- Email: [email protected]
Prognostic impact of increasing age and co-morbidity in cancer patients: A population-based approach
Critical Reviews in Oncology/Hematology 55 (2005) 231–240
Prognostic impact of increasing age and co-morbidity in cancer patients: A population-based approach Maryska L.G. Janssen-Heijnen a,∗ , Saskia Houterman a , Valery E.P.P. Lemmens a , Marieke W.J. Louwman a , Huub A.A.M. Maas b , Jan Willem W. Coebergh a,c a
Eindhoven Cancer Registry, Comprehensive Cancer Centre South, P.O. Box 231, 5600 AE Eindhoven, The Netherlands b Department of Clinical Geriatrics, Tweesteden Hospital, Tilburg, The Netherlands c Department of Public Health, University Medical Centre Rotterdam, Rotterdam, The Netherlands Accepted 22 April 2005
Contents 1. 2. 3.
4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Prevalence of co-morbidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Treatment (see overview in Table 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Complications of treatment (Fig. 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Survival (Table 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Validity of data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Prevalence of co-morbidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Treatment and complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reviewers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
232 232 233 233 233 233 237 237 237 237 237 238 238 238 239 240
Abstract This large population-based study focuses on the prognostic role of increasing age and co-morbidity in cancer patients diagnosed in the southern Netherlands. Data of patients diagnosed between 1995 and 2002 and recorded in the population-based Eindhoven Cancer Registry were used. Older patients (with serious co-morbidity) with non-small cell lung cancer or prostate cancer underwent surgery less often than younger patients. Elderly with stage III colon cancer, small cell lung cancer, FIGO II or III ovarian cancer or non-Hodgkin’s lymphoma (NHL) received (adjuvant) chemotherapy less often, probably because of the higher rate of haematological complications. Administration of adjuvant radiotherapy decreased with age and co-morbidity in patients with rectal cancer, limited small cell lung cancer or breast cancer. In general, elderly did not suffer from more complications than younger patients, except for cardiac complications (colorectal cancer and NHL) and postoperative death (non-small cell lung cancer). For most tumours relative survival was lower for the elderly, except for patients with colon cancer, prostate cancer or indolent NHL. Co-morbidity had an independent prognostic effect, except for tumours with a very poor prognosis. Future prospective studies should investigate whether the guidelines for cancer treatment should be adjusted for elderly with serious co-morbidity. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Elderly; Co-morbidity; Cancer; Treatment; Complications of treatment; Survival
∗
Corresponding author. Tel.: +31 40 2971616; fax: +31 40 2971610. E-mail address: [email protected] (M.L.G. Janssen-Heijnen).
1040-8428/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.critrevonc.2005.04.008
232
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
1. Introduction Due to ageing of the population and the rising incidence rates of most cancers with age, the mean age of patients diagnosed with cancer is increasing in western countries. This implies that they increasingly suffer from one or more other serious (chronic) diseases. Besides affecting the life expectancy, co-morbid conditions may complicate the treatment of cancer patients, especially when they are frail [1–3]. Since elderly patients are often excluded from clinical trials, little is known about treatment outcome, such as complications, quality of life and survival. Although the number of trials that include elderly is increasing, generally only relatively healthy elderly are included. This means that trial results may not be valid for many older cancer patients. This study focuses on the prognostic role of increasing age and co-morbidity in cancer patients diagnosed in the registration area of the population-based Eindhoven Cancer Registry.
2. Methods The Eindhoven Cancer Registry records data on all patients newly diagnosed with cancer in the southern part of the Netherlands, an area with 2.3 million inhabitants and only community hospitals [4]. Since 1993, serious co-morbidity with prognostic impact has been recorded for all patients. The Charlson co-morbidity index is most widely used for recording co-morbidity and was validated in various studies [5]. We used a slightly modified version of this index for recording co-morbidity (Table 1). Co-morbidity was defined as life-shortening diseases that were present at the time of cancer diagnosis. Table 1 Classification of co-morbidity, according to a modified version of the list of Charlson et al. [5] Chronic obstructive pulmonary disease (COPD) Cardiovascular disease: myocardial infarction, cardiac insufficiency, angina pectoris, coronary artery bypass graft (CABG) Peripheral arterial disease: intermittent claudication, abdominal aneurysm, surgical intervention Cerebrovascular diseases (cerebrovascular accident, hemiplegia) Other malignancies (except basal cell skin carcinoma) Hypertension Diabetes mellitus Other Autoimmune diseases: sarcoidosis, Wegener’s disease, systemic lupus erythematosis (SLE) Rheumatoid arthritis (only severe) Kidney diseases: glomerulonephritis, pyelonephritis Gastrointestinal: stomach ulcer and resection, colitis Liver diseases: cirrhosis, hepatitis Dementia Chronic infections
Trained registry personnel extracted all data from the medical records between 6 and 18 months after diagnosis. The medical record is generally regarded as the most complete source of information on the patient’s past and current health status [6]. Patients with cancer of the esophagus, stomach, colon or rectum, pancreas, lung, breast, cervix uteri, corpus uteri, ovary, prostate, bladder, kidney and non-Hodgkin’s lymphoma, newly diagnosed between 1995 and 2002 (N = 43,111), were included for this overview. Patients with cancer diagnosed at autopsy (N = 661 or 1.5%) were excluded. Treatment was classified as surgery (resection), radiotherapy, chemotherapy, endocrine therapy (or combinations of the aforementioned treatment options) and ‘other or none’. Surgery did not comprise diagnostic operations. For a random sample of 1686 patients with cancer of the colon/rectum, lung, breast, prostate and non-Hodgkin’s lymphoma additional information about grade 3 or 4 complications after treatment (according to the NCI CTC-toxicity criteria) within 1 year after diagnosis were gathered from the medical records. Survival analyses were restricted to patients with cancer of the colon or rectum, lung, breast, ovary, prostate or nonHodgkin’s lymphoma newly diagnosed between 1995 and 2001 (N = 29,249). Follow-up was completed until 1 January 2004. In addition to passive follow-up via the hospitals, this information was also obtained from the municipal registries in the area of the Eindhoven Cancer Registry and the Central Bureau for Genealogy. The latter is an institution that collects data on all deceased Dutch citizens via the civil municipal registries. In this way, information on patients who had moved outside the registry area was also obtained. Patients who died outside the Netherlands were wrongly considered as ‘being alive’. However, the estimated proportion of these patients was less than 0.3%. Survival time was defined as the time from diagnosis to death or the end of the study. Survival generally decreases with age, because other causes of death also take their share. Therefore, for describing the prognostic effect of age, we calculated relative survival rates for each age group, which are an estimation of disease-specific survival. In this way, survival of cancer patients is adjusted for mortality from all causes of death in the background population with the same age structure [7]. Mortality in the background population was estimated from life tables for regional male and female populations. The independent prognostic effect of the number of co-morbid conditions was estimated with a multivariable Cox regression model. The hazard rates for death were adjusted for age, gender, subsite (only for colorectal cancer), histology (only for NSCLC), stage and treatment. The SAS computer package (Version 8.2) was used for all statistical analyses (SAS Institute Inc., Cary, North Carolina, USA, 1999).
