- Email: [email protected]
Radical radiotherapy with or without gemcitabine in patients with early stage medically inoperable non-small cell lung cancer
Lung Cancer 77 (2012) 532–536
Contents lists available at SciVerse ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Radical radiotherapy with or without gemcitabine in patients with early stage medically inoperable non-small cell lung cancer Allan Price a,∗ , Ann Yellowlees b , Catriona Keerie b , Susan Russell c , Corinne Faivre-Finn d , David Gilligan e , Michael Snee f , Geraldine Skailes g , Matthew Hatton h , Sara Erridge a , Nazia Mohammed i a
Edinburgh Cancer Centre, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK QUANTICS Consulting, Logan Building, Roslin BioCentre, Roslin, Edinburgh EH25 9TT, UK c ISD Cancer Clinical Trials Team, 1st Floor, Gyle Square, 1 South Gyle Crescent, Edinburgh EH12 9EB, UK d Christie NHS Foundation Trust, Wilmslow Road, Manchester M20 4BX, UK e The Oncology Centre, Box 193, R4 Block, Addenbrooke’s Hospital, Hills Road, Cambridge CB2, UK f Yorkshire Cancer Centre, St. James’s Institute of Oncology, Leeds, UK g Royal Lancaster Infirmary, Ashton Road, Lancaster LA1 4RP, UK h Weston Park Hospital, Whitham Road, Sheffield S10 2SJ, UK i Beatson Oncology Centre, Gartnavel General Hospital, 1053 Great Western Road, Glasgow, Scotland G12 0NA, UK b
a r t i c l e
i n f o
Article history: Received 5 March 2012 Received in revised form 25 April 2012 Accepted 3 May 2012 Keywords: Non-small cell lung cancer Radiotherapy Chemotherapy Gemcitabine Randomised trial
a b s t r a c t Background: Preclinical and phase I data suggest gemcitabine to be a potent radiosensitiser. This multicentre study addressed whether the addition of low dose gemcitabine to radical radiotherapy improved 2 year event-free survival in patients with medically inoperable stages I–II non-small cell lung cancer. Aim: To determine whether low dose gemcitabine increased event–free survival in patients with T1-2 N0-1 M0 NSCLC deemed unfit for surgery. Methods: Patients with T1-2 N0-1 M0 NSCLC deemed unfit for surgery were randomised to 3D conformal radiotherapy delivering 55 Gy in 20 fractions over 4 weeks to known sites of cancer with (Arm B) or without (Arm A) 100 mg/m2 weekly gemcitabine. Results: Study entry was terminated early because of slow accrual. 111 patients were randomised between March 2003 and December 2005, of whom 4 withdrew consent and 2 were lost to follow-up. Median age was 75 (range 49–88) years and 67 (63%) were male. 86 (81%) were PS 0–1 and 31 (30%) Charlson index 2 or greater. Event-free survival in arm A and B, respectively, was 42% and 46% at 2 years and 20% and 31% at 5 years (p = 0.72), while overall survival was 56% and 52% at 2 years and 20% and 33% at 5 years (p = 0.87). Two deaths from accelerated interstitial lung disease were seen in arm B, but toxicity was otherwise mild. Conclusion: No evidence of an improvement in event-free survival was seen with the addition of weekly gemcitabine at this dose for patients with early stage NSCLC unfit for surgery, although the power of the study was low. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Gemcitabine is an anti-metabolite with an established role in the chemotherapy of a number of solid tumours, particularly non-small cell lung (NSCLC) and pancreatic cancer. It has also been shown to act as a potent radiosensitiser in vitro, and this was confirmed by the demonstration of significantly increased toxicity in a number of clinical trials, although to date no routine role has been established for this indication [1,2].
∗ Corresponding author. Tel.: +44 1315372205; fax: +44 1315372240. E-mail address: [email protected] (A. Price). 0169-5002/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.lungcan.2012.05.089
An early attempt to combine full dose gemcitabine and radiotherapy in non-small cell lung cancer resulted in a significant increase in severe or fatal toxicity [3]. Accordingly we conducted a phase I dose escalation study which established 300 mg/m2 weekly as a safe dose of gemcitabine to take forward into phase III trials combined with involved field thoracic radiotherapy [4]. To test the clinical effectiveness of gemcitabine as a radiosensitiser, we wished to identify a setting where local control was paramount in the survival of patients, and local failure easily identified as a component of cancer progression. Surgery is the standard of care for stages I–II NSCLC [5], but many patients are unfit for surgery because of poor cardiorespiratory function, or elect not to undergo surgery because of the perceived anaesthetic risks. Radiotherapy is the standard treatment in such patients, and the
A. Price et al. / Lung Cancer 77 (2012) 532–536
present trial was designed and carried out before the development of Stereotactic Ablation Radiotherapy in the UK [6]. We report the results of a randomised trial in this patient group, testing a hypofractionated radiotherapy regimen in routine use in the UK with and without weekly gemcitabine. 2. Methods This study was initially designed as an open-label, national multi-centre randomised phase III study to determine whether there was an advantage in progression-free survival associated with the addition of gemcitabine as a radiosensitiser to thoracic radiotherapy in participants with previously untreated medically inoperable T1-2 N0-1 M0 NSCLC, conducted according to the rules of Good Clinical Practice as defined by the Medicines for Human Use (Clinical Trials) Regulations, 2004. The safety and appropriate continuation of the trial was monitored by a 3 person Independent Data Monitoring Committee. Ethical approval was obtained from national and local ethics committees. It was registered as ISRCTN61727489 and EudraCT 2004-000136-86. 2.1. Patient eligibility Patients with T1-2 N0-1 M0 (UICC 1997) NSCLC were eligible if deemed unfit for resection in the opinion of a thoracic surgeon, according to the British Thoracic Society guidelines [7]. Participants required pathological documentation of NSCLC, had to be over 18 years of age, PS2 or better, have adequate cardiac, haematological and renal function, no significant infection, no prior non-surgical treatment for lung cancer, no other malignancy in the previous 5 years, not be pregnant or breast feeding and use adequate contraception if appropriate. Minimum lung function was an FEV1 either ≥0.7 l/s or >35% predicted for age and surface area and a transfer factor for carbon monoxide ≥40% of the predicted value.
