`
`(.';:\ Taylor&Francis
`'::::::::::;J T~ylor& Fr~nrnGroup
`
`ISSN: 0284-186X (Print) 1651-226X (Online) Journal homepage: http://www.tandfonline.com/loi/ionc20
`
`A Systematic Overview of Chemotherapy Effects in
`Non-Small Cell Lung Cancer
`
`Sverre Sorenson, Bengt Glimelius, Peter Nygren
`
`To cite this article: Sverre Sorenson, Bengt Glimelius, Peter Nygren (2001) A Systematic
`Overview of Chemotherapy Effects in Non-Small Cell Lung Cancer, Acta Oncologica, 40:2-3,
`327-339
`
`To link to this article: http://dx.doi.org/10.1080/02841860120014
`
`-
`
`Published on line: 08 Jul 2009.
`
`~~ Submit your article to this journal C?
`
`W!l Article views: 153
`
`!) View related articles C?
`
`Full Terms & Conditions of access and use can be found at
`http://www.tandfonline.com/action/journallnformation?journa1Code=ionc20
`
`Download by: [38.98.152.163]
`
`Date: 23 November 2015, At: 06:54
`
`Sandoz Inc.
`Exhibit 1042-0001
`
`
`
`SYSTEMATIC OVERVIEW ARTICLE
`
`A Systematic Overview of Chemotherapy Effects 1n
`Non-Small Cell Lung Cancer
`Sverre Sorenson, Bengt Glimelius and Peter Nygren for the SBU-group 1
`
`From the Department of Thoracic Medicine, Haukeland University Hospital, Bergen, Norway (S. Sorenson),
`and Department of Oncology, Radiology and Clinical Immunology, University Hospital, Uppsala, Sweden
`(B. Glimelius, P. Nygren)
`
`Correspondence to: Bengt Glimelius, Department of Oncology, Radiology and Clinical Immunology, Section of
`Oncology, University Hospital, SE-751 85 Uppsala, Sweden. Tel: + 46 18 611 55 13. Fax: + 46 18 611 55 28.
`E-mail: bengt.glimelius@onkologi.uu.se
`
`Acta Oncologica Vol. 40, No. 2/3, pp. 327-339, 2001
`
`•
`
`00
`0\
`00
`~
`
`A systematic review of chemotherapy trials in several tumour types was performed by The Swedish Council of Technology Assessment
`in Health Care (SBU). The procedures for the evaluation of the scientific literature are described separately (Acta Oncol 2001; 40:
`155-65). This overview of the literature on chemotherapy for non-small cell lung cancer (NSCLC) is based on 53 scientific publications
`including six meta-analyses based on 65 prospective randomised trials comprising 15 607 patients and an additional 32 prospective
`randomised studies including 8 902 patients. The conclusions reached can be summarised into the following points:
`•
`In stage IIlB-IV disease, published data demonstrate that cisplatin-based chemotherapy confers a modest, median 1.5-3 months,
`prolongation of survival. The closely related compound carboplatin seems to provide similar effects. Randomised studies indicate
`symptomatic relief and improvement of indices of quality of life (QoL) for patients who receive platinum-based combination
`chemotherapy or single drug therapy with more recent compounds. Data supporting the use of chemotherapy are not available for
`patients in poor general condition (WHO performance status 3-4) and evidence is limited for elderly patients (above 70- 75 years).
`Platinum-based chemotherapy can be recommended for selective use in routine care of advanced NSCLC although patients should be
`encouraged to participate in controlled clinical trials to further elucidate the role of chemotherapy in advanced disease.
`In advanced disease, recent data suggest that the newer agents gemcitabine, paclitaxel, irinotecan and vinorelbine, in combination with
`cisplatin, provide an additional survival benefit compared with earlier cisplatin-based regimens. Furthermore, paclitaxel, docetaxel and
`vinorelbine as single agents seemingly provide a survival benefit over supportive care alone comparable to that of older cisplatin-based
`combinations.
`• A standard regimen for advanced disease cannot yet be defined. Until more data are at hand, it is recommended to be platinum-based
`and preferably combined with one of the newer agents.
