Courting Controversy: Right Choice of Therapy for Metastatic Non-Small Cell Lung Cancer
The index patient is a 45 year old, male, smoker, with no comorbidities except well-controlled
diabetes mellitus, who presented with cough with hemoptysis. Clinical examination
was nonproductive; however, chest X-ray revealed blunting of left costophrenic angle.
Computed tomography (CT) scan and later positron emission tomography scan revealed
an evidence of 4 cm left lower lobe mass, 4 liver metastases (largest being 3 cm in
size), multiple left-sided pleural nodules, and mild pleural effusion. VATS-guided
biopsy was positive for adenocarcinoma. Tumor genomic testing was negative for targetable
EGFR, ALK, ROS, MET, and BRAF alterations. All blood tests were found to be within
normal limits, except for mild elevation of alkaline phosphatase. PDL1 testing with
PD-L1 22C3 pharmDx as well as SP142 assay, both reveal >90% positivity in tumor cells.
The patient is well educated and requests to offer him the best possible treatment
that can maximize his overall survival (OS) with least possible side effects and disruption
of his quality of life. The patient is reimbursable and can afford all treatments.
After carefully considering his requests, you decide to offer him:
-
Paclitaxel, platinum, atezolizumab, and bevacizumab
-
Single-agent pembrolizumab
-
Pemetrexed, platinum, pembrolizumab
-
Pemetrexed, platinum, atezolizumab
-
Nab-paclitaxel, platinum, atezolizumab
-
Any other.
I would offer this patient a single-agent pembrolizumab keeping in mind the histology,
PDL1 status, age, performance status, site of metastases, survival estimates, and
expected side effects of therapy. All of the options given are effective therapies
for patients of metastatic nonsmall cell lung cancer (NSCLC). This highlights the
revolution brought about by immunotherapy in the management of NSCLC in the last decade.
Historically, patients of metastatic NSCLC without actionable target mutations were
treated with platinum-based combination chemotherapy with median progression-free
survival (PFS) and OS not exceeding 9–12 months. The development of checkpoint inhibitors
and companion diagnostic tests has changed the landscape of metastatic NSCLC management,
and I have a big basket to choose from when offering therapy. I will try to rationalize
my choice in the paragraphs below using the available data.
Approximately 23%–28% of advanced NSCLC patients have a high PDL1 expression (defined
as membranous PDL1 expression on 50% or more tumor cells, regardless of intensity
of expression).[1] Pembrolizumab is a highly selective humanized anti-PD1 molecule which is now an
Food and Drug Administration-approved therapy as a single agent for advanced NSCLC
with PDL1 tumor proportion score of 50% or higher.[2] Keynote-024 study showed a superior OS, PFS, and overall response rates (ORRs) of
single-agent pembrolizumab over platinum-based chemotherapy as shown in [Table 1].[3],[4] The updated analysis for OS showed that the benefit was maintained, despite significant
crossover (approximately 54% of chemotherapy arm patients received pembrolizumab at
progression).[4]
Table 1
Comparison of key outcomes in KEYNOTE-024 and KEYNOTE-189 study
|
Trials
|
KEYNOTE-024 (pembrolizumab alone)
|
KEYNOTE-024 (chemotherapy alone)
|
KEYNOTE-189* (pembrolizumab plus chemotherapy)
|
KEYNOTE-189* (chemotherapy alone)
|
|
*PFS and OS shown are for subgroup of population with PDL1 > 50%. PFS - Progression-free
survival; OS - Overall survival; CR - Complete response; PR - Partial response; HR
- Hazard ratio; PDL1 - Programmed death-ligand 1; NR - Not reached
|
|
PFS (months, HR)
|
10.3 (0.50)
|
6.0
|
11.1 (0.36)
|
4.8
|
|
OS (months, HR)
|
30 (0.63)
|
14
|
NR (0.59)
|
10.1
|
|
OS rate at 2 years (%)
|
51.5
|
34.5
|
51.9
|
39.4
|
|
Response rates (CR + PR) (%)
|
44.8
|
27.8
|
62.1
|
24.3
|
|
Grade ¾ toxicity (%)
|
26.6
|
53.3
|
67.2
|
65.8
|
|
Treatment discontinuation rates (%)
|
7.1
|
10.7
|
13.8
|
7.9
|
Pembrolizumab has also been studied in combination with chemotherapy for the treatment
of PDL1 unselected advanced nonsquamous NSCLC in Phase 3 KEYNOTE-189 study.[5] It also demonstrated superior OS and PFS of pembrolizumab plus chemotherapy versus
chemotherapy alone in the entire population.[6] The median PFS, OS, and ORR were significantly higher in population subset with
PDL1 >50% as shown in [Table 1].[6]
There is no trial till date which has evaluated chemotherapy plus pembrolizumab versus
pembrolizumab alone in high PDL1 population. The conundrum to use pembrolizumab as
a monotherapy versus combination remains a controversial area for such patients. Cross-trial
comparisons between KEYNOTE-024 and the PDL1-high subgroup of KEYNOTE-189 suggest
comparable outcomes between pembrolizumab and chemotherapy and pembrolizumab alone
[Table 1].[3],[6] This is despite the fact that there was superior response rates in PDL1 high subset
of patients in KEYNOTE-189 (61%) as compared to monotherapy in KEYNOTE-024 (44%).[4],[6] However, this comparison is only hypothesis generating and formal conclusions can
only be drawn after a head-to-head trial comparisons between the two. It is also notable
that the corresponding chemotherapy control arms experienced somewhat different 12-month
OS rates (55% and 48% in KEYNOTE-024 and KEYNOTE-189 [PD-L1-high subgroup], respectively).[3],[5]
There was however a notable difference in rates of grade 3-4 toxicity between immunotherapy
arms of the two studies, and as shown in [Table 1], there was 40% more Grade ¾ toxicity observed with chemotherapy combination.[3],[5]
To answer this question, a meta-analysis of five randomized clinical trials was done
in patients with high PDL1 expression (>50% PDL1 score) comparing pembrolizumab monotherapy,
pembrolizumab combined with chemotherapy, versus chemotherapy alone.[7] Both the immunotherapy arms were superior to chemotherapy-alone arm in terms of
PFS and OS. Indirect comparison showed that pembrolizumab plus chemotherapy was superior
to pembrolizumab alone, in terms of ORR (relative risk 1.62, 1.18–2.23) and PFS (hazard
ratio [HR] 0.55, 0.32–0.97). A trend toward improved OS was also observed (HR 0.76,
0.51–1.14); however, it was not statistically significant.[7] The result of this meta-analysis can be very tempting for a physician to choose
the combination over single-agent pembrolizumab alone, but indirect comparisons and
potential for severe toxicities with a similar survival rates cast a doubt on the
results.