533
Table 1 Major and minor radiotherapy protocol variations. Parameter
Minor violation
Major violation
GTV margins Dose distribution Treatment time Cord dose Lung dose Oesophageal length Set up error
≤5 mm variation ±7–10% 5–7 additional days 75–80% prescribed dose V20 > 40% to <50% >12 to <15 cm 5–10 mm
>5 mm variation ±>10% >7 additional days >80% prescribed dose V20 > 50% >15 cm >10 mm
radiation oncologists, with variations defined as major or minor violations in the protocol (Table 1). In this period pencil beam algorithms were the UK standard. Gemcitabine 100 mg/m2 was administered intravenously over approximately 30 min 2 h prior to commencing radiotherapy on days 1, 8, 15 and 22. 2.4. Randomisation Telephone randomisation was carried out centrally by the Scottish Cancer Trials Network (later CACTUS). There was no stratification. 2.5. Response and toxicity assessments During radiotherapy patients had weekly full blood count, biochemistry, toxicity assessment using the CG-CTC v2.0 scale, history and physical examination. Subsequent to treatment participants had 3 monthly review for 1 year then annually with history, examination and toxicity assessment, CT thorax and abdomen at 3, 4 (if response at 3) and 24 months, and pulmonary function tests at 3 and 12 months. 2.6. Endpoints and statistics
2.2. Baseline assessments Pre-randomisation assessments included history and examination, chest X-ray, CT scan thorax and abdomen, full blood count including differential white cell count and platelet count, biochemistry including Mg2+ , Ca2+ , ALT, AST and ALP, and pulmonary function tests including transfer factor for carbon monoxide. During the period of the study PET scans were not available in the majority of UK centres for the staging of lung cancer. 2.3. Treatment Initially investigators were given a choice whether to deliver 60 Gy in 30 daily fractions over 6 weeks, or 55 Gy in 20 daily fractions over 4 weeks. The former schedule was withdrawn when it became apparent that it would be used by only one centre. Radiotherapy was CT planned with immobilisation techniques left to the investigators choice and the participants breathing freely. A planning target volume with 1.5 cm margins laterally and 2 cm supero-inferiorly was constructed around radiologically visible primary tumour and involved lymph nodes (1.5 cm in short axis). Elective nodal irradiation was not permitted. Defined organs at risk were the spinal cord (no point greater than 41.25 Gy), lung (less than 40% to receive 20 Gy), oesophagus (<12 cm within high dose volume) and heart (less than 30% to receive 55 Gy). Heterogeneity corrections were mandatory, and the PTV to be within ±7% of the dose prescribed at the ICRU reference point. Maximum overall treatment time was 30 days, with twice daily treatments with 6 h gap mandated if longer interruptions occurred. Quality assurance required central review of a pre-determined number of plans, contours and beam films by the Chief Investigator and two other
Initially the trial had an accrual target of 230 patients in each arm of the study, to give a 90% power at the 5% level of detecting an improvement from 30% to 45% in 2 year progression-free survival, and for either 55 Gy or 60 Gy 100 patients per arm to give an 80% power at the 5% level of detecting an improvement from 30% to 50% in 2 year progression-free survival. Subsequently, when the deadline for the withdrawal of funding was shortened this was revised to 52 patients per arm, to give a 90% power at the 5% level of detecting an improvement from 30% to 60% in event-free survival at 2 years. Analysis was on an intention to treat basis, from the date of randomisation. A detailed statistical analysis plan was drawn up before the trial opened. Patients who were alive at the end of the trial and had not progressed were considered to be event-free and were censored at the date of last contact. Patients who had died by the end of the trial with no record of a progression were not censored. For these patients date of death was used in the analysis of overall survival and event-free survival. For those patients (alive or dead) who experienced a progression during the trial, the date of the first progression was used in the analysis of event-free survival. 3. Results The disposition of participants in the trial is shown in Fig. 1. One hundred and eleven participants were randomised between March 2003 and December 2005. 107 were available for analysis, 55 randomised to radiotherapy alone and 52 to radiotherapy plus weekly gemcitabine. Demographic and clinical characteristics are shown in Table 2. The trial closed in June 2011, over 5 years after final patient entry.
534
A. Price et al. / Lung Cancer 77 (2012) 532–536
Fig. 1. CONSORT diagram of disposition of participants in GRiN trial.
Table 2 Demographics and clinical characteristics. Factor
Arm A (radiotherapy alone)
Arm B (radiotherapy plus gemcitabine)
p
Number of participants Age, years (median, range) Gender (% male) Smokers, % ongoing PS, % 0, 1, 2 Charlson index (median, range) FEV1, % predicted (median, range) FEV1, l/s (median, range) Transfer factor, % predicted (median, range) Primary cancer, mm (median, range) Hilar node involved
56 75 (49–88) 67 36 24, 54, 22 1 (0–6) 60 (29–134) 1.5 (0.8–3.0) 67 (42–152) 38 (3–89) 19
55 74 (58–88) 59 24 18, 67, 14 1 (0–6) 63 (30–135) 1.4 (0.8–4.1) 79 (54–154) 30 (3–55) 20
0.36 0.42 0.21 0.86 0.19 0.88 0.53 0.05 0.4 –
3.1. Treatment delivery
in arm A and 14 in arm B (Table 4) were thought possibly related to treatment (for number of participants, 2 = 6.85, df = 1, p < 0.01).