`• At progression after platinum-based chemotherapy for advanced disease, limited data indicate a small survival benefit from docetaxel
`over supportive care alone. Such second-line chemotherapy of advanced disease can be recommended for selected patients but should
`preferably be confined to controlled clinical trials.
`In stage III disease, published data show that induction cisplatin-based chemotherapy before radical radiotherapy modestly prolongs
`long-term survival and lowers the incidence of distant metastases compared with radiotherapy alone. Furthermore, published data
`show that concurrent chemo- and radiotherapy with cisplatin or carboplatin may enhance local control and long-term survival.
`Chemotherapy in this setting can be recommended for selected patients but treatment should preferably be given within a controlled
`clinical trial.
`In stage IIIAN2 disease, data from pilot studies demonstrate that surgery after induction chemotherapy is feasible. Pathologically
`complete remissions have been confirmed in 10-20% of treated patients. Two small randomised studies demonstrate a significant
`survival advantage for induction chemotherapy followed by surgery compared with surgery alone. Induction chemotherapy can be
`recommended for selected patients but treatment should preferably be given within a controlled clinical trial. The superiority of
`induction chemotherapy plus surgery compared with combined chemotherapy and radical irradiation has not been proven in a
`randomised trial but currently such studies are under way.
`In the adjuvant setting, published data suggest that cisplatin-based chemotherapy after radical surgery may increase five-year survival
`from around 50% by a further 5% but the confidence interval for this estimate is too wide for firm conclusions. Large-scale prospective
`randomised trials are under way to resolve this important issue and adjuvant chemotherapy is, thus, not recommended for routine
`treatment.
`
`•
`
`•
`
`•
`
`1 Other members of the SEU-group were: Jonas Bergh, Radiumhemmet, Stockholm; Lars Brandt, Dept of Oncology, University Hospital,
`Lund; Bengt Brorsson, SBU, Stockholm; Barbro Gunnars, Dept of Oncology, University Hospital, Lund; Larsolof Hafstrom, Dept of
`Surgery, University Hospital, Umeii; Ulf Haglund, Dept of Surgery, University Hospital, Uppsala; Thomas Hogberg, Dept of
`Gynaecological Oncology, University Hospital, Linkoping; Karl-Gunnar Janunger, Dept of Surgery, University Hospital, Umeii;
`Per-Ebbe fonsson, Dept of Surgery, Helsingborgs lasarett, Helsingborg; Goran Karlsson, Handelshogskolan, Stockholm; Eva Kimby,
`Dept of Haematology, University Hospital, Huddinge; Gunilla Lamnevik, SBU, Stockholm; Sten Nilsson, Radiumhemmet, Stockholm;
`Johan Permert, Dept of Surgery, University Hospital, Huddinge; Peter Ragnhammar, Radiumhemmet, Stockholm.
`
`© Taylor & Francis 2001. ISSN 0284-186X
`
`Acta Oncologica
`
`Sandoz Inc.
`Exhibit 1042-0002
`
`
`
`328
`
`S. Sorenson et al.
`
`Acta Oncologica 40 (2001)
`
`In a global perspective, lung cancer is a major cause of
`cancer-related death. In Sweden, about 2 500 new cases of
`lung cancer are diagnosed clinically every year, with an
`additional 300 cases detected at autopsy. One-third of the
`patients are women and the incidence in females is steadily
`rising. In males, overall incidence has levelled off since
`about 1975. A more detailed analysis shows a statistically
`significant decline among men in larger Swedish cities, but
`not in the country as a whole (!). 80-90% of all lung
`cancer is caused by smoking alone or in combination with
`other factors.
`The major histologic types of lung cancer according to
`the WHO classification, which was revised in 1999, are
`squamous cell carcinoma, small cell carcinoma, adenocar(cid:173)
`cinoma, large cell carcinoma, and adeno-squamous car(cid:173)
`cinoma. While small cell carcinoma exhibits specific
`characteristics with respect to clinical features and treat(cid:173)
`ment principles, the other histologic types are frequently
`grouped together as non-small cell lung cancer (NSCLC)
`for practical purposes. In contrast to SCLC, which shows
`indisputable short-term sensitivity to antineoplastic drugs,
`NSCLC displays low or moderate sensitivity to chemother(cid:173)
`apy and the clinical relevance of such treatment has been
`questioned. Since NSCLC accounts for about 80% of all
`lung cancer, acceptance of a role for chemotherapy in this
`tumour type would have considerable impact on the econ(cid:173)
`omy and organisation of care for lung cancer patients.