In the absence of direct comparative data and keeping in mind the patient's desire
for maximal survival with minimal toxicity, I have chosen pembrolizumab monotherapy
for the patient, thus allowing for the option of using a platinum-based doublet in
the second-line setting.
Atezolizumab in combination with nab-paclitaxel and carboplatin is also approved as
a first-line therapy for patients with nonsquamous NSCLC in PDL1 unselected population.
The approval for this is based on the pivotal Phase 3 Study IMPower 130.[8] The combination of atezolizumab with nab-paclitaxel and carboplatin significantly
improved PFS and OS versus chemotherapy alone as shown in [Table 2].[8] Nab-paclitaxel was chosen as it does not require a steroid premedication, which
may affect the response to immunotherapy. Grade 3–4 adverse events noted in this study
were 71.3%, making it a toxic therapy. Liver metastases were present in approximately
15% of each arm of the study. The subgroup of patients with baseline liver metastases
failed to show the PFS and OS benefit.[8] The high incidence of Grade 3–4 toxicity and poorer outcomes of patients with baseline
liver metastases makes the use of atezolizumab in combination with nab-paclitaxel
and carboplatin, a less preferred option for our patient who has multiple liver metastases.
Table 2
Key outcomes in IMPower 130 and IMPower 150 studies
|
IMPOWER 130* Atezolizumab + chemotherapy
|
IMPOWER 130* Chemotherapy
|
IMPOWER 150 ABCP
|
IMPOWER 150 BCP
|
|
*For PDL1 high subgroups. PFS - Progression-free survival; OS - Overall survival;
HR - Hazard ratio; PDL1 - Programmed death-ligand 1; mets - Metastases
|
|
PFS (months, HR)
|
6.4 (0.51)
|
4.6
|
8.3 (0.62)
|
6.8
|
|
OS (months, HR)
|
17.3 (0.84)
|
16.9
|
19.2 (0.78)
|
14.7
|
|
PFS in baseline liver mets (months, HR)
|
4.2 (0.93)
|
4.4
|
8.2 (0.41)
|
5.4
|
|
OS in baseline liver mets (months, HR)
|
10.0 (1.04)
|
8.8
|
13.3 (0.52)
|
9.4
|
The presence of hemoptysis which is a contraindication to the use of bevacizumab makes
atezolizumab, bevacizumab, paclitaxel, and platinum (ABCP) an unacceptable option
for our patient. ABCP had otherwise shown superiority over combination of bevacizumab,
paclitaxel, and platinum (BCP) in IMPower 150 study of advanced, PDL1 unselected nonsquamous
NSCLC patients.[9] An updated subgroup analysis of IMPower 150 demonstrated improved PFS and OS in
patients with baseline liver metastases [Table 2], a finding which was not observed in the subset analysis of immunotherapy arm of
IMPower 130.[8],[10] IMPower 150 had a third arm of atezolizumab, paclitaxel, and platinum (ACP). The
ACP arm failed to show superiority over BCP in patients with baseline liver metastases.[10]
The combination of atezolizumab with pemetrexed and carboplatin has been compared
to pemetrexed and carboplatin in advanced nonsquamous NSCLC in IMPower 132 study which
demonstrated improved PFS (7.6 months vs. 5.2 months; HR 0.60), but there was only
trend toward difference in OS at the time of first analysis (18.1 vs. 13.5 months;
HR 0.46).[11] In view of nonavailability of robust data for OS for this regimen at this time,
it cannot be recommended as a first-line therapy for our patient.
We have seen how there are a plethora of choices while choosing therapy for metastatic
NSCLC in the first-line setting. The right choice of therapy can be decided by the
presence of molecular driver alterations, PDL1 expression, comorbidities, performance
status, burden of disease, and most importantly financial abilities of the patient.
The various trials discussed above can guide us in choosing therapy, but the final
decision must always be individualized after an informed decision-making with patient
and physician. For example, in a young patient with high burden of disease, a significant
response will be an ideal primary goal and a combination of chemotherapy and immunotherapy
will be the best option. On the other hand, for an elderly gentleman with comorbidities
and borderline performance status, single-agent immunotherapy will be a preferred
choice where minimal toxicity and good quality of life with maximal survival advantage
are desired. Unfortunately, not much data are available regarding the efficacy of
immunotherapy drugs in Indian patients, but we can always extrapolate the data available
from studies from the west. Another major concern for patients in a third world country
is financial status of the family. A good fraction of patients presenting in Indian
hospitals cannot afford traditional chemotherapy; immunotherapy will be out of reach
of most patients which is 40–80 times costlier. It will still be long before chemotherapy
can be completely done away in the management of lung cancer, particularly in an Indian
healthcare system.