All but one participant received the full radiotherapy dose in arm A. In arm B, one participant received only 50 Gy and one 54 Gy, and 4 had gemcitabine dose reductions. Major and minor protocol variations from radiotherapy, blinded to treatment arm, are shown in Table 3. 3.2. Toxicity Two participants in arm B died of acceleration of pre-existing interstitial lung disease. Otherwise grade 3 pneumonitis occurred in 2 (4%) participants in arm A and none in arm B, and grade 2 in 4 (7% and 8%) in each arm. Grade 3 oesophagitis occurred in 1 (2%) participant in arm A and 2 (4%) in arm B, and grade 2 in 3 (5%) and 4 (8%) in arm A and B, respectively. There were 10 serious adverse events in 9 participants in arm A and 25 in 19 participants in arm B (for number of participants, 2 = 5.02, df = 1, p < 0.03), of which 4 Table 3 Radiotherapy protocol variations. Parameter
Minor violation
Major violation
Insufficient data
GTV margins Dose distribution Treatment time Cord dose Lung dose Oesophageal length Set up error
1 (3%) 7 (18%) 0 0 0 0 3 (8%)
1 (3%) 11 (29%) 0 0 0 0 0
7 (18%) 3 (8%) 8 (21%) 1 (3%) 2 (5%) 3 (8%) 18 (47%)
Total
11 (4%)
12 (5%)
42 (16%)
3.3. Lung function Data was available for 82 (77%) participants at 3 months and 40 (37%) at 12 months. FEV1 was reduced by a median of 4% (range −30% to +22%) in arm A and by 2% (range −32% to +34%) in arm B at 3 months, and by 2% (range −19% to +16%) in arm A at 12 months. In arm B at 12 months, FEV1 was increased by a median of 2% (range −12% to +34%). Transfer factor was reduced by a median of 7% (range −41% to +50%) in arm A and by 4% (range −37% to +36%) in arm B at 3 months, and by 9% (range −33% to +23%) in arm A, and by 7% (−75% to +21%) in arm B at 12 months. Table 4 Serious adverse events with possible attribution to protocol treatment. Arm A
Arm B
Event
Number
Event
Number
Dyspnoea Chest pain
1 1
1 2
Rib insufficiency fracture Myocardial infarction Perforated viscus
1 1 1
Hypotension Pulmonary fibrosis/interstitial lung disease Dyspnoea Pneumonitis Pneumothorax COPD exacerbation Rash Oesophagitis Subcutaneous reaction Chest pain Transient ischaemic attack
2 1 1 1 1 2 1 1 1
A. Price et al. / Lung Cancer 77 (2012) 532–536
535
Fig. 2. Event-free survival of participants in GRiN trial.
3.4. Progression and survival At the time of final analysis in November 2011, 34 participants remained alive when last seen. The primary endpoint for the study was not met; at 2 years 22 participants had progressed in arm A and 21 in arm B, while 23 had died in arm A and 22 in arm B. Event-free survival in arm A and B, respectively, was 42% and 46% at 2 years and 20% and 31% at 5 years (p = 0.72) (Fig. 2). Actuarial overall survival was 56% and 52% at 2 years and 20% and 33% at 5 years (p = 0.87) in arms A and B, respectively (Fig. 3). A component of local relapse was observed in 14 (25%) participants in arm A, and 9 (18%) in arm B, and of distant relapse in 10 (18%) and 11 (22%), respectively.
4. Discussion This investigation of the radiosensitising role of gemcitabine in patients with early NSCLC failed to show any evidence of benefit, although the combination of radiotherapy and gemcitabine was
well tolerated and acceptable to participants. The non-significant result may reflect a lack of any radiosensitising effect in the clinic, but there are also a number of possible reasons why this trial might have failed to detect any such benefit if it did exist. First and foremost, the eventual trial was much smaller than originally planned, because accrual was slower than expected [8] and because this led to an early withdrawal of funding when it became clear that planned accrual would not be reached. This has been a feature of many recent radiotherapy trials in lung cancer [9–11], for reasons which are unclear. The dose of gemcitabine chosen for this trial was rather lower than that established in phase I trials in participants with stage III disease [4]. This choice was made because of concerns about the fitness of the expected study population and data in head and neck cancer showing no increase in intra-tumoural levels of gemcitabine with increased gemcitabine doses from 50 to 300 mg/m2 weekly, at all of which doses significant tumour responses and marked radiation toxicity was seen [1]. No similar data from tumour biopsies is available, nor could it be easily obtained, in patients
Fig. 3. Overall survival of participants in GRiN trial.
536
A. Price et al. / Lung Cancer 77 (2012) 532–536
with lung cancer. Both the Charlson indices and the lung function data suggest this was a less unfit population than anticipated and a higher dose might have been possible. Yarnold has argued that all radiotherapy trials should include two experimental arms to facilitate radiobiological assessments, but the numbers achieved here would have precluded such a design. A recent study has suggested gemcitabine administration 1 h before radiotherapy [12], although in vitro data at the time of study design suggested a 24 h interval between gemcitabine and radiotherapy with effects lasting up to 72 h [13], with radiosensitisation seen at non-cytotoxic doses [14]. We cannot exclude the hypothesis that the gemcitabine dose chosen for this study was inadequate, although even at this dose there appeared to be an increased number of serious adverse events in the experimental arm. The study design was based on the prognostic information available at that time, much of it dating from much earlier studies. The 2 year progression-free survival seen in the control arm was above 40%, approaching that we had aimed to see in the experimental arm, further reducing the power of the study. This may have reflected the greater fitness seen than expected in the study population, but also improvements in staging, although CT-PET was not widely available in the UK in 2003–2005, and in management of non-cancer co-morbidities in this group. More recent data from series of stereotactic radiotherapy, suggest up to 10% of this group will relapse locally, about 20% of this group will die from systemic relapse, and a similar number from non-cancer causes [15]. The treatment delivered here appeared to give a much lower relapse rate than anticipated, 18–25%, and given the very low levels of toxicity, suggests further escalation of the radiotherapy dose with further reduction in local failure might be possible with conventional radiotherapy. CALGB 39904, in which participants with T1-2N0M0 NSCLC and poor lung function had radiotherapy dose escalated from 70 Gy in 29 to 70 Gy in 17 fractions reported 3 local relapses in 39 participants, median overall survival of 38.5 and median progression-free survival of 28.6 months [16]. Our data do not indicate a large number of competing events (i.e. non-cancer deaths). While there is always significant uncertainty about the quality of death certification in patients with lung cancer, there seemed to be a reasonable match between the number who had developed progressive disease and the number who had died at 2 years. One would not expect prolonged survival in this group once relapse had been diagnosed. 5. Conclusion We did not see any indication of an improvement in outcome due to weekly gemcitabine 100 mg/m2 added to radical radiotherapy in this population. The rate of local relapse was lower, and the observed survival higher, in both groups than had been expected before the trial. Conflict of interest statement During the period of trial design and patient accrual, the Chief Investigator (AP) and a co-author (MPS) served on Advisory Boards with Lilly Oncology. Acknowledgements The study was funded by a Grant based on patient accrual from Lilly Oncology to the University of Edinburgh. The manuscript was reviewed by Lilly Oncology before submission. Our thanks to all participants who consented to take part in this trial.
