Keywords Gefitinib - pemetrexed - second-line lung cancer
Introduction
Platinum doublet chemotherapy has been the historical backbone of therapy for advanced
metastatic non-small cell lung cancer (NSCLC).[1 ] Pemetrexed platinum doublet is accepted standard of care for chemotherapy-naive
nonsquamous NSCLC.[2 ] However, recently, use of tyrosine kinase inhibitors (TKIs) in epidermal growth
factor receptor (EGFR) mutation-positive metastatic NSCLC has shown to significantly
improve progression-free survival (PFS) when used upfront compared to chemotherapy.[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]
[10 ]
[11 ] However, caveat to these studies is that, except for LUX-Lung 3, all have nonpemetrexed
doublet in comparator arm and have few patients from Indian Subcontinent.[12 ]
[13 ] Moreover, none had maintenance pemetrexed in control arm, which is now an accepted
standard worldwide. We had earlier reported outcomes of Phase III randomized trial
comparing first-line TKI and pemetrexed doublet in EGFR mutation-positive NSCLC.[14 ] We further report post hoc analysis of outcomes and toxicities of patients who progressed
after first-line therapy including maintenance pemetrexed and managed to receive second-line
chemotherapy or TKIs in EGFR mutation positive NSCLC.
Materials and Methods
Our group from tertiary cancer center at Mumbai, India, has initially reported outcomes
of Phase III, double-arm, single-center, open-label randomized control trial comparing
gefitinib versus pemetrexed platinum doublet in first-line EGFR mutation-positive
nonsquamous advanced metastatic NSCLC. Out of 290 patients enrolled, 257 progressed
after first-line therapy. One hundred and eighty-seven out of these 257 went on to
receive second-line physician choice agents; remaining others were not eligible due
to various reasons such as poor performance status, progressive brain metastasis,
asymptomatic progression, patient not willing for further therapy, and clinician decision
[CONSORT diagram [Figure 1 ]. This study was approved by the Institutional Ethics Committee of our Institute.
Figure 1: CONSORT diagram: Schematic representation of patients on prospective randomised
trial who received second line therapy after progression to first line
Demographic and clinical profile of patients who went on to receive physician choice
second-line therapy was recorded including age, sex, performance status before initiation
of second-line therapy, smoking status, EGFR mutation, and tobacco chewing. For EGFR
mutation analysis, DNA was extracted from formalin fixed, paraffin-embedded tumor
block and amplified for exon 18, 19, and 21 using nested PCR which has been standardized
and validated at our institute molecular laboratory.[15 ]
[16 ]
As per physician choice, any of the following schedules of drug administration was
followed:
Gefitinib 250 mg daily per oral until progressive disease, intolerable toxicity, or
other approved criteria for discontinuation
Pemetrexed 500 mg/m2 intravenous over 10 min with or without carboplatin (area under
the curve [AUC] =5) over 60 min intravenous infusion, with appropriate antiemetic
prophylaxis along with B12 (1000 mcg intramuscularly 1 week before chemotherapy and
repeated every 12 weeks) and folic acid (5 mg once a day daily) continued till 3 weeks
after last dose of pemetrexed
Weekly paclitaxel at 100 mg/m2 infused over 60 min, continued till progressive disease,
Grade II neuropathy or other intolerable toxicity
Gemcitabine 1000 mg/m2 infused over 20 min D1 and D8, with or without cisplatin 75
mg/m2 over 60 min with adequate intravenous hydration or carboplatin (AUC = 5) over
60 min intravenous infusion
Response evaluation was done at every 12 weeks with contrast-enhanced computerized
tomography of chest and associated metastatic sites in patient receiving second-line
therapy including gefitinib and chemotherapy, recorded and compared as per RECIST
criteria version 1.1. Toxicity with the use of second-line agents was prospectively
recorded at every cycle visit of chemotherapy and every 2 weeks for the first 2 months
for TKIs and thereafter monthly. Patients were followed through clinical charts in
each outpatient department visit during and after completion of therapy including
telephonic calls.
PFS was defined as time interval from the start of second-line agent to objective
progressive disease and change in treatment or death from any cause. Overall survival
(OS) was defined as time interval from start of second-line agent till death from
any cause. PFS and OS were calculated as per Kaplan–Meir statistics. Log-rank test
was used to compare the impact of several prognostic factors such as age, sex, smoking,
tobacco use, EGFR mutation, performance status, response rates, on PFS and OS.
