Sydney Cancer Centre
Royal Prince Alfred Hospital
At a time when our understanding of the molecular biology of cancer was supposed to lead to major advances in therapy, it is disappointing to report that the major lung cancer ‘story’ of 2002 (table one) is likely to be the failure of the new molecularly-targeted agent Iressa (an inhibitor of Epidermal Growth Factor Receptor EGFR-signalling) to improve the survival of patients with metastatic non-small cell lung cancer (NSCLC) above that obtained with ‘modern’ chemotherapy. It is also perhaps a sign of this time that the release of this information1, in advance of any publication or presentation at a scientific meeting, appears to have been made because of the sensitivity of the result for the price of the sponsor’s shares and the need to avoid any hint of withholding price-sensitive data – in a year when corporate governance in general (and in biotechnology firms in particular) has been a major theme.
By the time you read this, the results of the two Iressa phase III trials will have been presented at the ESMO meeting in October. The design of the studies tells us a lot about where we are and where we (might) be going. These were big (>500 patients) international studies. Patients having first-line chemotherapy for stage IV or ‘wet’ IIIB NSCLC were randomised to standard chemotherapy or standard chemotherapy plus Iressa. Iressa was given continuously and there was a matching placebo. The trials were double blind. Survival was the primary endpoint, but symptom control and quality of life were also assessed.
What was considered ‘standard chemotherapy’? The two Iressa phase III trials had carboplatin plus paclitaxel and cisplatin plus gemcitabine as their standard chemotherapy regimens – Australian centres were included in the trial where it was carboplatin and paclitaxel.
So what went wrong? Until we see all the final data it is impossible to satisfactorily answer that question. Perhaps it was something very mundane, like very poor compliance with the daily oral medication. But I would speculate that the problem was the failure to really select the patients where inhibition of EGFR signalling could be therapeutically useful. The Iressa trials did not select patients whose cancers overexpressed EGFR, but included all NSCLC. While over 80% of NSCLC are found to actually overexpress EGFR2, this does not necessarily mean that 80% of NSCLC will benefit from inhibiting this pathway. Advanced cancers, particularly epithelial cancers, are genetically very heterogeneous, and dependence on a single aberrant pathway for cell survival is likely to be the exception rather than the rule. We cannot expect the same paradigm in NCSLC as in cancers like GISTs where there is a very narrow spectrum of genetic changes and one clear causative activating mutation3. To make progress in molecular therapy in NSCLC we will need to understand not only the range of genetic changes that can occur, but what is actually occurring and important for an individual patient. Individual tumour profiling will be costly and complex, but perhaps necessary.
Over the past 10 years we have seen the introduction of many new cytotoxics with some activity as single agents in advanced NSCLC. This resulted in the potential for multiple combinations with cisplatin, with carboplatin, with other older agents like ifosfamide, and among the new cytotoxics themselves. All the two-drug and many of the three-drug combinations have now been tested in phase III trials, and in 2002 we can at last make some overall conclusions about the ‘best’ ones. There is no evidence for any clinically important efficacy differences between the five regimens suggested in table two, and any one can be considered a suitable standard of care for metastatic NSCLC patients with good performance status. Differences in the spectrum of side effects or in the number of treatment visits per cycle may lead oncologists to favour one over another for individual patients.
Three presentations at ASCO 2002 reported phase III trials where paclitaxel, gemcitabine or docetaxel alone was compared to that drug plus a platinum. The message from all these trials was the same – the two-drug combination was better than the single agent. Most important was the trial comparing paclitaxel plus carboplatin to paclitaxel4. Five hundred and sixty one patients were entered and randomised to paclitaxel/carboplatin or paclitaxel. Response rate, time to progression, and median survival were all significantly better with paclitaxel/carboplatin. Although laboratory haematological toxicity was worse with combination, there was no difference in clinically important haematological toxicity, non-haematological toxicity or quality of life.
