Management and Survival Trends in Advanced Colorectal Cancer

Clinical Oncology (2008) 20: 626e630 doi:10.1016/j.clon.2008.04.014

Original Article

Management and Survival Trends in Advanced Colorectal Cancer1 T. Price*y, K. Pittman*, W. Patterson*y, M. Colbeck*, N. Rieger*, P. Hewett*, D. Rodda*, A. Townsend*, G. Maddern*, C. Lukez, D. Roderx *The Queen Elizabeth Hospital, Woodville, Australia; yLyell McEwin Hospital, Elizabeth Downs, Australia; zDepartment of Health, SA, Australia; xCancer Council, SA, Australia

ABSTRACT: Aims: Significant improvements in the outcome for patients with advanced colorectal cancer (CRC) have been achieved. The median survival for advanced CRC reported in clinical trials now approaches 2 years, but there is often a question as to whether this partly represents patient selection. We aimed to explore whether the availability of new chemotherapy drugs (irinotecan and oxaliplatin) and surgical advances have affected survival in a normal clinical setting. Materials and methods: A review of the Queen Elizabeth and Lyell McEwin health service prospective CRC database from 1992 to 2004 was carried out to assess outcome differences between two time cohorts (1 January 1992e31 December 1997 and 1 January 1998e31 December 2004). Results: For all patients (n [ 744) overall survival was seen to improve over time and is maintained out to 5 years. There have been a number of trends over time (1992e1997 vs 1998e2004) that have probably contributed to this gain; increased overall chemotherapy use (33% vs 43%); use of combination chemotherapy (i.e. oxaliplatin and irinotecan regimens); increased hepatic resection rates (1.9% vs 10.8%) and increased clinical trial uptake (0.6% vs 14.5%). Conclusion: This current analysis confirms an improvement in survival over time for advanced CRC and this is seen in unselected patients including those over 70 years of age. Price, T. et al. (2008). Clinical Oncology 20, 626—630 ª 2008 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Chemotherapy, clinical trials, colorectal cancer, elderly, hepatic surgery, survival

Introduction Significant improvements in the outcome for patients with advanced colorectal cancer (CRC) have been achieved over the last decade. Median survival as reported in clinical trials has improved from 5 months with ‘best supportive care’ alone [1] to 12.1 months in the era of 5-fluorouracil (5-FU) alone [2] and 19.5e21.5 months when new, noncross-resistant agents are used [3,4]. Five-year survivals of O30% are also reported after resection of hepatic and lung metastases [5]. There is, however, a trend to younger patients being included in chemotherapy clinical trials with the median age frequently close to 60 years [3,4], suggesting a degree of patient selection. Although improvements in survival over the past decade have been reported in multiple clinical trials, there is a lack of published data on the outcome in the community oncology setting. Epidemiological data from the USA and Germany show an improvement in survival for all stages of CRC over the past

1 Presented in part at the 42nd annual meeting of the American Society of Clinical Oncology, Atlanta, Georgia, USA, 2e6 June 2006.

0936-6555/08/200626þ05 $35.00/0

decade [6], but this does not address the question of overall improvements in patients with metastatic disease. Our group has previously shown a survival gain over time for all patients diagnosed with CRC (5-year survival 48% in 1980e1986 vs 56% in 1995e2002) [7], but this was not specific for patients with metastatic CRC. We have reviewed the prospective CRC database at our institutions from 1992 to 2004 to explore whether the availability of new chemotherapy drugs (irinotecan and oxaliplatin) and surgical advances have affected survival for patients with metastatic CRC in the general population, including the elderly group (those over 70 years). 5-FU had been the only active agent until 1997. On the basis of phase II data indicating activity of irinotecan in 5-FU failure [8], the potential treatment options altered. Subsequently, a survival advantage was seen when irinotecan was compared with best supportive care alone [9]. As the landscape for advanced CRC changed as of 1997/1998 we took this time point to assess progress, before and after 31 December 1997. An initial analysis confirmed that there was a trend to improved outcomes [10]. Although irinotecan was available as of 1998, the use of chemotherapy and new agents was expected to increase with further time, particularly

ª 2008 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

ADVANCED COLORECTAL CANCER SURVIVAL

when irinotecan (1 May 2001 and 1 December 2004) and oxaliplatin (1 December 2001 and 1 May 2004) became fully reimbursed in Australia for second-line and, subsequently, frontline therapy, respectively. Therefore, we planned an updated analysis.