In addition to the affiliated centres of the authors of the manuscript, participants were entered by the investigators at the following centres: Dr. K. Fife, Dr. S. Old, Addenbrooke’s Hospital, Cambridge; Dr. A. Lamont, Southend Hospital; Dr. P. Burt, Christie Hospital, Manchester; Professor M. Brada, Royal Marsden Hospital, Sutton; Dr. F. Little, Western General Hospital, Edinburgh; Dr. P. Mulvenna, Dr. P. Atherton, Newcastle General Hospital; Dr. V. Laurence, Poole Hospital; Dr. E. Toy, Torbay District Hospital; Dr. N. Shah, Mount Vernon Hospital, Northwood; Dr. S. Upadhyay, Diana Princess of Wales Hospital, Grimsby; Dr. Samanchi, New Cross Hospital, Wolverhampton; Dr. C. Martin, Norfolk & Norwich Hospital; Dr. P. Wells, St. Bartholomew’s Hospital, London; Dr. A. Chetiwardena, Queen Elizabeth Hospital, Birmingham; Dr. D. Bissett, Aberdeen Royal Infirmary; Dr. J. Simpson, Royal Sussex County Hospital, Brighton; Dr. S. Beesley, Kent Oncology Centre, Maidstone; Dr. M. Illsley, Royal Surrey County Hospital, Guildford; Dr. J. Littler, Clatterbridge Centre for Oncology, Liverpool; Dr. M. Hocking, Walsgrave Hospital, Coventry; Dr. Z. Mallik, Whiston Hospital, St. Helens. Our thanks also to the trial coordinators Bryn Dixon, Tracy McEleney, and Leanne Ferrigan. References [1] Eisbruch A, Shewach DS, Bradford CR, Littles JF, Teknos TN, Chepeha DB, et al. Radiation concurrent with gemcitabine for locally advanced head and neck cancer: a phase I trial and intracellular drug incorporation study. J Clin Oncol 2001;19:792–9. [2] Shewach DS, Lawrence TS. Antimetabolite radiosensitizers. J Clin Oncol 2007;25:4043–50. [3] Scalliet P, Goor C, Galdermans D, et al. Gemzar (gemcitabine) with thoracic radiotherapy – a phase II pilot study in chemonaive patients with advanced non-small cell lung cancer (NSCLC). Proc ASCO 1998;17:499a. [4] van Putten JW, Price A, van der Leest AH, Gregor A, Little FA, Groen HJ. A phase I study of gemcitabine with concurrent radiotherapy in stage III, locally advanced non-small cell lung cancer. Clin Cancer Res 2003 Jul;9(7):2472–7. [5] Whitson BA, Groth SS, Duval SJ, Swanson SJ, Maddaus MA. Surgery for earlystage non-small cell lung cancer: a systematic review of the video-assisted thoracoscopic surgery versus thoracotomy approaches to lobectomy. Ann Thorac Surg 2008;86:2008–16. [6] Rowell NP, Williams CJ. Radical radiotherapy for stage I/II non-small cell lung cancer in patients not sufficiently fit for or declining surgery (medically inoperable): a systematic review. Thorax 2001;56:628–38. [7] British Thoracic Society. Guidelines on the selection of patients with lung cancer for surgery. Thorax 2001;56:89–98. [8] Price A, Dixon B, Erridge SC, Mohammed N. GRiN: a trial and tribulation in respiratory radiotherapy research. Clin Oncol (R Coll Radiol) 2005;17:328–31. [9] Hatton M, Nankivell M, Lyn E, Falk S, Pugh C, Navani N, et al. Induction chemotherapy and continuous hyperfractionated accelerated radiotherapy (CHART) for patients with locally advanced inoperable non-small-cell lung cancer: the MRC INCH randomized trial. Int J Radiat Oncol Biol Phys 2010;(October). [10] Belderbos J, Uitterhoeve L, van Zandwijk N, Belderbos H, Rodrigus P, van de Vaart P, et al. EORTC LCG and RT group randomised trial of sequential versus concurrent chemo-radiotherapy in patients with inoperable non-small cell lung cancer (EORTC 08972-22973). Eur J Cancer 2007;43:114–21. [11] Belani CP, Wang W, Johnson DH, Wagner H, Schiller J, Veeder M, et al. Eastern Cooperative Oncology Group Phase III study of the Eastern Cooperative Oncology Group (ECOG 2597): induction chemotherapy followed by either standard thoracic radiotherapy or hyperfractionated accelerated radiotherapy for patients with unresectable stage IIIA and B non-small-cell lung cancer. J Clin Oncol 2005;23:3760–7. [12] Wouters A, Pauwels B, Lardon F, Pattyn GG, Lambrechts HA, Baay M, et al. In vitro study on the schedule-dependency of the interaction between pemetrexed, gemcitabine and irradiation in non-small cell lung cancer and head and neck cancer cells. BMC Cancer 2010;10:441. [13] Milas L, Fujii T, Hunter N, Elshaikh M, Mason K, Plunkett W, et al. Enhancement of tumor radioresponse in vivo by gemcitabine. Cancer Res 1999;59:107–14. [14] Trodella L, D’Angelillo RM, Ramella S, Cellini F, Ciresa M, Cesario A, et al. Chemoradiotherapy in non-small cell lung cancer: the role of gemcitabine. Ann Oncol 2006;17(Suppl. 5):v52–4. [15] Haasbeek CJ, Lagerwaard FJ, Antonisse ME, Slotman BJ, Senan S. Stage I nonsmall cell lung cancer in patients aged > or =75 years: outcomes after stereotactic radiotherapy. Cancer 2010;116:406–14. [16] Bogart JA, Hodgson L, Seagren SL, Blackstock AW, Wang X, Lenox R, et al. Phase I study of accelerated conformal radiotherapy for stage I non-small-cell lung cancer in patients with pulmonary dysfunction: CALGB 39904. J Clin Oncol 2010;28:202–6.