Results
Patients were enrolled in prospective randomized Phase III trial from February 2012
to April 2016, all having EGFR classical mutation and metastatic advanced nonsquamous
NSCLC. One hundred and eighty-seven out of 257 (73%) patients, who progressed after
first-line therapy, went on to receive physician choice second-line therapy. Gefitinib
was the most common agent used (n = 113) followed by pemetrexed platinum doublet (n = 58), weekly paclitaxel (n = 6), pemetrexed alone (n = 1), and other agents (n = 9). Male patients and smokers were 110 and 39, respectively. One hundred and forty-three
(76%) patients had a good performance status (PS 0–1), whereas 14 (7.5%) had PS 3
before starting second-line agents. Eighty-eight out of one hundred and eighty-seven
(47%) patients underwent re-biopsy after progression on first-line agents. Most common
site of biopsy was lung (n = 79), followed by extrathoracic site (n = 6) and lymph node (n = 3).
Of all patients re-biopsied, three patients had nonadenocarcinoma histology, with
squamous cell, small cell, and poorly differentiated carcinoma each, while remaining
77 had similar adenocarcinoma histology, and in 8 patients, sample was not representative
[Table 1 ]. Gefitinib was used in 36/64, 76/116, and 1/4 patients with exon 21, exon 19, and
exon 18 mutation, respectively, while remaining patients received nongefitinib chemotherapeutic
agents.
Table 1
Demographic, clinical, and treatment profile of patients on second-line therapy in
nonsquamous mutation-positive nonsmall cell lung cancer
Profile (n =187)
Characteristics
Remarks (%)
SD – Stable disease; PD – Progressive disease; PR – Partial response; CR – Complete
response; PS – Performance status
Age
Median 54 years
Range 27-76 year
Sex
Male
110
59
Female
77
41
Smoking
Yes
39
21
No
148
79
Tobacco chewing
Yes
75
40
No
112
60
EGFR mutation
Exon 21
64
34
Exon 19
119
64
Exon 18
4
2
PS
PS 0-1
143
76
PS 2
30
16
PS 3
14
7.5
Second line agents
Gefitinib
113
60
Others (nongefitinib)
74
40
Pemetrexed platinum
58
31
Other chemotherapy
16
9
Response rates
Gefitinib
Nongefitinib
SD+ PD
53
56
CR + PR
60
18
Median follow-up of patients is 14 months. Patients receiving gefitinib had better
response rates’ complete response partial response (CR + PR) compared with nongefitinib
group, 60/113 (53%) versus 18/74 (24%) when used in second line in EGFR mutation positive,
metastatic NSCLC. Among nongefitinib group (n = 74), pemetrexed platinum doublet had better response rate (CR + PR), 16/58 (28%)
compared to other agents 2/16 (12%). Median PFS and OS of entire cohort of patients
are 5.9 months and 13 months, respectively. Gefitinib group had significantly better
PFS (7.4 months vs. 4.4 months, P = 0.001) and OS (14 months vs. 9.7 months, P = 0.03) compared to nongefitinib group.
Response to therapy (CR + PR) was significantly associated with improved PFS (10.8
vs. 3.9 months, P = 0.001) in gefitinib as well as in nongefitinib group (6.2 months vs. 3.8 months,
P = 0.007). Similarly, OS was significantly improved in both gefitinib (21.4 months
vs. 8.9 months, P = 0.01) and nongefitinib group (11.7 months vs. 8.8 months, P = 0.03) for patients with good response to therapy. None of the other prognostic
factors such as age, sex, performance status, EGFR mutation status, smoking, and tobacco
use had a significant impact on outcome, except gefitinib group had better survival
among nonsmokers [Table 2 ].