Materials and Methods The prospectively derived cancer registry database from the Queen Elizabeth and Lyell McEwin Hospitals (academic tertiary referral teaching hospitals) was analysed comparing two time cohorts, 1 January 1992e31 December 1997 and 1 January 1998e31 December 2004. The Queen Elizabeth Hospital is a 367 bed public hospital that services a population base of about 250 000 people, living primarily in the western suburbs of Adelaide, South Australia. The Lyell McEwin Health Service is a public hospital with 198 beds in the northern suburbs of Adelaide. The two hospitals service more than 700 000 people as of 30 June 2004, just over 50% of the state’s population. The Index of Relative Socio-economic Disadvantage score for the western and northern regions is lower than the state as a whole. Disease-specific survival was analysed from the date of diagnosis for stage D, and from the date of distant recurrence for stages A, B and C, with a date of censoring of live cases at 31 December 2004. Cases dying of other causes before 30 December 2004 were censored at the time of death. The KaplaneMeier product-limit estimate was used for univariate analyses and Cox proportional hazards regression for multivariable analyses.

Results The current analysis was of 744 patients; 1992e1997 ¼ 313, 1998e2004 ¼ 431. There were no meaningful differences in age or pathology distribution. The median age was 69 and 70 years, respectively, for the time points. There was, however, a greater proportion of initial diagnosis with metastatic disease in the 1998e2004 cohort (Table 1). Inclusion in clinical trials increased significantly over time, as did the rate of hepatic resection. For all patients, overall survival was seen to improve over time and is maintained out to 5 years (Fig. 1). Survival for the respective time periods was 47.6 and 54.9% at 12 months; 28.0 and 34.8% at 24 months; 18.9 and 23.0% at 36 months; 12.6 and 17.2% at 48 months; and 10.4 and 14.9% at 60 months. Cox proportional hazards regression indicated a lower risk of case fatality for 1998e2004 than for 1992e1997 cases (P ¼ 0.048) after adjusting for age measured in years (i.e. age at diagnosis for stage D and at the time of distant recurrence for stages AeC). The key predictors of case fatality were found to be period (i.e. 1992e1997/ 1998e2004), age, and stage of disease at the time of initial diagnosis. Patients treated in the period 1998e2004 had an 18% reduction of risk of death (Table 2). For the entire group, chemotherapy was seen to reduce the risk of death by 27% (95% confidence interval: 0.61e0.88), with the median survival improving from 7.9

627

Table 1 e Patient characteristics 1992e1997 [n (%)]

1998e2004 [n (%)]

Number Relapsed Duke’s A, B and C Metastatic/Duke’s D

314 101 (32%) 213 (68%)

434 73 (17%) 361 (83%)

Age (years) !50 50e59 60e69 70e79 O80

26 41 93 108 46

29 66 108 162 69

Pathology Well Moderate Poor

11 (3.5%) 170 (54.1%) 133 (42.4%)

10 (2.3%) 248 (57.1%) 176 (40.6%)

Chemotherapy use

104 (33%)

182 (42%)

Hepatic resection Relapsed A, B and C Metastatic Clinical trials

(8.3%) (13.1%) (29.6%) (34.4%) (14.6%)

(6.7%) (15.2%) (24.9%) (37.3%) (15.9%)

0 4 (1.9%)

9 (12.3%) 38 (10.5%)

18 (0.6%)

63 (14.5%)

months for all patients not receiving chemotherapy to 17.8 months if chemotherapy was received. Chemotherapy use was seen to increase over time (33% vs 43%, P ¼ 0.023) with greater use of newer agents, although this was more frequent in second-line therapy (Table 3). With time there has been an increase in the use of oxaliplatin-containing regimens compared with irinotecan when compared with a prior analysis [9]. Non-single-agent 5-FU chemotherapy

1992-1997 1998-2004 100

80

Median survival 1992-1997 11.2 months 1998-2004 14.3 months

60

40

20

0 Mths

0

6

12

18

24

30

36

42

48

54

60

92-97 98-04

313 431

210 295

149 237

116 177

88 150

74 118

59 99

51 85

39 74

37 70

33 64

Duration of follow-up *Date of censoring of live cases: December 31st, 2004

Fig. 1 e Survival from the time of diagnosis of distant colorectal cancer: time period 1992e1997 vs 1998e2004.