Subgroup analyses of elderly patients (age ≥ 70) and performance status two patients gave the same conclusions. In patients belonging to either of these important patient groups, cisplatin-based regimens are unacceptably toxic and performance status two patients were not included in the major phase III trial evaluating carboplatin and docetaxel. For them, either paclitaxel/carboplatin or gemcitabine/carboplatin is therefore my recommendation. There will still be a ‘borderline’ group of patients, with performance status two and elderly with some comorbidity in whom supportive care or single agent chemotherapy will be appropriate.
What about adding a third cytotoxic agent? While all conceivable triplets have not been tested, it is now clear that there is unlikely to be a significant benefit from doing so. Additional toxicity and (importantly) additional cost means that three-drug combinations must remain experimental. Similarly, there is no current evidence to support using a two-drug, non-platinum combination. The impetus to do so has been the hypothesis that (particularly cisplatin) combinations are more likely to impair patients’ quality of life than non-platinum ones. At ASCO 2002, an important European/Canadian trial was reported where quality of life was the primary endpoint5. Patients (n=502) received vinorelbine plus gemcitabine or either drug plus cisplatin. There was no difference in quality of life. Tumour response rates and time to progression were better with cisplatin, resulting in more tumour symptoms balancing out the less toxicity in the vinorelbine/gemcitabine arm. The financial costs to the health care system of chemotherapy for metastatic NSCLC patients have been debated before – we must therefore await clear evidence to use non-platinum combinations.
Finally, for chemotherapy of metastatic NSCLC, 2002 saw the presentation of the results of the UK Big Lung Trial (BLT)6. The BLT should be the last of the trials where ‘best supportive care’ is an acceptable treatment option for good performance status NSCLC patients. Seven hundred and twenty five patients were randomised in the UK to chemotherapy or supportive care, 80% of the chemotherapy patients received an old three-drug combination of cisplatin plus mitomycin C plus either ifosfamide or vinblastine. Reflecting the unusual nature of NSCLC practice in the UK, less than 40% of the patients entered on the trial actually had stage IV NSCLC; 7% had stage I or II disease! (These figures don’t include the companion trial where patients with localised disease could be randomised to chemotherapy followed by radiation therapy or radiation therapy alone.) Survival was significantly better with chemotherapy, and the results look pretty identical to the 1995 meta-analysis. Chemotherapy did not produce any detrimental effect on patients’ quality of life.
It hasn’t been a good year for small cell lung cancer. Recent survival gains in this disease have come from defining the optimum use and timing of radiation therapy for limited disease patients7,8. Apart from the Japanese trial (published this year but really old news from 2001) showing benefit from irinotecan9, there are no positive results to report. A large US trial (587 patients) compared standard platinum/etoposide chemotherapy to the same chemotherapy plus paclitaxel for extensive disease patients10. Not only was survival not improved by adding paclitaxel, there was a significant increase in toxic deaths in the three-drug arm despite the routine use of G-CSF.
Which leads us back to Iressa. Despite the failure of this drug to improve on the results of first-line chemotherapy, there has been clear evidence presented in 2002 that when given alone, it can induce significant tumour shrinkage and result in improvement in symptoms in patients with NSCLC who have previously been treated with chemotherapy. Two large but uncontrolled multinational phase II studies were done (there was randomisation in both studies to different doses of Iressa but not to a control arm)11,12. The combined total of patients was 416. The response rates in terms of tumour shrinkage were between 10% and 20% of patients (higher in patients treated in Japan than elsewhere), while 30%-40% of patients reported improved symptoms and quality of life. These results are clearly not due to chance or spontaneous remissions of disease, but right now it is unknown how regulatory agencies will view this uncontrolled data if (when) it is submitted to them. My bet is that in Australia there will be difficulty obtaining registration and subsidised reimbursement without a controlled trial.