Contents lists available at SciVerse ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Radical radiotherapy with or without gemcitabine in patients with early stage medically inoperable non-small cell lung cancer Allan Price a,∗ , Ann Yellowlees b , Catriona Keerie b , Susan Russell c , Corinne Faivre-Finn d , David Gilligan e , Michael Snee f , Geraldine Skailes g , Matthew Hatton h , Sara Erridge a , Nazia Mohammed i a
Edinburgh Cancer Centre, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK QUANTICS Consulting, Logan Building, Roslin BioCentre, Roslin, Edinburgh EH25 9TT, UK c ISD Cancer Clinical Trials Team, 1st Floor, Gyle Square, 1 South Gyle Crescent, Edinburgh EH12 9EB, UK d Christie NHS Foundation Trust, Wilmslow Road, Manchester M20 4BX, UK e The Oncology Centre, Box 193, R4 Block, Addenbrooke’s Hospital, Hills Road, Cambridge CB2, UK f Yorkshire Cancer Centre, St. James’s Institute of Oncology, Leeds, UK g Royal Lancaster Infirmary, Ashton Road, Lancaster LA1 4RP, UK h Weston Park Hospital, Whitham Road, Sheffield S10 2SJ, UK i Beatson Oncology Centre, Gartnavel General Hospital, 1053 Great Western Road, Glasgow, Scotland G12 0NA, UK b
a r t i c l e
i n f o
Article history: Received 5 March 2012 Received in revised form 25 April 2012 Accepted 3 May 2012 Keywords: Non-small cell lung cancer Radiotherapy Chemotherapy Gemcitabine Randomised trial
a b s t r a c t Background: Preclinical and phase I data suggest gemcitabine to be a potent radiosensitiser. This multicentre study addressed whether the addition of low dose gemcitabine to radical radiotherapy improved 2 year event-free survival in patients with medically inoperable stages I–II non-small cell lung cancer. Aim: To determine whether low dose gemcitabine increased event–free survival in patients with T1-2 N0-1 M0 NSCLC deemed unfit for surgery. Methods: Patients with T1-2 N0-1 M0 NSCLC deemed unfit for surgery were randomised to 3D conformal radiotherapy delivering 55 Gy in 20 fractions over 4 weeks to known sites of cancer with (Arm B) or without (Arm A) 100 mg/m2 weekly gemcitabine. Results: Study entry was terminated early because of slow accrual. 111 patients were randomised between March 2003 and December 2005, of whom 4 withdrew consent and 2 were lost to follow-up. Median age was 75 (range 49–88) years and 67 (63%) were male. 86 (81%) were PS 0–1 and 31 (30%) Charlson index 2 or greater. Event-free survival in arm A and B, respectively, was 42% and 46% at 2 years and 20% and 31% at 5 years (p = 0.72), while overall survival was 56% and 52% at 2 years and 20% and 33% at 5 years (p = 0.87). Two deaths from accelerated interstitial lung disease were seen in arm B, but toxicity was otherwise mild. Conclusion: No evidence of an improvement in event-free survival was seen with the addition of weekly gemcitabine at this dose for patients with early stage NSCLC unfit for surgery, although the power of the study was low. © 2012 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Gemcitabine is an anti-metabolite with an established role in the chemotherapy of a number of solid tumours, particularly non-small cell lung (NSCLC) and pancreatic cancer. It has also been shown to act as a potent radiosensitiser in vitro, and this was confirmed by the demonstration of significantly increased toxicity in a number of clinical trials, although to date no routine role has been established for this indication [1,2].
∗ Corresponding author. Tel.: +44 1315372205; fax: +44 1315372240. E-mail address: [email protected] (A. Price). 0169-5002/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.lungcan.2012.05.089
An early attempt to combine full dose gemcitabine and radiotherapy in non-small cell lung cancer resulted in a significant increase in severe or fatal toxicity [3]. Accordingly we conducted a phase I dose escalation study which established 300 mg/m2 weekly as a safe dose of gemcitabine to take forward into phase III trials combined with involved field thoracic radiotherapy [4]. To test the clinical effectiveness of gemcitabine as a radiosensitiser, we wished to identify a setting where local control was paramount in the survival of patients, and local failure easily identified as a component of cancer progression. Surgery is the standard of care for stages I–II NSCLC [5], but many patients are unfit for surgery because of poor cardiorespiratory function, or elect not to undergo surgery because of the perceived anaesthetic risks. Radiotherapy is the standard treatment in such patients, and the
A. Price et al. / Lung Cancer 77 (2012) 532–536
present trial was designed and carried out before the development of Stereotactic Ablation Radiotherapy in the UK [6]. We report the results of a randomised trial in this patient group, testing a hypofractionated radiotherapy regimen in routine use in the UK with and without weekly gemcitabine. 2. Methods This study was initially designed as an open-label, national multi-centre randomised phase III study to determine whether there was an advantage in progression-free survival associated with the addition of gemcitabine as a radiosensitiser to thoracic radiotherapy in participants with previously untreated medically inoperable T1-2 N0-1 M0 NSCLC, conducted according to the rules of Good Clinical Practice as defined by the Medicines for Human Use (Clinical Trials) Regulations, 2004. The safety and appropriate continuation of the trial was monitored by a 3 person Independent Data Monitoring Committee. Ethical approval was obtained from national and local ethics committees. It was registered as ISRCTN61727489 and EudraCT 2004-000136-86. 2.1. Patient eligibility Patients with T1-2 N0-1 M0 (UICC 1997) NSCLC were eligible if deemed unfit for resection in the opinion of a thoracic surgeon, according to the British Thoracic Society guidelines [7]. Participants required pathological documentation of NSCLC, had to be over 18 years of age, PS2 or better, have adequate cardiac, haematological and renal function, no significant infection, no prior non-surgical treatment for lung cancer, no other malignancy in the previous 5 years, not be pregnant or breast feeding and use adequate contraception if appropriate. Minimum lung function was an FEV1 either ≥0.7 l/s or >35% predicted for age and surface area and a transfer factor for carbon monoxide ≥40% of the predicted value.