628

CLINICAL ONCOLOGY

(1998e2004). For patients aged 80 years and over (n ¼ 40 for 1992e1997 and n ¼ 67 for 1998e2004) there was still the indication of a trend for improved survival at 24 months, survival (standard error) being 18.6% (6.7%) and 26.4% (6.9%), respectively (P O 0.2). Overall survival trends are summarised in Table 4.

Table 2 e Relative risk (95% confidence interval) of case fatality from distant colorectal cancer (CRC) (The Queen Elizabeth Hospital, TQEH, 1992e2004*): multivariable proportional hazards regression Predictors

Relative risk

Age at diagnosis of distant CRC (years) Under 70 (reference) (n ¼ 359) 70e79 (n ¼ 270) 80þ (n ¼ 115)

1.00 1.20 (0.99, 1.45) 1.68 (1.30, 2.18)

Duke’s stage at diagnosis A (reference) (n ¼ 13) B (n ¼ 73) C (n ¼ 86) D (n ¼ 572)

1.00 1.59 (0.72, 3.52) 2.26 (1.04, 4.91) 2.26 (1.07, 4.79)

Time period 1992e1997 (reference) (n ¼ 313) 1998e2004 (n ¼ 431)

1.00 0.82 (0.69, 0.98)

Discussion Our data have shown that there are improvements in survival that are both time specific reflecting overall improvements in health care and also relate to improved chemotherapy and surgical intervention. This survival gain has occurred in an unselected group with a median age at least 10 years older than current clinical trials. Importantly, the improvement has been seen in the group as a whole and in patients who received chemotherapy. For patients receiving chemotherapy, these data compare favourably with those reported from the USA, where 48% of patients received biological agents and significantly more irinotecan and oxaliplatin in first line [11]. A median survival of 17 months was reported in this group where all patients received chemotherapy, compared with 17.8 months for our institution. There have been five major trends over time (1998e2004 vs 1992e1997) that have probably contributed to this survival gain: (1) increased overall chemotherapy use (43% vs 33%), including the elderly group; (2) increased use of combination chemotherapy (i.e. oxaliplatin and irinotecan regimens); (3) increased hepatic resection rates (10.8% vs 1.9%); (4) increased clinical trial enrolment (14.5% vs 0.6%); (5) improved staging represented by stage migration. Although not captured in the registry database, improvements in supportive or palliative care have also probably played a part in the improved outcome. Despite improved chemotherapy, the survival gain remains modest and possibly reflects the inability in Australia to routinely use irinotecan and/or oxaliplatin in first-line chemotherapy until May and December 2004, respectively, although both drugs had been available for second-line use

*Date of censoring: 31 December 2004.

use increased over time from 23 to 29% in the current analysis. Analysis of patients who had received chemotherapy indicated an improvement in survival for patients commencing therapy after 1 January 1998, with 1- and 2-year survivals improving from 59.9% and 34.4% to 68.8% and 42.4%, respectively (Table 4). Patients with advanced rectal cancer were more likely to receive chemotherapy in both time points. Patients with recurrence after initial Duke’s A, B or C pathology had a median survival of 17.5 months (from diagnosis of metastatic disease) compared with 11.4 months if metastatic at first presentation (Fig. 2). Although an upward trend in survival was recorded for all ages, it was most pronounced for 70e79 year olds (n ¼ 272), where the increase in 24-month survival was from 21.1% for 1992e1997 to 36.1% for 1998e2004 (P ¼ 0.015). This corresponded to a significant increased rate of chemotherapy for this age group (P ¼ 0.042), increasing from 24.5% (1992e1997) to 36.1%