`Staging of lung cancer is performed by use of the TNM
`system (Table I). A modification of stage grouping accord(cid:173)
`ing to TNM, based on data from a heterogeneous set of
`5 319 patients diagnosed and/or treated at the University
`of Texas M. D. Anderson Cancer Center between 197 5
`and 1988, was published in 1997 (2). In the absence of
`impaired lung function or other medical contraindications
`to surgical therapy, patients with stages I and II disease
`are usually offered radical resection. Stage lllA is a hetero(cid:173)
`geneous subset where some patients may be suitable for
`surgery. In stage lllB, tumour growth is localised but too
`advanced for resection. In stage IV, distant metastases are
`confirmed. Non-surgical treatment, which may be consid(cid:173)
`ered in patients with early stage disease who cannot be
`operated for medical reasons and in patients with stage
`Ill- IV disease, offers at best very limited prospects for
`long-term survival.
`Without treatment, the prognosis of NSCLC is poor.
`Based on follow-up of 130 patients who did not receive
`any specific anticancer therapy, median survival in stage
`lB was 17 months and all patients in this group were dead
`within three years from diagnosis. Patients with more
`advanced tumour stage (stages II-IV) had a median sur(cid:173)
`vival of eight months and all were deceased at 2.5 years
`(3).
`After a basic evaluation, about 20% of lung cancer
`patients can be considered for surgery. In this group,
`
`five-year survival is in the order of 40-50% (4, 5). The vast
`majority of deaths during the first few years are due to
`lung cancer. In non-resected patients, five-year survival
`amounts to about I%. The resection rate is, however,
`lower in some countries, and in the EUROCARE study,
`which reported 167 068 cases of lung cancer from 30
`cancer registries in 11 countries between 1978 and 1985,
`relative five-year survival varied between 6 and 14% (6).
`These differences are too large to be explained solely by
`non-uniform methods of registration and implicate subop(cid:173)
`timal standards for lung cancer detection and care in some
`parts of Europe. However, in all countries more than 50%
`of patients present with late stage disease (stage Ill/IV).
`The role of chemotherapy in NSCLC has been the
`subject of considerable debate. This issue, as well as a
`number of other topics related to treatment of unresectable
`NSCLC, was reviewed in clinical practice guidelines for(cid:173)
`warded by the American Society of Clinical Oncology in
`May, 1997. The conclusions were derived from an exten(cid:173)
`sive review of published literature through April 1997,
`performed by a multidisciplinary panel (7).
`It should be noted that response rate, which is a com(cid:173)
`monly used endpoint in studies of cancer chemotherapy,
`should be interpreted with caution in NSCLC. One major
`reason is the lack of consistent relationships between re(cid:173)
`sponse rate and survival in randomised trials. In a study
`by the Eastern Cooperative Oncology Group (ECOG), 743
`patients with stage IV NSCLC were randomised to receive
`one of five different regimens (8). When results for patients
`on a given regimen were compared with survival for all
`other patients, after adjustment for prognostic factors,
`carboplatin as a single drug was associated with the most
`favourable survival (median survival time, 32 weeks; p =
`0.008). Response rate with carboplatin was, however, only
`9%. In contrast, the combination of mitomycin, vin(cid:173)
`blastine, and cisplatin (MVP) showed a higher response
`rate of 20% but a trend for shorter survival (median
`survival time, 23 weeks; p = 0.09). Other examples of poor
`correlation between response rate and survival time in
`NSCLC may be found below.