533
Table 1 Major and minor radiotherapy protocol variations. Parameter
Minor violation
Major violation
GTV margins Dose distribution Treatment time Cord dose Lung dose Oesophageal length Set up error
≤5 mm variation ±7–10% 5–7 additional days 75–80% prescribed dose V20 > 40% to <50% >12 to <15 cm 5–10 mm
>5 mm variation ±>10% >7 additional days >80% prescribed dose V20 > 50% >15 cm >10 mm
radiation oncologists, with variations defined as major or minor violations in the protocol (Table 1). In this period pencil beam algorithms were the UK standard. Gemcitabine 100 mg/m2 was administered intravenously over approximately 30 min 2 h prior to commencing radiotherapy on days 1, 8, 15 and 22. 2.4. Randomisation Telephone randomisation was carried out centrally by the Scottish Cancer Trials Network (later CACTUS). There was no stratification. 2.5. Response and toxicity assessments During radiotherapy patients had weekly full blood count, biochemistry, toxicity assessment using the CG-CTC v2.0 scale, history and physical examination. Subsequent to treatment participants had 3 monthly review for 1 year then annually with history, examination and toxicity assessment, CT thorax and abdomen at 3, 4 (if response at 3) and 24 months, and pulmonary function tests at 3 and 12 months. 2.6. Endpoints and statistics
2.2. Baseline assessments Pre-randomisation assessments included history and examination, chest X-ray, CT scan thorax and abdomen, full blood count including differential white cell count and platelet count, biochemistry including Mg2+ , Ca2+ , ALT, AST and ALP, and pulmonary function tests including transfer factor for carbon monoxide. During the period of the study PET scans were not available in the majority of UK centres for the staging of lung cancer. 2.3. Treatment Initially investigators were given a choice whether to deliver 60 Gy in 30 daily fractions over 6 weeks, or 55 Gy in 20 daily fractions over 4 weeks. The former schedule was withdrawn when it became apparent that it would be used by only one centre. Radiotherapy was CT planned with immobilisation techniques left to the investigators choice and the participants breathing freely. A planning target volume with 1.5 cm margins laterally and 2 cm supero-inferiorly was constructed around radiologically visible primary tumour and involved lymph nodes (1.5 cm in short axis). Elective nodal irradiation was not permitted. Defined organs at risk were the spinal cord (no point greater than 41.25 Gy), lung (less than 40% to receive 20 Gy), oesophagus (<12 cm within high dose volume) and heart (less than 30% to receive 55 Gy). Heterogeneity corrections were mandatory, and the PTV to be within ±7% of the dose prescribed at the ICRU reference point. Maximum overall treatment time was 30 days, with twice daily treatments with 6 h gap mandated if longer interruptions occurred. Quality assurance required central review of a pre-determined number of plans, contours and beam films by the Chief Investigator and two other
Initially the trial had an accrual target of 230 patients in each arm of the study, to give a 90% power at the 5% level of detecting an improvement from 30% to 45% in 2 year progression-free survival, and for either 55 Gy or 60 Gy 100 patients per arm to give an 80% power at the 5% level of detecting an improvement from 30% to 50% in 2 year progression-free survival. Subsequently, when the deadline for the withdrawal of funding was shortened this was revised to 52 patients per arm, to give a 90% power at the 5% level of detecting an improvement from 30% to 60% in event-free survival at 2 years. Analysis was on an intention to treat basis, from the date of randomisation. A detailed statistical analysis plan was drawn up before the trial opened. Patients who were alive at the end of the trial and had not progressed were considered to be event-free and were censored at the date of last contact. Patients who had died by the end of the trial with no record of a progression were not censored. For these patients date of death was used in the analysis of overall survival and event-free survival. For those patients (alive or dead) who experienced a progression during the trial, the date of the first progression was used in the analysis of event-free survival. 3. Results The disposition of participants in the trial is shown in Fig. 1. One hundred and eleven participants were randomised between March 2003 and December 2005. 107 were available for analysis, 55 randomised to radiotherapy alone and 52 to radiotherapy plus weekly gemcitabine. Demographic and clinical characteristics are shown in Table 2. The trial closed in June 2011, over 5 years after final patient entry.
534
A. Price et al. / Lung Cancer 77 (2012) 532–536
Fig. 1. CONSORT diagram of disposition of participants in GRiN trial.
Table 2 Demographics and clinical characteristics. Factor
Arm A (radiotherapy alone)
Arm B (radiotherapy plus gemcitabine)
p
Number of participants Age, years (median, range) Gender (% male) Smokers, % ongoing PS, % 0, 1, 2 Charlson index (median, range) FEV1, % predicted (median, range) FEV1, l/s (median, range) Transfer factor, % predicted (median, range) Primary cancer, mm (median, range) Hilar node involved
56 75 (49–88) 67 36 24, 54, 22 1 (0–6) 60 (29–134) 1.5 (0.8–3.0) 67 (42–152) 38 (3–89) 19
55 74 (58–88) 59 24 18, 67, 14 1 (0–6) 63 (30–135) 1.4 (0.8–4.1) 79 (54–154) 30 (3–55) 20
0.36 0.42 0.21 0.86 0.19 0.88 0.53 0.05 0.4 –
3.1. Treatment delivery
in arm A and 14 in arm B (Table 4) were thought possibly related to treatment (for number of participants, 2 = 6.85, df = 1, p < 0.01).
