From Wiki Journal Club
Jump to: navigation, search
Arriagada R, et al. "Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer". The New England Journal of Medicine. 2004. 350(4):351-360.
PubMedFull textPDF

Clinical Question

Do patients with completely resected non-small cell lung cancer have improved survival with cisplatin-based adjuvant chemotherapy compared to observation alone?

Bottom Line

Adjuvant cisplatin-based chemotherapy improves overall survival at five years in patients with complete surgical resection of stage 1-3 non-small cell lung cancer.

Major Points

Surgery, chemotherapy, and radiation therapy all play a role in the treatment of patients with non-small-cell lung cancer (NSCLC). Particularly for patients with stage 1 and 2 disease, surgical resection offers the best chance at a cure. Nevertheless relapse occurs in a substantial portion of patients following complete resection, and postoperative chemotherapy may play a role in eradicating micrometastatic disease. Published in 2004, the International Adjuvant Lung Cancer Trial (IALT) was the first large randomized clinical trial to demonstrate that adjuvant cisplatin-based chemotherapy, after complete surgical resection, improved 5-year overall survival compared to observation alone. The trial randomized 1,867 patients with completely resected stage 1-3 NSCLC to either postoperative chemotherapy or observation alone. Participating centers chose the adjuvant therapy provided, and about half of patients received 3-4 cycles of cisplatin plus etoposide. Most patients had undergone pneumonectomy or lobectomy as their cancer surgery, and most (90%) had WHO performance status 0-1. Pathologic stage was 1 in about 35%, stage 2 in 25%, and stage 3 in the remaining 40%. With a median follow-up of 56 months, the primary outcome of overall survival favored adjuvant chemotherapy (HR 0.86; 95% CI 0.76-0.98; P=<0.03). As a result of IALT and similar studies, adjuvant chemotherapy became the standard of care for patients with resected stage 1-3 disease.

Of interest, accrual in this trial that administered chemotherapy after surgery (i.e., adjuvant) slowed towards the latter years of the trial. This was attributed to more patients electing to undergo preoperative (i.e., neoadjuvant) chemotherapy and therefore forgoing enrollment in adjuvant trials. This trend in physician and patient preference followed observations that preoperative chemotherapy may be better tolerated than postoperative chemotherapy, while yielding similar survival benefits. For example, NATCH (2010)[1] demonstrated greater completion rates for patients receiving preoperative than those receiving postoperative chemotherapy (90% vs. 60%).


NCCN Guidelines on Non-Small Cell Lung Cancer (version 4.2019, adapted)[2]

  • Observation is recommended for patients with completely resected stage 1A tumors.
  • Postoperative chemotherapy is recommended for patients with resected T2ab,N0 tumors and negative surgical margins who have high-risk features such as poorly differentiated tumors, vascular invasion, tumor >4 cm, visceral pleural involvement, and unknown lymph node status. (Category 2A)
  • Postoperative chemotherapy (with or without concurrent RT, depending on circumstances) is recommended for patients with resected stage 2B and 3 disease.
  • When postoperative chemotherapy is administered, cisplatin is recommended in combination with either docetaxel, etoposide, gemcitabine, or vinorelbine for all histologies; pemetrexed is also an option for nonsquamous NSCLC.
    • If cisplatin cannot be tolerated, carboplatin may be used in combination with pemetrexed (for nonsquamous), paclitaxel, or gemcitabine.


  • Multicenter, non-blinded, non-placebo-controlled, randomized trial
  • N=1,867 patients with resected stage 1-3 NSCLC
    • Cisplatin-based chemotherapy (n=932)
    • Observation (n=935)
  • Setting: 148 centers in 33 countries
  • Enrollment: 1995-2000
  • Median follow-up: 56 months
  • Analysis: Intention-to-treat
  • Primary outcome: Overall survival
  • Secondary outcomes: Disease-free survival and second primary cancers


Inclusion Criteria

  • Age 18-75 years
  • Pathological stage 1-3 NSCLC
  • Complete surgical resection

Exclusion Criteria

  • Prior chemotherapy or radiotherapy
  • Contraindication to chemotherapy
  • No prior cancer, aside from nonmelanoma skin cancer or cervical carcinoma in situ

Baseline Characteristics

From cisplatin group (similar to control).