`A second limitation of response rate as a tool for
`evaluation of chemotherapy in NSCLC is observer varia(cid:173)
`tion. Application of WHO criteria for definition of re(cid:173)
`sponse may be difficult in the presence of poorly defined
`boundaries of
`tumour and secondary
`inflammatory
`changes. Thirdly, it has been documented in studies of
`induction chemotherapy before surgery that the antineo(cid:173)
`plastic effect may be underrated by chest roentgenograms
`or CT examinations. Persistent radiographic changes do
`not exclude a pathological complete response. A reason(cid:173)
`able explanation is that NSCLC tumours are composed of
`neoplastic cells as well as variable amounts of stromal
`tissue and the stromal component will not be affected by
`
`00
`0\
`00
`~
`
`Sandoz Inc.
`Exhibit 1042-0003
`
`
`
`Acta Oncologica 40 (2001)
`
`Chemotherapy effects in non-small cell lung cancer
`
`329
`
`Table 1
`
`International TNM staging of lung cancer
`
`TNM descriptors
`
`TO
`Tis
`Tl
`
`T2
`
`Primary tumour (T)
`TX
`Primary tumour cannot be assessed, or tumour proven by the presence of malignant cells in sputum or bronchial
`washings but not visualized by imaging or bronchoscopy.
`No evidence of primary tumour.
`Carcinoma in situ.
`Tumour < 3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of
`invasion more proximal than the lobar bronchus' (i.e., not in the main bronchus).
`Tumour with any of the following features of size or extent: > 3 cm in greatest dimension; involves main bronchus,
`> 2 cm distal to the carina; invades the visceral pleura; associated with atelectasis or obstructive pneumonitis that
`extends to the hilar region but does not involve the entire lung.
`Tumour of any size that directly invades any of the following: chest wall (including superior sulcus tumours),
`diaphragm, mediastinal pleura, parietal pericardium; or tumour in the main bronchus <2 cm distal to the carina,
`but without involvement of the carina; or associated atelectasis or obstructive pneumonitis of the entire lung
`Tumour of any size that invades any of the following: mediastinum, heart, great vessels, trachea, esophagus,
`vertebral body, carina; or tumour with a malignant pleural or pericardia! effusion', or with satellite tumour
`nodule(s) within the ipsilateral primary-tumour lobe of the lung.
`
`T3
`
`T4
`
`Regional lymph nodes (N)
`NX Regional lymph nodes cannot be assessed.
`NO
`No regional lymph node metastasis.
`NI Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes, and intrapulmonary nodes involved by
`direct extension of the primary tumour.
`N2 Metastasis to ipsilateral mediastinal and/or subcarinal lymph nodes(s).
`N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular
`lymph node(s).
`
`Distant metastasis (M)
`MX Presence of distant metastasis cannot be assessed.
`MO No distant metastasis.
`Distant metastasis present. 3
`Ml
`
`Stage grouping - TNM subsets
`
`00
`0\
`00
`~
`
`Stage
`0
`IA
`IB
`IIA
`IIB
`
`TNM subset4
`Carcinoma in situ
`TINOMO
`T2NOMO
`TINIMO
`T2NIMO
`T3NOMO
`IIIA T3NIMO
`TIN2MO
`T2N2MO
`T3N2MO
`IIIB T4NOMO
`T4NIMO
`T4N2MO
`TIN3MO
`T2N3MO
`T3N3MO
`T4N3MO
`Any T Any N Ml
`
`IV
`
`'The uncommon superficial tumour of any size with its invasive component limited to the bronchial wall, which may extend
`proximal to the main bronchus, is also classified Tl.
`'Most pleural effusions associated with lung cancer are due to tumour. However, there are a few patients in whom multiple
`cytopathologic examinations of pleural fluid show no tumour. In these cases, the fluid is non-bloody and is not an exudate. When
`these elements and clinical judgement dictate that the effusion is not related to the tumour, the effusion should be excluded as a
`staging element and the patient's disease should be staged Tl, T2, or T3. Pericardia! effusion is classified according to the same
`rules.
`3 Separate metastatic tumour nodule(s) in the ipsilateral non primary-tumour lobe(s) of the lung also are classified Ml.
`4 Staging is not relevant for occult carcinoma, designated TXNOMO.
`
`Sandoz Inc.
`Exhibit 1042-0004
`
`
`
`330
`
`S. Sorenson et al.