Table 2
Impact of prognostic factors in gefitinib and nongefitinib group on median progression-free
survival and median overall survival - log rank test (pair-wise over each stratum)
Factors
Group
PFS (months)
OS (months)
Gefitinib
Nongefitinib
P
Gefitinb
Nongefitinib
P
# OS is significantly better in nonsmokers compared to smokers in gefitinib group, P=0.03,
$CR+PR responses had significantly better PFS and OS in both gefitinb and nongefitinib
group. PS – Performance status, EGFR mut – Epidermal Growth Factor receptor mutation
type; SD – Stable disease; PD – Progressive disease; PR – Partial response; CR – Complete
response; NS – Not significant, OS – Overall survival
Age (years)
≤54
7.4
4.7
NS/NS
15.2
10.7
NS/NS
>54
8.3
4.9
13.4
7.4
Sex
Male
6.5
4.1
NS/NS
13.7
8.86
NS/NS
Female
8.7
4.5
15.2
11.5
Smoking
Yes
6.2
3.6
NS/NS
11.1#
9.7
0.03/NS
No
8.0
4.4
15.2#
9.7
Tobacco
Yes
6.5
4.5
NS/NS
13.8
11.2
NS/NS
No
7.6
4.4
15.2
9.5
EGFR mut
Exon 21
5.7
4.1
NS/NS/NS
11.5
8.2
NS/NS/NS
Exon 19
8.0
4.5
14.8
9.7
Exon 18
8.4
2.3
14.0
9.1
PS
0-1
7.5
5.2
NS/NS
15.7
11.3
NS/NS
2-3
6.1
4.3
11.5
7.2
Response
SD+PD
3.9
3.8
0.00/0.07$
8.9
8.8
0.00/0.03$
CR+PR
10.8
6.2
21.4
11.7
Patients receiving gefitinib had diverse but mostly low-grade toxicity compared to
nongefitinib group such as diarrhea, rash, paronychia, pruritus, anorexia, fatigue,
and dry skin [Table 3 ]. Among chemotherapy group, neutropenia was the most common toxicity including life-threating
febrile neutropenia (5.5%). After progression from second-line therapy, 87/187 (46%)
went on to receive third-line agents. Weekly paclitaxel (100 mg/m2) was the most common
regimen (43/87, 49%) followed by docetaxel (15/87) and gefitinib (10/87).
Table 3
Toxicity recorded among patients receiving gefitinib versus nongefitinib therapy,
as per Common toxicity criteria v4.03
Toxicity
Gefitinib group (n =113)
Nongefitinib group (n =74)
Group 1
Group 2
Group 3
Group 4
Group 1
Group 2
Group 3
Group 4
SGPT – Serum glutamate pyruvate transaminase; SGOT – Serum glutamate-oxaloacetate
transaminase
Anaemia
43
12
5
0
28
15
3
0
Thrombocytopena
7
4
1
1
8
3
2
1
Neutropenia
1
4
0
0
10
2
2
4
SGOT increased
31
4
2
0
14
2
0
0
SGPT increased
31
5
5
0
18
4
0
0
Hypoalbuminemia
14
12
2
0
15
6
0
0
Diarrhea
14
7
3
0
7
6
6
0
Skin rash
34
22
8
0
20
4
2
0
Paronychia
3
2
1
0
1
1
0
0
Mucositis
8
1
0
0
11
0
2
0
Anorexia
39
4
1
0
21
4
0
0
Pruritus
33
2
0
0
16
4
0
0
Nausea
8
2
0
0
9
2
0
0
Vomiting
8
3
1
0
4
0
2
0
Fatigue
45
10
2
0
24
15
3
0
Constipation
9
0
1
0
11
2
0
0
Bilirubin increase
5
1
0
0
6
0
0
0
Creatinine increase
6
3
0
0
6
3
0
0
Hypokalemia
6
1
2
0
6
2
0
0
Pneumonitis
1
2
0
0
1
1
0
1
Dry skin
38
11
1
0
13
3
1
0
Discussion
First-line pemetrexed platinum doublet and TKIs are both approved as a standard of
care in metastatic nonsquamous NSCLC.[2 ]
[3 ] However, upfront TKI has shown significant benefit in improving PFS when used upfront
compared to chemotherapy in EGFR mutation-positive NSCLC.[3 ]
[4 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]
[10 ]
[11 ] Despite the use of TKIs or pemetrexed doublet, almost all patients subsequently
suffer disease progression and become eligible to receive second-line therapy with
respect to age, performance status, and comorbidities. TKIs are a valid second-line
option for patients who progressed on first-line chemotherapy and harbor EGFR classical
mutation. Similarly, patients who progressed on first-line TKI, merit systemic chemotherapy
where pemetrexed is more rational choice compared to docetaxel in nonsquamous NSCLC,
having similar outcomes with reduced toxicities.[17 ] We report outcomes and toxicities of physician choice second-line agents who progressed
after first-line TKIs or pemetrexed platinum doublet with or without maintenance pemetrexed
in patients with EGFR mutation-positive advanced metastatic nonsquamous NSLC exclusively
from the Indian subcontinent.