All but one participant received the full radiotherapy dose in arm A. In arm B, one participant received only 50 Gy and one 54 Gy, and 4 had gemcitabine dose reductions. Major and minor protocol variations from radiotherapy, blinded to treatment arm, are shown in Table 3. 3.2. Toxicity Two participants in arm B died of acceleration of pre-existing interstitial lung disease. Otherwise grade 3 pneumonitis occurred in 2 (4%) participants in arm A and none in arm B, and grade 2 in 4 (7% and 8%) in each arm. Grade 3 oesophagitis occurred in 1 (2%) participant in arm A and 2 (4%) in arm B, and grade 2 in 3 (5%) and 4 (8%) in arm A and B, respectively. There were 10 serious adverse events in 9 participants in arm A and 25 in 19 participants in arm B (for number of participants, 2 = 5.02, df = 1, p < 0.03), of which 4 Table 3 Radiotherapy protocol variations. Parameter
Minor violation
Major violation
Insufficient data
GTV margins Dose distribution Treatment time Cord dose Lung dose Oesophageal length Set up error
1 (3%) 7 (18%) 0 0 0 0 3 (8%)
1 (3%) 11 (29%) 0 0 0 0 0
7 (18%) 3 (8%) 8 (21%) 1 (3%) 2 (5%) 3 (8%) 18 (47%)
Total
11 (4%)
12 (5%)
42 (16%)
3.3. Lung function Data was available for 82 (77%) participants at 3 months and 40 (37%) at 12 months. FEV1 was reduced by a median of 4% (range −30% to +22%) in arm A and by 2% (range −32% to +34%) in arm B at 3 months, and by 2% (range −19% to +16%) in arm A at 12 months. In arm B at 12 months, FEV1 was increased by a median of 2% (range −12% to +34%). Transfer factor was reduced by a median of 7% (range −41% to +50%) in arm A and by 4% (range −37% to +36%) in arm B at 3 months, and by 9% (range −33% to +23%) in arm A, and by 7% (−75% to +21%) in arm B at 12 months. Table 4 Serious adverse events with possible attribution to protocol treatment. Arm A
Arm B
Event
Number
Event
Number
Dyspnoea Chest pain
1 1
1 2
Rib insufficiency fracture Myocardial infarction Perforated viscus
1 1 1
Hypotension Pulmonary fibrosis/interstitial lung disease Dyspnoea Pneumonitis Pneumothorax COPD exacerbation Rash Oesophagitis Subcutaneous reaction Chest pain Transient ischaemic attack
2 1 1 1 1 2 1 1 1
A. Price et al. / Lung Cancer 77 (2012) 532–536
535
Fig. 2. Event-free survival of participants in GRiN trial.
3.4. Progression and survival At the time of final analysis in November 2011, 34 participants remained alive when last seen. The primary endpoint for the study was not met; at 2 years 22 participants had progressed in arm A and 21 in arm B, while 23 had died in arm A and 22 in arm B. Event-free survival in arm A and B, respectively, was 42% and 46% at 2 years and 20% and 31% at 5 years (p = 0.72) (Fig. 2). Actuarial overall survival was 56% and 52% at 2 years and 20% and 33% at 5 years (p = 0.87) in arms A and B, respectively (Fig. 3). A component of local relapse was observed in 14 (25%) participants in arm A, and 9 (18%) in arm B, and of distant relapse in 10 (18%) and 11 (22%), respectively.
4. Discussion This investigation of the radiosensitising role of gemcitabine in patients with early NSCLC failed to show any evidence of benefit, although the combination of radiotherapy and gemcitabine was
well tolerated and acceptable to participants. The non-significant result may reflect a lack of any radiosensitising effect in the clinic, but there are also a number of possible reasons why this trial might have failed to detect any such benefit if it did exist. First and foremost, the eventual trial was much smaller than originally planned, because accrual was slower than expected [8] and because this led to an early withdrawal of funding when it became clear that planned accrual would not be reached. This has been a feature of many recent radiotherapy trials in lung cancer [9–11], for reasons which are unclear. The dose of gemcitabine chosen for this trial was rather lower than that established in phase I trials in participants with stage III disease [4]. This choice was made because of concerns about the fitness of the expected study population and data in head and neck cancer showing no increase in intra-tumoural levels of gemcitabine with increased gemcitabine doses from 50 to 300 mg/m2 weekly, at all of which doses significant tumour responses and marked radiation toxicity was seen [1]. No similar data from tumour biopsies is available, nor could it be easily obtained, in patients
Fig. 3. Overall survival of participants in GRiN trial.
536
A. Price et al. / Lung Cancer 77 (2012) 532–536
with lung cancer. Both the Charlson indices and the lung function data suggest this was a less unfit population than anticipated and a higher dose might have been possible. Yarnold has argued that all radiotherapy trials should include two experimental arms to facilitate radiobiological assessments, but the numbers achieved here would have precluded such a design. A recent study has suggested gemcitabine administration 1 h before radiotherapy [12], although in vitro data at the time of study design suggested a 24 h interval between gemcitabine and radiotherapy with effects lasting up to 72 h [13], with radiosensitisation seen at non-cytotoxic doses [14]. We cannot exclude the hypothesis that the gemcitabine dose chosen for this study was inadequate, although even at this dose there appeared to be an increased number of serious adverse events in the experimental arm. The study design was based on the prognostic information available at that time, much of it dating from much earlier studies. The 2 year progression-free survival seen in the control arm was above 40%, approaching that we had aimed to see in the experimental arm, further reducing the power of the study. This may have reflected the greater fitness seen than expected in the study population, but also improvements in staging, although CT-PET was not widely available in the UK in 2003–2005, and in management of non-cancer co-morbidities in this group. More recent data from series of stereotactic radiotherapy, suggest up to 10% of this group will relapse locally, about 20% of this group will die from systemic relapse, and a similar number from non-cancer causes [15]. The treatment delivered here appeared to give a much lower relapse rate than anticipated, 18–25%, and given the very low levels of toxicity, suggests further escalation of the radiotherapy dose with further reduction in local failure might be possible with conventional radiotherapy. CALGB 39904, in which participants with T1-2N0M0 NSCLC and poor lung function had radiotherapy dose escalated from 70 Gy in 29 to 70 Gy in 17 fractions reported 3 local relapses in 39 participants, median overall survival of 38.5 and median progression-free survival of 28.6 months [16]. Our data do not indicate a large number of competing events (i.e. non-cancer deaths). While there is always significant uncertainty about the quality of death certification in patients with lung cancer, there seemed to be a reasonable match between the number who had developed progressive disease and the number who had died at 2 years. One would not expect prolonged survival in this group once relapse had been diagnosed. 5. Conclusion We did not see any indication of an improvement in outcome due to weekly gemcitabine 100 mg/m2 added to radical radiotherapy in this population. The rate of local relapse was lower, and the observed survival higher, in both groups than had been expected before the trial. Conflict of interest statement During the period of trial design and patient accrual, the Chief Investigator (AP) and a co-author (MPS) served on Advisory Boards with Lilly Oncology. Acknowledgements The study was funded by a Grant based on patient accrual from Lilly Oncology to the University of Edinburgh. The manuscript was reviewed by Lilly Oncology before submission. Our thanks to all participants who consented to take part in this trial.