`
`Acta Oncologica 40 (2001)
`
`non-surgical treatment. If irradiation is a treatment modal(cid:173)
`ity, postradiation fibrosis may further add to the difficulty
`of response evaluation by usual imaging methods.
`In this review, the role of chemotherapy in NSCLC will
`be evaluated in four different contexts: (I) advanced disease
`(stage lllB-IV), not amenable to surgical treatment or
`radical radiotherapy, (2) locally advanced disease (stage
`lllA-B) in combination with radiotherapy, (3) as induction
`therapy before surgery (neoadjuvant therapy, in particular
`for stage lllAN2), and (4) after radical surgery (stage I, II
`and subsets of stage lllA).
`The basis for the evaluation will be prospective ran(cid:173)
`domised trials and meta-analyses of such studies. With few
`exceptions, only studies fully published in peer-reviewed
`journals will be taken into account. Endpoints will be QoL
`and duration of survival in advanced disease and long-term
`survival in settings 2-4 above.
`
`ADVANCED DISEASE (STAGE lllB-lV)
`Role of cisplatin
`
`A major international collaborative meta-analysis of the
`role of chemotherapy in NSCLC was published in 1995 (5).
`Individual updated information on 9 387 patients in 52
`randomised trials was retrieved and the effect of chemother(cid:173)
`apy on survival was reported in four different settings:
`chemotherapy vs supportive care in advanced disease,
`chemotherapy +radiotherapy vs radiotherapy alone for
`locally advanced disease, radical surgery followed by
`chemotherapy vs surgery alone, and surgery+ postopera(cid:173)
`tive irradiation followed by chemotherapy vs surgery +
`postoperative irradiation alone. The role of chemotherapy
`as related to surgery+ postoperative irradiation will not
`receive major attention in this review. The principal reason
`is that postoperative radiotherapy has undergone critical
`re-evaluation in a recent meta-analysis (9).
`In advanced disease, data were available from 11 trials
`(1190 patients and I 144 deaths). Eight trials used cisplatin(cid:173)
`based chemotherapy and seven of these studies employed a
`combination of cisplatin with vinca alkaloids or etoposide.
`The results of trials using long-term alkylating agents
`suggested a detrimental effect of chemotherapy. However,
`cisplatin-based treatment compared with supportive care
`alone reduced the risk of death by 27% (p < 0.0001) and
`survival at one year was improved by 10% (95% confidence
`interval, Cl: 5 to 15%) from 15% to 25%. Median survival
`increased from 4 to 5.5 (95% Cl: 5 to 6.5) months. There
`was no evidence that any group of patients as defined by
`age, sex, histological type, performance status, or stage, had
`more benefit from chemotherapy.
`In another analysis of comparative randomised trials
`confined to patients with advanced disease, survival at six
`months was chosen as endpoint. Individual patient data
`were not analysed. There were eight trials comprising a total
`of 712 patients and all studies but one used cisplatin-based
`
`chemotherapy in the comparison with best supportive care
`(BSC). The criteria for selection of trials were slightly
`different from those used in the study cited above. Seven of
`the trials were included in both analyses. The estimated
`pooled odds ratio of death was 0.44 (95% Cl, 0.32-0. 59) for
`patients receiving chemotherapy compared with patients
`treated by BSC alone, corresponding to an increase of
`median survival from 3.9 to 6.7 months (10).
`A third analysis of randomised studies was performed by
`Souquet and others (11). Seven randomised studies compris(cid:173)
`ing a total of 706 patients were examined with respect to
`mortality at 3, 6, 9, 12 and 18 months. Individual patients
`were not taken into account. Studies on single-drug treat(cid:173)
`ment were excluded. Cisplatin was a component of
`chemotherapy in six trials. The seven studies form a subset
`of the 11 trials in the first meta-analysis. Mortality was
`significantly reduced during the first six months (three
`months; p < 0.001; six months; p < 0.00001) in patients
`receiving chemotherapy compared with BSC. The relative
`received combined
`risk of death for patients who
`chemotherapy was 0.65 at three months, 0.73 at six months,
`0.86 at nine months, 0.91 at 12 months, and 0. 96 at 18
`months.