Table 3 e Chemotherapy trends: 1992e1997 vs 1998e2004 1992e1997

Chemotherapy regimen 5-fluorouracil  leuocovorin Floxuridine (FUDr) Mitomycin C Mitomycin C þ fluoropyrimidine Irinotecan  fluoropyrimidine Capecitabine Oxaliplatin þ fluoropyrimidine Other/unknown

1998e2004

First line

Second/third line

First line

Second/third line

(n ¼ 103)

(n ¼ 7)

(n ¼ 183)

(n ¼ 57)

90.3% 1% 2.9% 0 0 1% 0 4.8%

57% 0 14.3% 0 14.3% 0 0 14.3%

53.6% 0 2.2% 8.2% 7.7% 8.7% 14.2% 5.4%

10.5% 0 3.5% 1.7% 28% 22.8% 21% 7%

ADVANCED COLORECTAL CANCER SURVIVAL Table 4 e Survival over time: 1992e1997 vs 1998e2004 1992e1997

1998e2004

Median overall survival 1-year survival 2-year survival

11.2 months 47.6% 28%

14.3 months 54.9% 34.8%

Chemotherapy Median 1-year survival 2-year survival

16.9 months 59.9% 34.4%

20.8 months 68.8% 42.4%

since May and December 2001, respectively. Survival gains, as seen in clinical trials, have reflected use of all three active drugs [12] and the delay in availability in Australia will probably have affected the magnitude of gain. Despite this, the median survival of 14.8 months seen in this group (1998e2004) is not dissimilar to that seen in early combination chemotherapy trials when irinotecan/5-FU/ leucovorin was first reported to show a survival gain over 5-FU alone [13]. Further gains should occur with access to biological agents. The number of patients receiving chemotherapy remains relatively low. Age and comorbidities, as well as patient wishes, will affect this figure, but rates of under 50% for advanced CRC seem low and improved education of patients and clinicians may assist in increasing access to active therapy. In addition, the increase in chemotherapy use was primarily in the older age group associated with a survival gain, although comorbidity and tolerance factors probably limited the improvement in overall survival to a degree. These data do, however, support a similar gain in

Duke's A, B & C-recurrence Duke's D

100

80

Median Survival Duke’s A-C 17.5 months Duke’s D 11.4 months

60

40

the elderly population and, thus, can be used when discussing therapy options in the palliative setting. Resection of hepatic metastases leads to 5-year survival of 30e40% [5] and has thus significantly contributed to improvements in survival for this patient group. Hepatic resection has now become an accepted and important component of multidisciplinary management of metastatic CRC. In this unselected group, 10.8% of patients underwent hepatic resection in the later cohort, indicating that there has been an appropriate increase in referral to hepatic surgeons. The improved overall 3-, 4- and 5-year survival seen in this data set probably reflects the increased rate of hepatic surgery seen over time. There is supportive evidence to indicate that patients treated in clinical trial programmes seem to have an improved outlook [14] and these data indicate that access to clinical trials has increased over time. Clinical trial participation directly improving outcome remains controversial, with a recent analysis suggestive of insufficient data to make this conclusion [15]. For our population, clinical trial participation probably allowed earlier access to firstline therapy with irinotecan and oxaliplatin and it may have been this factor that had a greater direct affect on survival. Although not specifically measured as part of the database, the increased use of multi-slice computed tomography, magnetic resonance imaging and positron emission tomography will probably lead to upstaging of early disease, reflected by the greater percentage of Duke’s D patients in the later time period. Although this study focussed on improvements in outcome over time, our data also suggested a poorer outcome overall when comparing patients with a diagnosis of metastatic disease at diagnosis with patients who recurred after previous surgery with curative intent (Fig. 2). This may reflect a difference in the biology of these groups or potentially earlier diagnosis due to surveillance. This finding does, however, differ to two recent reports where patients with relapse did worse overall. Koeller et al. [11] reported survival of 19 months for patients with newly diagnosed metastatic disease compared with 14.3 months for recurrence. Michael et al. [16] showed that patients who had had previous adjuvant chemotherapy and subsequently had a recurrence had a significantly poorer progression-free survival.