In addition to the affiliated centres of the authors of the manuscript, participants were entered by the investigators at the following centres: Dr. K. Fife, Dr. S. Old, Addenbrooke’s Hospital, Cambridge; Dr. A. Lamont, Southend Hospital; Dr. P. Burt, Christie Hospital, Manchester; Professor M. Brada, Royal Marsden Hospital, Sutton; Dr. F. Little, Western General Hospital, Edinburgh; Dr. P. Mulvenna, Dr. P. Atherton, Newcastle General Hospital; Dr. V. Laurence, Poole Hospital; Dr. E. Toy, Torbay District Hospital; Dr. N. Shah, Mount Vernon Hospital, Northwood; Dr. S. Upadhyay, Diana Princess of Wales Hospital, Grimsby; Dr. Samanchi, New Cross Hospital, Wolverhampton; Dr. C. Martin, Norfolk & Norwich Hospital; Dr. P. Wells, St. Bartholomew’s Hospital, London; Dr. A. Chetiwardena, Queen Elizabeth Hospital, Birmingham; Dr. D. Bissett, Aberdeen Royal Infirmary; Dr. J. Simpson, Royal Sussex County Hospital, Brighton; Dr. S. Beesley, Kent Oncology Centre, Maidstone; Dr. M. Illsley, Royal Surrey County Hospital, Guildford; Dr. J. Littler, Clatterbridge Centre for Oncology, Liverpool; Dr. M. Hocking, Walsgrave Hospital, Coventry; Dr. Z. Mallik, Whiston Hospital, St. Helens. Our thanks also to the trial coordinators Bryn Dixon, Tracy McEleney, and Leanne Ferrigan. References [1] Eisbruch A, Shewach DS, Bradford CR, Littles JF, Teknos TN, Chepeha DB, et al. Radiation concurrent with gemcitabine for locally advanced head and neck cancer: a phase I trial and intracellular drug incorporation study. J Clin Oncol 2001;19:792–9. [2] Shewach DS, Lawrence TS. Antimetabolite radiosensitizers. J Clin Oncol 2007;25:4043–50. [3] Scalliet P, Goor C, Galdermans D, et al. Gemzar (gemcitabine) with thoracic radiotherapy – a phase II pilot study in chemonaive patients with advanced non-small cell lung cancer (NSCLC). Proc ASCO 1998;17:499a. [4] van Putten JW, Price A, van der Leest AH, Gregor A, Little FA, Groen HJ. A phase I study of gemcitabine with concurrent radiotherapy in stage III, locally advanced non-small cell lung cancer. Clin Cancer Res 2003 Jul;9(7):2472–7. [5] Whitson BA, Groth SS, Duval SJ, Swanson SJ, Maddaus MA. Surgery for earlystage non-small cell lung cancer: a systematic review of the video-assisted thoracoscopic surgery versus thoracotomy approaches to lobectomy. Ann Thorac Surg 2008;86:2008–16. [6] Rowell NP, Williams CJ. Radical radiotherapy for stage I/II non-small cell lung cancer in patients not sufficiently fit for or declining surgery (medically inoperable): a systematic review. Thorax 2001;56:628–38. [7] British Thoracic Society. Guidelines on the selection of patients with lung cancer for surgery. Thorax 2001;56:89–98. [8] Price A, Dixon B, Erridge SC, Mohammed N. GRiN: a trial and tribulation in respiratory radiotherapy research. Clin Oncol (R Coll Radiol) 2005;17:328–31. [9] Hatton M, Nankivell M, Lyn E, Falk S, Pugh C, Navani N, et al. Induction chemotherapy and continuous hyperfractionated accelerated radiotherapy (CHART) for patients with locally advanced inoperable non-small-cell lung cancer: the MRC INCH randomized trial. Int J Radiat Oncol Biol Phys 2010;(October). [10] Belderbos J, Uitterhoeve L, van Zandwijk N, Belderbos H, Rodrigus P, van de Vaart P, et al. EORTC LCG and RT group randomised trial of sequential versus concurrent chemo-radiotherapy in patients with inoperable non-small cell lung cancer (EORTC 08972-22973). Eur J Cancer 2007;43:114–21. [11] Belani CP, Wang W, Johnson DH, Wagner H, Schiller J, Veeder M, et al. Eastern Cooperative Oncology Group Phase III study of the Eastern Cooperative Oncology Group (ECOG 2597): induction chemotherapy followed by either standard thoracic radiotherapy or hyperfractionated accelerated radiotherapy for patients with unresectable stage IIIA and B non-small-cell lung cancer. J Clin Oncol 2005;23:3760–7. [12] Wouters A, Pauwels B, Lardon F, Pattyn GG, Lambrechts HA, Baay M, et al. In vitro study on the schedule-dependency of the interaction between pemetrexed, gemcitabine and irradiation in non-small cell lung cancer and head and neck cancer cells. BMC Cancer 2010;10:441. [13] Milas L, Fujii T, Hunter N, Elshaikh M, Mason K, Plunkett W, et al. Enhancement of tumor radioresponse in vivo by gemcitabine. Cancer Res 1999;59:107–14. [14] Trodella L, D’Angelillo RM, Ramella S, Cellini F, Ciresa M, Cesario A, et al. Chemoradiotherapy in non-small cell lung cancer: the role of gemcitabine. Ann Oncol 2006;17(Suppl. 5):v52–4. [15] Haasbeek CJ, Lagerwaard FJ, Antonisse ME, Slotman BJ, Senan S. Stage I nonsmall cell lung cancer in patients aged > or =75 years: outcomes after stereotactic radiotherapy. Cancer 2010;116:406–14. [16] Bogart JA, Hodgson L, Seagren SL, Blackstock AW, Wang X, Lenox R, et al. Phase I study of accelerated conformal radiotherapy for stage I non-small-cell lung cancer in patients with pulmonary dysfunction: CALGB 39904. J Clin Oncol 2010;28:202–6.