`It is apparent from these studies that cisplatin-based
`chemotherapy confers a modest prolongation of survival in
`advanced NSCLC. The set of regimens is heterogeneous, as
`are entry criteria for various studies, and the optimal effect
`of chemotherapy on survival is probably underestimated. In
`addition, individual variability in response to treatment
`implies that gain of life may be more substantial in a
`proportion of cases. Questions related to optimal choice of
`drugs, doses, duration of chemotherapy as well as QoL and
`cost/benefit relationships are left unanswered by the meta(cid:173)
`analyses.
`The role of single-agent vs combination chemotherapy in
`advanced NSCLC was evaluated in an analysis by Marino
`and others (12). Nine studies comprising a total of 2199
`patients were reviewed and individual patient data were not
`taken into account. One-year survival was chosen as end(cid:173)
`point. The estimated pooled odds ratio of death for combi(cid:173)
`nation chemotherapy was 0.8 (95% confidence interval,
`0.6-1.0). It should be noted that the only two trials showing
`a significant difference in favour of combined therapy
`compared single agent vinca alkaloid therapy with vinca
`akaloid + cisplatin (13, 14). Since the role of cisplatin has
`already been demonstrated in the meta-analysis by the
`Non-small Cell Lung Cancer Collaborative Group, it is
`doubtful if the study by Marino provides further informa(cid:173)
`tion. With the advent of new active agents, further analyses
`of monotherapy vs combination chemotherapy in older
`trials are of limited interest.
`
`Role of carhoplatin
`
`Cisplatin has been used in the vast majority of platinum(cid:173)
`based treatment regimens. Comparable results may, how(cid:173)
`ever, be obtained with the analogue carboplatin (15),
`
`00
`0\
`00
`~
`
`Sandoz Inc.
`Exhibit 1042-0005
`
`
`
`Acta Oncologica 40 (2001)
`
`Chemotherapy effects in non-small cell lung cancer
`
`331
`
`which is more easily administered and is associated with
`less need for antiemetics as well as less nephro- and
`neurotoxicity. Carboplatin is, however, more myelosup(cid:173)
`pressive than cisplatin. In a comparative trial, 109 patients
`were randomised to receive cisplatin 120 mg/m2 or carbo(cid:173)
`platin 325 mg/m2 combined with etoposide 100 mg/m2 iv
`days one to three. Response rate was higher with cisplatin
`(27 vs 16%; p = 0.07) but survival was similar (median
`value, 30 vs 27 weeks). Leucopenia, nausea, diarrhoea and
`nephrotoxicity were less common with carboplatin (16).
`Single agent treatment with etoposide was compared
`with etoposide + carboplatin in a randomised study com(cid:173)
`prising 120 patients with NSCLC stage IV (17). The dose
`of etoposide was 50 mg/m2
`, orally, days 1-21 and carbo(cid:173)
`platin was given at a dose of 400 mg/m2 on day 1. Cycle
`length was four weeks and duration of treatment was up to
`six cycles. Response rates were 20 vs 31 % (p = 0.19) and
`median survival was significantly prolonged by the addi(cid:173)
`tion of carboplatin, nine vs five months (p = 0.015).
`
`Platinum-based chemotherapy, QoL and survival
`
`QoL has been a neglected dimension in the majority of
`previous trials. Data are, however, emerging from more
`recent studies. Improvement of QoL scores is suggested in
`a number of uncontrolled studies but randomised trials
`comparing chemotherapy with BSC which report QoL as
`an endpoint are few. In addition, the analysis of QoL data
`is associated with a number of methodological problems,
`related to loss of information when the condition of the
`patients is worsened and the definition of clinically rele(cid:173)
`vant changes in QoL scores (18).
`In a randomised study, 288 patients with NSCLC stage
`lllB or IV and performance status 0-2 were randomised
`to receive BSC or combination chemotherapy consisting of
`ifosfamide, epirubicin, and cisplatin or mitomycin C, cis(cid:173)
`platin, and vinblastine. Serial estimates of QoL were made
`by using a modified functional living index-cancer (T(cid:173)
`FLIC) and a modified QoL index (T-QLI) (19). Patients
`treated with chemotherapy experienced longer survival
`(median seven months vs two and a half month; p =
`0.006). Scores of T-FLIC and T-QLI were improved (p =
`0.0001 for T-FLIC after three months of chemotherapy
`and at two months after termination of treatment as
`compared with pretreatment values) for patients receiving
`chemotherapy (20).