Table two gives my perspective on current standard care and what new systemic agents might lead to future advances. All the approaches in table two are in phase III trials. Metastatic NSCLC continues to be a priority for companies developing new agents, because of the very high incidence of the disease and the paradigm from other cancers that agents useful in metastatic disease may increase cure rates in earlier stages. We will see results from many more trials of standard chemotherapy versus standard chemotherapy plus molecular/biologic agent announced in the next few years. For the smaller group of patients, with good performance status and locally advanced disease, concurrent platinum chemotherapy and radiation is unlikely to be displaced but we are less clear on exactly what drug combination to use concurrently to maximise efficacy and minimise in-field toxicity. Specific radiation enhancers (and protectors) may provide further therapeutic gain in this groups of patients, as may further refinements in radiation planning and delivery systems. For operable patients, neoadjuvant chemotherapy still lacks definitive proof of benefit, but results of trials where the chemotherapy used is one of, or close to, the current standards for metastatic disease will provide more substantive data. We will also expect to see results from more trials of adjuvant chemotherapy following surgery, and within two to three years a new meta-analysis of surgery ± adjuvant chemotherapy.
One hesitates to even speculate on where any advance in systemic therapy of SCLC will come from. An international phase III trial is currently open to (hopefully) validate the result of the Japanese irinotecan trial. If so, we will have a new two-drug standard for extensive disease patients and a strong lead for limited disease patients. Adding a third cytotoxic agent (paclitaxel, topotecan) as part of initial therapy or after response has not proved fruitful. SCLC is truly a malignancy where we need a new approach, and great hope rests on the large international trial of a ganglioside-based vaccine for limited disease patients who have completed combined modality therapy. If this shows benefit, then we have reason to look with much more interest at immunological approaches to this disease. If not, then perhaps the only good news will be that the incidence of SCLC seems to be falling!
At least it was a good year for mesothelioma.
Michael Millward has received research support from, acted as a consultant for, or received travel grants or honoraria from the following companies whose products are discussed in this article: Aventis, Eli-Lilly, Pharmacia, Bristol-Myers Squibb, Merck, and Astra-Zeneca.
4. RC Lilenbaum, J Herndon, M List, et al. “Single agent versus combination chemotherapy in advanced non-small cell lung cancer: A CALGB randomized trial of efficacy, quality of life, and cost-effectiveness.” Proc Am Soc Clin Oncol, 21 (2002): 1a.
5. C Gridelli, F Shepherd, F Perrone, et al. “Gemvin III: A Phase III study of gemcitabine plus vinorelbine compared to cisplatin plus vinorelbine or gemcitabine chemotherapy for stage IIIb or IV non-small cell lung cancer: an Italo-Canadian study.” Proc Am Soc Clin Oncol, 21 (2002): 292a.
6. RJ Stephens, D Fairlamb, N Gower, et al. “The Big Lung Trial (BLT): Determining the value of cisplatin-based chemotherapy for all patients with non-small cell lung cancer. Preliminary results in the supportive care setting.” Proc Am Soc Clin Oncol, 21 (2002): 291a.
7. M Takada, M Fukuoka, M Kawahara, et al. “Phase III study of concurrent verus sequential thoracic radiotherapy in combination with cisplatin and etoposide for limited-stage small cell lung cancer: Results of the Japan Clinical Oncology Group study 9104.” J Clin Oncol, 20 (2002) :3054-60.
10. HB Niell, JE Herndon, AA Miller, et al. “Randomized Phase III intergroup trial (CALGB 9732) of etoposide and cisplatin with or without paclitaxel and G-CSF in patients with extensive stage small cell lung cancer.” Proc Am Soc Clin Oncol, 21 (2002): 293a.
11. M Fukuoka, S Yano, G Giaccone, et al. “Final results from a Phase II trial of ZD1839 (Iressa) for patients with advanced non-small cell lung cancer (IDEAL 1).” Proc Am Soc Clin Oncol, 21 (2002): 298a.
12. MG Kris, RB Natale, RS Herbst, et al. “A Phase II trial of ZD1839 (Iressa) in advanced non-small cell lung cancer patients who had failed platinum- and docetaxel-based regimens (IDEAL 2).” Proc Am Soc Clin Oncol, 21 (2002): 292a.