Conclusion

20

0 0

6

12

18

24

30

36

42

48

54

Duke`s A-C

172

126

105

82

66

53

42

37

26

24

21

Duke`s D

572

381

280

212

170

139

116

99

86

81

74

Mths

629

60

Duration of follow-up *Date of censoring of live cases: December 31st 2004

Fig. 2 e Survival from the time of diagnosis of distant colorectal cancer: Duke’s A, B and C recurrence vs Duke’s D.

Clinical trials have shown improvements in survival for highly selected patients. This current analysis confirms an improvement in survival over time for advanced CRC and this was seen in unselected patients, including the elderly (those aged over 70 years). In the community setting, patients have been able to access active agents within the guidelines reflected by the period and have had improved access to clinical trials. These data suggest that a number of factors have contributed to the trend of improved survival and these include increased use of chemotherapy, new chemotherapeutic agents, hepatic resection and

630

CLINICAL ONCOLOGY

staging methods. Increased clinical trial participation has probably also affected survival gains. Author for correspondence: T. Price, Department of Oncology, The Queen Elizabeth Hospital, Woodville Road, Woodville, South Australia 5011, Australia. Tel: þ61-8-82228429; Fax: þ61-882227054; E-mail: [email protected] Received 10 January 2008; received in revised form 31 March 2008; accepted 1 April 2008

References 1 Scheithauer W, Rosen H, Kornek GV, et al. Randomised comparison of combination chemotherapy plus supportive care with supportive care alone in patients with metastatic colorectal cancer. Br Med J 1993;306(6880):752e755. 2 Meta-analysis Group in Cancer. Efficacy of intravenous continuous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. J Clin Oncol 1998;16(1):301e308. 3 Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004;22(1):23e30. 4 Tournigand C, Andre T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004;22(2):229e237. 5 Simmonds PC, Primrose JN, Colquitt JL, et al. Surgical resection of hepatic metastases from colorectal cancer: a systematic review of published studies. Br J Cancer 2006;94(7):982e999. 6 Gondos A, Arndt V, Holleczek B, et al. Cancer survival in Germany and the United States at the beginning of the 21st century: an up-to-date comparison by period analysis. Int J Cancer 2007;121(2):395e400.

7 Luke CG, Koczwara B, Moore JE, et al. Treatment and survival from colorectal cancer: the experience of patients at South Australian teaching hospitals between 1980 and 2002. Clin Oncol 2005;17:372e381. 8 Rothenberg ML, Eckardt JR, Kuhn JG, et al. Phase II trial of irinotecan in patients with progressive or rapidly recurrent colorectal cancer. J Clin Oncol 1996;14(4):1128e1135. 9 Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998;352(9138):1413e1418. 10 Price TJ, Pittman K, Patterson K, et al. Survival and treatment trends for advanced colorectal cancer treatment in a university hospital, 1992e2001. J Clin Oncol 2004;22(Suppl):3707a. 11 Koeller JM, Kwan P, McKibbin T, et al. Advanced colorectal cancer treatment and outcome in the community oncology setting. J Clin Oncol 2006;20(24):16040a. 12 Grothey A, Sargent D, Goldberg RM, et al. Survival of patients with advanced colorectal cancer improves with the availability of fluorouracileleucovorin, irinotecan, and oxaliplatin in the course of treatment. J Clin Oncol 2004;22(7):1209e1214. 13 Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000;343(13):905e914. 14 Hebert-Croteau N, Brisson J, Lemaire J, et al. The benefit of participating to clinical research. Br J Cancer Res Treat 2005; 91(3):279e281. 15 Peppercorn J, Weeks J, Cook EF, et al. Comparison of outcomes in cancer patients within and outside clinical trials: conceptual framework and structured review. Lancet 2004; 363:263e270. 16 Michael M, Goldstein D, Clarke SJ, et al. Prognostic factors predictive of response and survival to a modified FOLFOX regimen: importance of an increased neutrophil count. Clin Colorectal Cancer 2006;6:297e304.