`In a British investigation, patients with previously un(cid:173)
`treated unresectable NSCLC, age up to 75 years and
`performance status 0-2 were divided into two groups (21).
`Patients with no evidence of metastatic disease, no pleural
`effusion and tumours encompassable in a radical radio(cid:173)
`therapy port were eligible for the MIC! trial, and the
`remainder were eligible for the MIC2 trial. In the MIC!
`trial, 446 patients were randomised to receive up to four
`, ifos(cid:173)
`cycles of MIC chemotherapy (mitomycin 6 mg/m2
`, and cisplatin 50 mg/m2
`) every 21 days,
`famide 3 g/m2
`
`Table 2
`
`WHO performance status scale
`
`0 Able to carry out all normal activity without restriction.
`Restricted in physically strenuous activity but ambulatory
`and able to carry out light work.
`2 Ambulatory and capable of all self-care but unable to
`carry out any work; up and about more than 50 per
`cent of waking hours.
`3 Capable of only limited self-care; confined to bed or chair
`more than 50 per cent of waking hours.
`4 Completely disabled; cannot carry out any self-care; totally
`confined to bed or chair.
`
`followed by radiotherapy, or to receive radiotherapy
`alone. In the MIC2 trial, 351 patients were randomised to
`BSC plus the same chemotherapy or BSC alone. Short(cid:173)
`term change in QoL was assessed in a subgroup of pa(cid:173)
`tients. Median survival time in the MIC! trial was 11. 7
`months with chemotherapy + irradiation and 9.7 months
`with radiotherapy alone (p = 0.14) whereas in MIC2 me(cid:173)
`dian survival amounted to 6.7 and 4.8 months, respectively
`(p = 0.03). QoL, as assessed in a subset of 134 patients
`from start of trial to week six, improved with chemother(cid:173)
`apy and deteriorated without chemotherapy. Another
`study compared BSC with a combination of carboplatin
`and etoposide every four weeks to a maximum of eight
`cycles in patients with NSCLC stage lllB and IV (15).
`Forty-eight patients were randomised and 102 other pa(cid:173)
`tients were treated on an individual preference basis, due
`to difficulties to gain acceptance for randomisation. Me(cid:173)
`dian survival was 29 weeks with chemotherapy and 11
`weeks with BSC and one-year survival was 28 vs 8%.
`Patients who received chemotherapy reported less dysp(cid:173)
`noea, cough, and fatigue and improved physical function.
`Results were similar in randomised and non-randomised
`patients.
`The last three studies were not part of the meta-analyses
`presented above. The prolongation of median survival by
`up to 4.5 months and improvement of QoL by platinum(cid:173)
`based combination chemotherapy suggest that symp(cid:173)
`tomatic palliation from treatment is more important than
`side-effects from cytotoxic compounds. Other randomised
`studies reporting QoL data, comparing single agent treat(cid:173)
`ment with BSC, will be cited below.
`It should be noted that entrance criteria for randomised
`studies usually exclude patients older than 70- 75 years as
`well as patients with performance status 3-4 according to
`the WHO scale (Table 2). Such individuals form, however,
`a sizeable proportion of the population of patients with
`advanced NSCLC and very limited documentation is
`available from randomised studies concerning the potential
`role of chemotherapy in these groups. An exception is an
`Italian trial, which specifically studied patients who were
`70 years of age or older (22).
`
`00
`0\
`00
`~
`
`Sandoz Inc.
`Exhibit 1042-0006
`
`
`
`332
`
`S. Sorenson et al.
