable disease)
 Duration of response
 Rate of disease progression
 Time to progression
 Safety and tolerability.

Table 7: For investigative services
Prior tests
Initial biopsy and tests to confirm diagnosis of NSCLC and EGFR/ALK testing
if nonsquamous or NOS.
Reference standard
The evidentiary standard is the Market Ready Assay
Healthcare resources
Healthcare resources that are most likely to be affected, should PD-L1 testing and treatment with
pembrolizumab become available include (see Table 8):







Cost of the PD-L1 antibody and pathologists time in interpreting and reporting the
results. Pathology laboratories are likely to have all the required equipment for IHC as it is
routinely performed.
Costs of a second biopsy if there is insufficient tissue for PD-L1 testing
Costs of treating PD-L1 positive patients with pembrolizumab
Cost offsets from reduced use of displaced treatments.
Costs for treating adverse events from treatment (with any therapeutic agent).
Costs associated with ongoing patient monitoring, e.g. physician visits.
Health care resources associated with initial diagnosis are assumed to remain unchanged
and may be excluded from the analysis accordingly.
Questions for public funding
Primary question for public funding
What is the safety, effectiveness, and cost-effectiveness of PD-L1 testing to determine eligibility for
pembrolizumab treatment in NSCLC patients whose tumours are EGFR wild type and ALK
translocation negative and express PD-L1 (≥50%
MSAC website: www.msac.gov.au
Page 21 of 23
Table 8: List of resources to be considered in the economic analysis
Provider of
resource
Setting in which
resource is
provided
Resources provided to identify eligible population
Equivalent to current practice
Proportion
of patients
receiving
resource
Number of Disaggregated unit cost
units of
resource per
relevant
Other
Private
Safety
time horizon MBS
government health
nets*
per patient
budget
insurer
receiving
resource
To be
provided in
submission
To be
provided in
submission
Resources provided to deliver proposed intervention (PD-L1 IHC test and pembrolizumab)
To be
To be
PD-L1 IHC testing
MBS
Pathology lab
Additional biopsy (if required)
MBS
Public or private
hospital
provided in
submission
To be
provided in
submission
Resources provided in association with proposed intervention
To be
Pembrolizumab for patients
PBS
Outpatient
provided in
deemed eligible based on PBS
submission
criteria
Administration cost for
Hospitals/MBS Inpatient/outpatient To be
provided in
pembrolizumab
and public and
submission
private hospitals
To be
Physician visits (Oncologist or
MBS
Outpatient
provided in
respiratory physician)
Clinical monitoring (radiological or MBS
other imaging, blood counts)
Outpatient
Treatment of adverse events
Outpatient
PBS
submission
To be
provided in
submission
To be
provided in
submission
To be
provided in
submission
To be
provided in
submission
To be
provided in
submission
To be
provided in
submission
To be
provided in
submission
To be
MSAC website: www.msac.gov.au
Patient
Total cost
Page 22 of 23
Provider of
resource
Setting in which
resource is
provided
Proportion
of patients
receiving
resource
Number of Disaggregated unit cost
units of
resource per
relevant
Other
Private
Safety
time horizon MBS
government health
nets*
per patient
budget
insurer
receiving
resource
provided in
submission
provided in
submission
Patient
Total cost
Resources provided in association with comparator 1 (no testing followed by docetaxel)
(e.g., pre-treatments, co-administered interventions, resources used to monitor or in follow-up, resources used in management of adverse events, resources used for
treatment of down-stream conditions)
To be
To be
Pharmaceuticals (relevant prePBS
Outpatient
provided in
provided in
medications,
submission
submission
carboplatin/cisplatin+gemcitabine)
followed by pemetrexed
maintenance if appropriate
To be
Administration cost for
Hospitals/MBS Inpatient/outpatient To be
provided
in
provided in
carboplatin/cisplatin+gemcitabine
and public and
submission
submission
followed by pemetrexed
private hospitals
maintenance if appropriate
To be
To be
Physician visits (Oncologist or
MBS
Outpatient
provided in
provided in
respiratory physician)
Clinical monitoring (radiological or MBS
other imaging, blood counts)
Outpatient
Treatment of adverse events
Outpatient
PBS
submission
To be
provided in
submission
To be
provided in
submission
submission
To be
provided in
submission
To be
provided in
submission
* Include costs relating to both the standard and extended safety net.
MSAC website: www.msac.gov.au
Page 23 of 23
References
Garon EB, Rizvi NA, Hui R, et al. Pembrolizumab for the treatment of non–small-cell lung cancer. N
Engl J Med. DOI: 10.1056/NEJMoa1501824
Gandhi et al, 2014, MK-3475 (anti-PD-! Monoclonal antibody for non-small cell lung cancer (NSCLC):
Antitumour activity and association with tumor PD-L1 expression, Cancer Res 2014;74: CT105.
Abstract
Herbst et al, 2015, Pembrolizumab versus docetaxel for previously treated,PD-L1-positive, advanced
non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial, DOI:
http://dx.doi.org/10.1016/S0140-6736(15)01281-7
MSAC website: www.msac.gov.au

Similar documents

×

Report this document