Role of Adjuvant Therapy in Resected Stage II/IIIA Non-Small-Cell Lung Cancer
Role of Adjuvant Therapy in Resected Stage II/IIIA Non-Small-Cell Lung Cancer
The first part of this
article began an exploration of adjuvant therapy for completely resected
(margin-negative) stage II/IIIA (node-positive) non-small-cell lung cancer (NSCLC).
Using an evidence-based approach, the review sorted though key studies on the
role of adjuvant treatment, focusing primarily on randomized trials of
postoperative radiotherapy and postoperative chemotherapy, as well as two large
meta-analysesthe postoperative radiotherapy (PORT) meta-analysis and the
NSCLC Cooperative Group meta-analysis. The second part of this article will
describe several studies evaluating the role of adjuvant chemoradiation, and
will address currently active phase III trials of adjuvant therapy in this
setting. Part 2 will conclude with a discussion of future strategies for the
adjuvant treatment of NSCLC.
Only a few studies have assessed the role of adjuvant chemoradiation in the
phase III setting.
Lung Cancer Study Group Trial
The Lung Cancer Study Group (LCSG)-791 trial compared split-course
radiotherapy (2,000 cGy in five fractions × 2 separated by 3 weeks) and
concurrent CAP chemotherapy (cyclophosphamide, doxorubicin [Adriamycin],
cisplatin) to radiotherapy alone. This study focused on patients with an
incomplete resection, defined as either tumor in the highest resected
mediastinal node or the presence of a positive surgical margin. Most patients (N
= 151) had stage III disease (119 had pN2 disease), 10 had stage II, and 3 had
stage I. As in the previous LCSG study, CAP was administered every 4 weeks
for 6 months (with the first two cycles administered on day 1 of each
The chemoradiation group experienced a nonsignificant improvement in median
survival (20 vs 13 months) but a significant improvement in median time to
recurrence (14 vs 8 months, P = .004). Analysis of failure patterns showed a
significant decrease in distant metastases in the chemotherapy arm (P = .01).
Subset analysis demonstrated a slight advantage for patients with pN2 disease
who received both radiation and chemotherapy. However, with further follow-up,
the survival curves converged at about 2.5 years.
Memorial Sloan-Kettering Trial
A study at Memorial Sloan-Kettering Cancer Center randomized 72 patients with
pathologic T1-3, N2 (stage IIIA) NSCLC to postoperative radiotherapy vs
radiotherapy and chemotherapy with vindesine and cisplatin. No significant
differences were found in overall survival (16.3 months for radiotherapy
plus chemotherapy vs 19.1 months for radiotherapy alone, P = .42) or
time to progression (9.2 vs 9.0 months, respectively, P = .35).
In a trial from France, 267 patients with resected NSCLC were randomized
to postoperative radiotherapy alone (6,000 cGy in 6 weeks) vs three cycles of
chemotherapy (cycles 1 and 3: doxorubicin, vincristine, cisplatin, lomustine;
cycle 2: vincristine, cisplatin, cyclophosphamide) followed by radiotherapy.
Most patients (N = 189) had stage III disease, 70 had stage II, and 8 had stage
There were no significant differences in overall survival or disease-free
survival. Although the analysis was not stratified by stage, among the 137
patients who were pN2, a significant improvement in both survival and
disease-free survival was observed in the chemoradiation group (P = .003 and
.002, respectively), apparently due to a decreased incidence of metastases (P = .003).
Intergroup Trial 0115
Perhaps the biggest setback to the use of adjuvant chemoradiation came with
the results of Intergroup Trial 0115 (E-3590). In this large randomized
trial, 488 patients who had undergone complete resection of stage II and IIIA
NSCLC, as well as a thorough mediastinal lymph node sampling or dissection, were
randomized to receive either four cycles of monthly cisplatin (60 mg/m² on day
1) and etoposide (120 mg/m² on days 1 to 3) administered concurrently with
thoracic radiotherapy vs radiotherapy alone. Radiotherapy on both arms consisted
of standard fractions (180 cGy) to a total dose of 5,040 cGy. An additional
1,080 cGy in six fractions was administered to nodal levels in which
extracapsular extension was documented.