`
`Novel agents -
`
`gemcitahine
`
`Gemcitabine is an analogue of cytosine arabinoside (ara(cid:173)
`C) with greater membrane permeability, greater affinity for
`deoxycytidine kinase, and longer intracellular retention
`with prolonged inhibition of DNA synthesis. Gemcitabine
`has undergone extensive evaluation in phase II trials (six
`studies in a total of 501 chemotherapy-naive patients),
`with a remarkably consistent response rate around 20%
`after extramural review (23). Toxicity has been mild. The
`dose-response relationship is currently under study in a
`randomised trial.
`A randomised comparison between gemcitabine as sin(cid:173)
`gle agent therapy and BSC has been performed in the
`United Kingdom (24). One hundred and fifty patients were
`treated with gemcitabine 100 mg/m2 on days one, eight
`and 15 of a 28-day schedule and 149 patients were allo(cid:173)
`cated to BSC. In addition, patients were treated with
`palliative radiotherapy as needed. Irradiation was exten(cid:173)
`sively used in the BSC arm. Median time to palliative
`radiotherapy was 29 weeks in the gemcitabine arm vs 3.8
`weeks in the BSC arm. Survival curves were identical
`(median survival time 24.4 weeks for gemcitabine vs 25.5.
`weeks for BSC) but QoL was improved by gemcitabine.
`Toxicity from gemcitabine was mild.
`, administered on days one,
`Gemcitabine I 000 mg/m2
`eight and 15 of a four-week cycle was compared with
`cisplatin 100 mg/m2 day one plus etoposide 100 mg/m2
`intravenously on days one, two and three in a randomised
`multicentre trial in 147 patients (25). In an early analysis,
`response rates were 17% and 15%, respectively. Time to
`progression was similar in both arms (4.2 vs 4.1 months).
`Median survival was 6.3 and 7.3 months, respectively.
`None of these differences was statistically significant.
`There was significantly more alopecia and nausea and
`vomiting in the cisplatin + etoposide arm. It was con(cid:173)
`cluded that single agent gemcitabine was at least as effec(cid:173)
`tive as cisplatin and etoposide but less toxic.
`In a Spanish trial, 135 previously untreated patients with
`advanced NSCLC were randomised to receive either gem(cid:173)
`citabine I 250 mg/m2 iv days one and eight plus cisplatin
`100 mg/m2 day one or etoposide 100 mg/m2 iv days one to
`three plus cisplatin 100 mg/m2 day one on a three-week
`cycle (26). Response rate, externally validated, was 41 vs
`22% (p = 0.02) and time to progression was significantly
`prolonged by the gemcitabine combination (6.9 vs 4.3
`months; p = 0.01). Survival was, however, similar in both
`arms (median survival time 8.7 vs 7.2 months; p = 0.18).
`The toxicity profile was similar in both arms.
`A randomised trial including 522 patients with advanced
`or metastatic NSCLC compared gemcitabine I 000 mg/m2
`days one, eight and 15 in combination with cisplatin 100
`mg/m2 day one of the 28-day cycle with cisplatin at this
`dosage alone (27). Tumour response rate was 30% for the
`combination compared with 11 % for cisplatin alone (p <
`
`Acta Oncologica 40 (2001)
`
`0.0001). Median time to progression was longer for the
`combination, 5.6 vs 3. 7 months (p = 0.001) as was median
`overall survival, 9. I vs 7 .6 months (p = 0.04).
`
`Novel agents -
`
`taxanes
`
`Significant activity against advanced NSCLC has been
`observed in phase II studies of the taxanes, docetaxel and
`paclitaxel, with response rates mostly in the order of
`20-30% (28). Two prospective randomised comparisons
`between paclitaxel + cisplatin and podophyllotoxin +
`cisplatin have been carried out, one by the European
`Organisation on Research and Treatment in Cancer
`(EORTC) and one by ECOG.
`In the EOR TC trial, 332 patients were randomised to
`receive A: cisplatin 80 mg/m2 on day one+ teniposide 100
`mg/m2 days one, three and five or B: cisplatin in the same
`dose+ paclitaxel 175 mg/m2 as a three-hour infusion on
`day one. Cycles were repeated every three weeks. Sixty per
`cent of the patients had stage IV disease and the arms were
`similar with respect to prognostic factors. Response rates
`were 28% and 41%, respectively (p = 0.018). Survival was
`similar in both arms (median surviva