In the initial studies, gefitinib in unselected patients produces response rate of
15%–20% after progression on 1–2 lines of platinum-based chemotherapy.[18 ]
[19 ] Gefitinib when compared with docetaxel as second-line therapy in randomized Phase
II/III trial did not show any improvement in PFS and OS in unselected patients.[20 ]
[21 ]
[22 ]
[23 ] However, post hoc subset analysis showed better response rate (42% vs. 21%) and
PFS with gefitinib in EGFR mutation-positive subset of NSCLC.[24 ] Similarly, our study demonstrated superior response rates in selected EGFR classical
mutation NSCLC with gefitinib compared to chemotherapy (53% vs. 24%, P = 0.03) which translated into improved PFS (7.4 months vs. 4.4 months, P = 0.001) and OS (14 months vs. 9.7 months, P = 0.03) as second-line therapy.
Several randomized Phase III trial has now compared gefitinib with nonpemetrexed doublet
in first line and shown better response rates (60%–84% vs. 30%–47%) and PFS.[7 ]
[8 ]
[9 ]
[10 ]
[11 ] LUX-Lung 3 is the only randomized Phase III trial which has compared TKI (afatinib)
with the standard of care pemetrexed platinum doublet, without maintenance, and shown
significantly improved PFS of 11.6 months versus 6.9 months.[12 ] Our study, with second-line gefitinib, consolidates the evidence that TKIs after
progression from first-line chemotherapy retains its drug sensitivity and provides
a better outcome in EGFR mutation-positive NSCLC compared to any other second-line
chemotherapy.
Hanna et al . reported largest randomized trial of pemetrexed versus docetaxel in second-line
NSCLC, where pemetrexed has shown modest response rates of 9% with PFS of 2.9 months
and OS of 8.3 months which was clinically equivalent to docetaxel albeit with reduced
toxicities.[17 ] Our study produced much better response rates (28%) with PFS and OS of 4.4 months
and 9.7 months, respectively, with pemetrexed platinum doublet second-line therapy.
This discordant outcome can be explained by the fact that all patients in our study
receiving pemetrexed platinum doublet in second line were platinum naive and have
not been exposed to any chemotherapeutic agent before. Moreover, as against Hanna
et al ., majority of our patients have received pemetrexed platinum doublet (58/74) compared
to pemetrexed alone, being platinum naive.
Majority of our patients who received second-line pemetrexed doublet had prior received
gefitinib in first line as per the provision of Phase III randomized trial and had
shown better response rate and PFS in first line.[14 ] Similar to Hanna et al ., where better response to second-line chemotherapy, was demonstrated in the same
cohort of patients who had complete or PR to first-line agents, in our study better
response to second-line pemetrexed doublet could also be attributed to higher response
produced by first-line gefitinib (65% vs. 43%) in EGFR mutation-positive NSCLC.[14 ]
[17 ] This further contributes to the fact that response to first-line agents is strong
prognostic and predictive factor of outcome with second-line agents. Pemetrexed doublet
in our study produced <5% incidence of Grade ¾ anemia and thrombocytopenia, while
febrile neutropenia was 1.3%. This is comparable to toxicity reported in cisplatin-pemetrexed
arm of LUX-Lung 3 trial and marginally more than Hanna et al ., where single agent pemetrexed was used.[12 ]
[17 ]
Rash, diarrhea, fatigue, pruritus, and paronychia are the most common toxicities seen
with gefitinib while nongefitinib chemotherapy had higher hematological toxicities.
In nongefitinib group, while 78% had received pemetrexed platinum doublet, remaining
patients received weekly paclitaxel, pemetrexed, docetaxel, and gemcitabine, thus
making the overall comparator (nongefitinib group) heterogeneous when compared to
second-line gefitinib. As physician choice was exercised to select patients and chemotherapeutic
agents, selection bias cannot be ruled out with PS 2–3 patients receiving single-agent
chemotherapy compared to pemetrexed doublet otherwise. After progression from second-line
agents, 46% patients also received third-line agents, mainly weekly paclitaxel. Hence,
impact of third line chemotherapy on OS cannot be completely nullified.
Conclusion
Among exclusive EGFR mutation-positive NSCLC, our study is the first to report outcomes
of gefitinib versus physician choice chemotherapy as second line among patients from
Indian subcontinent. Favorable response rate after second-line agents is one of the
strongest prognostic factors for overall outcome. TKIs after progression on first-line
chemotherapy improves response rates, PFS, and OS compared to chemotherapy in EGFR
mutation-positive NSCLC.