With a median follow-up of 44 months, the median survival was 39 months in
the radiotherapy arm vs 38 months in the combined-modality arm (P = .56). An
important finding on multivariate analysis was that survival was significantly
influenced by extent of lymph node involvement (multiple vs single), type of
lymph node dissection (sampling vs complete), age (60 or more vs < 60), and
histology (other vs squamous). The authors concluded that postoperative
chemoradiotherapy with cisplatin and etoposide does not prolong survival in
patients with completely resected stage II/IIIA NSCLC compared to postoperative
Potential caveats in interpreting this negative trial relate to the
relatively high rate of ineligible patients (23%) and the relatively low rate of
compliance with all the planned chemotherapy (69%). It is interesting to note
that patients without evidence of K-ras mutations appeared to benefit from the
addition of chemotherapy to radiotherapy (median survival: 42 vs 25 months), P =
.066 on multivariate analysis. This suggests that the underlying problem may
not be the lack of effective adjuvant therapies, but rather the difficulty in
properly selecting patients who truly benefit from these treatments.
Yet another recent randomized trial further corroborates the lack of benefit
of chemotherapy in addition to radiotherapy in the adjuvant setting. In this
study, patients with resected pN2 disease were randomized to either radiotherapy
alone (50 Gy in 5 weeks) or chemotherapy with cisplatin, 75 mg/m² on day 1, plus
ifosfamide (Ifex), 1.5 g/m² on days 1 to 4, for three cycles every 4 weeks
followed by the same radiotherapy. An interim analysis showed a median overall
survival of 34 months and a 3-year survival rate of 46%, with no differences
between the groups (P = .7).
NSCLC Cooperative Group Meta-analysis
Seven trials (807 patients) included in an NSCLC Cooperative Group
meta-analysis randomized patients with resected NSCLC to surgery plus
radiotherapy vs surgery plus radiotherapy plus chemotherapy. Six of these trials
used a cisplatin-based regimen. The total planned doses of radiation therapy
ranged from 4,000 cGy in 10 fractions to 6,500 cGy in 33 fractions.
The meta-analysis demonstrated a hazard ratio of 0.94, or a 6% reduction in the
risk of death, favoring the chemotherapy arm, which translated into a 2%
absolute benefit at 5 years but was not statistically significant.
Summary of Evidence-Based Data
The role of adjuvant therapy following a complete resection for stage II/IIIA
NSCLC remains controversial. To date, there is no convincing evidence that any
therapy consistently improves survival in the adjuvant setting. Thus,
observation remains a viable alternative for such patients. Postoperative
radiotherapy has been associated with a significant improvement in local
control, particularly in patients with pathologic N2 disease. Several studies
suggest that the pN2 subset may also have an improvement in recurrence-free
survival. Patients with pathologic N0 disease (with negative margins) have a low
risk of locoregional recurrence and, therefore, should not be offered
The benefit-risk ratio of adjuvant radiotherapy in the setting of pathologic
N1 disease appears to be intermediate. Other factors may help in making
treatment recommendations in this situation. For example, Yano et al
reported a 5-year survival of 40% for patients with hilar N1 disease, compared
to 65% for those with lobar N1 disease (P = .014). As the survival for patients
with lobar N1 disease approaches that of patients with pN0 disease, the benefit
of postoperative radiotherapy is likely small in this group of patients. In the
future, detection of lymph node micrometastases using immunohistochemical or
molecular methods may be instrumental in further selecting which patients will
benefit most from adjuvant therapies.
If postoperative radiotherapy is administered, it should be done in the
context of CT-based treatment planning with careful attention to the
radiotherapy technique, including dose, fractionation, and volume. The most
commonly used regimen in this country is 5,000 to 5,040 cGy in 25 to 28
fractions. The use of posterior spinal cord shields must be avoided. Special
care should be taken to limit the amount of normal lung radiation, particularly
on the contralateral side. Thus, lateral radiation fields should not be
utilized. Similarly, the dose and volume of heart radiation should be minimized.
Only patients with adequate performance status and pulmonary function
(generally FEV1 > 1 liter) after surgery should be considered for
postoperative radiotherapy. The role of adjuvant chemotherapy (with or without
radiation therapy) is less clear. Chemotherapy, at this time, should be offered
to patients in appropriate clinical trials.