  • Demographics: Age 59 years, 19.3% female
  • Type of cancer: Squamous (46%), adenocarcinoma (41%), large-cell carcinoma (6%), mixed (4%)
  • TNM pathological stage: 1 (36%), 2 (25%), 3 (40%)
  • WHO performance status: 0 (54%), 1 (38%), 2 (7%)
  • Type of surgery: Pneumonectomy (35%), Lobectomy (64%), Segmentectomy (1%)


  • Randomization to postoperative chemotherapy versus observation
    • Randomization was stratified by center, surgery type, and pathologic stage
    • Each center individually determined their policy for dose of cisplatin and choice/dose of second agent for each stage of NSCLC
      • Drug combined with cisplatin: etoposide (56.5%), vinorelbine (26.7%), vinblastine (10.9%), vindesine (5.8%)
    • Experimental group received cisplatin-based adjuvant chemotherapy as pre-planned by each center within 60 days of surgery
    • Control group received observation only (no placebo given)
  • All patients were to complete a one-page follow-up form on adverse events at 6 months and annually


Comparisons are postoperative chemotherapy vs. observation.

Primary Outcomes

Overall survival
5 years: 44.5% vs. 40.4% (HR 0.86; 95% CI 0.76-0.98; P<0.03)
2 years: 70.3% vs. 66.7%

Secondary Outcomes

Disease-free survival
5 years: 39.4% vs. 34.3% (P<0.003)
2 years: 61.0% vs. 55.5%

Subgroup Analysis

The intervention exhibited a nonsignificant trends towards mortality benefit across a multitude of subgroup analyses including age, gender, TNM stage, histologic type, dose of cisplatin, and choice of second agent. Three subgroups with a nonsignificant trend toward harm includes worse performance status, higher dose of cisplatin (120 mg/m2), and use of vindesine as a second agent.

Mortality by TNM stage
Stage 1: 34.5% vs. 35% (nonsignificant)
Stage 2: 53.4% vs. 56.7% (nonsignificant)
Stage 3: 62.6% vs 70.1% (P<0.05)

In the subgroup analysis, the P for interaction between stage and survival was 0.41.

Adverse Events

Lethal toxic effects
7 patients (0.8%) died due to chemotherapy-related adverse events: bone marrow aplasia (5), renal failure (1), hyponatremia (1).
2.4% of patients receiving 120mg/m2 of cisplatin vs. 0.6% for 100 mg/m2 or less (P=0.15)
Grade 4 toxic effects (22.6%)
Neutropenia (17.5%), thrombocytopenia (2.6%), vomiting (3.3%)


  • Study was underpowered to detect/reject survival benefit in subgroups, so magnitude of benefit for stage 1-2 patients is less certain.
  • No data was reported on grade 2-3 adverse events which would inform the risk/benefit ratio and informed consent discussion with patients.
  • Choice of optimal adjuvant chemotherapy regimen remains unclear given the design of this trial alone.
  • No placebo-control, concealment, or blinding were performed, introducing multiple sources of bias.
  • Second primary cancers were not reported despite being a secondary outcome that was assessed.


This was a large, multi-center study spanning multiple nations and institutions with a wide array of non-corporate funding sources. Conflicts of interest appear to be inevitable but minimal.

Further Reading

  1. Felip E et al. Preoperative chemotherapy plus surgery versus surgery plus adjuvant chemotherapy versus surgery alone in early-stage non-small-cell lung cancer. J. Clin. Oncol. 2010. 28:3138-45.
  2. NCCN Guidelines on Non-Small Cell Lung Cancer, 4.2019