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
3. Results 3.1. Prevalence of co-morbidity The prevalence of co-morbidity usually increased with age (Table 2), but remained stable or decreased above age 80 for registered patients with pancreatic or lung cancer. About 60% of all new cancer patients older than 65 suffered from at least one other serious disease. The most frequent concomitant diseases were previous cancers, heart disease, hypertension, COPD and diabetes mellitus, with prevalence rates up to 20, 23, 26, 17 and 16%, respectively. The prevalence of cardiovascular diseases was higher among men compared to women, and was highest for patients with cancer of the digestive tract, lung and kidney and non-Hodgkin’s lymphoma. The prevalence of COPD was relatively high among patients with lung cancer, and also among men with esophageal or bladder cancer. The prevalence of hypertension was highest among women with cancer of the corpus uteri or kidney. High prevalence rates of diabetes in older patients were observed for cancer of the pancreas, cervix uteri, corpus uteri and kidney. 3.2. Treatment (see overview in Table 3) Patients with colon cancer underwent surgery regardless of age or the number of co-morbid conditions: more than 95% with stages I–III disease and about 75% of patients with stage IV disease (Table 3). However, patients with stage III disease received adjuvant chemotherapy less frequently with the rise of age: from 82% of patients in age group 50–64 years to 42% of patients aged 65–79 years (P < 0.01, data 1997–2002). The proportion of patients with stage III disease aged 50–79 years receiving adjuvant chemotherapy also decreased from 69% of patients without co-morbidity to 53% of those with co-morbidity (P < 0.01, data 1997–2002). The proportion of rectal cancer patients stages I–III receiving adjuvant radiotherapy decreased from 57% of patients younger than 65 years to 36% of patients aged 65 years or older (P < 0.01), and also decreased with co-morbidity from 50% of patients without co-morbidity to 39% in case of co-morbidity (P < 0.01). The proportion of patients with localised non-small cell lung cancer who underwent surgery with or without radiotherapy was only 11% of those aged 80 years or older versus 91%, 79% and 59% of the age groups <60, 60–69 and 70–79 years, respectively (P < 0.01). Patients aged 60–69 and 70–79 years also underwent surgery less often when co-morbidity was present (P < 0.01 and P = 0.01, respectively). Most patients with non-localised non-small cell lung cancer received only radiotherapy (36%). The proportion receiving chemotherapy (with or without radiotherapy) was considerably higher among patients younger than 60 years (28%) than among those aged 80 years or older (2%) (P < 0.01). The number of co-morbid conditions had no substantial influence on treatment chosen for patients with
233
non-localised disease. Elderly patients with limited small cell lung cancer received combined chemoradiation less often and chemotherapy alone more often (P < 0.01). Among patients with limited small cell lung cancer aged 70–79 years the proportion receiving adjuvant radiotherapy also decreased in the presence of co-morbidity (P = 0.05). Among patients with breast cancer younger than 80 years over 95% underwent surgery, compared with only 76% of those aged 80 years or older (P < 0.01). Among those undergoing surgery, the proportion of breast conserving surgery decreased from 54% of those younger than 60 years to 29% of those aged 80 years or older (P < 0.01). Axilliary dissection for those undergoing breast conserving surgery decreased in the presence of co-morbidity (78% of patients with at least two co-morbid conditions compared to 97% of those without co-morbidity, P < 0.01), as did adjuvant radiotherapy (from 94% of patients without co-morbidity to 78% of those with at least two co-morbid conditions, P < 0.01). Eighty-three percent of patients younger than 70 years with FIGO II or III ovarian cancer received the recommended treatment (combination of surgery and chemotherapy): 87% of those without co-morbidity and 73% of those with at least one co-morbid condition (P < 0.01). In contrast, only 45% of the patients aged 70 years or older received the recommended treatment (54% of those without co-morbidity and only 38% of the patients with co-morbidity (P < 0.05). The proportion of prostate cancer patients undergoing prostatectomy decreased with increasing age, from 41% of patients younger than 60 years to 1% of patients aged 80 years or older (P < 0.01). The proportion of patients receiving curative radiotherapy also decreased from 30% of those younger than 60 years to 5% of those aged 80 years or older (P < 0.01). Endocrine treatment increased from 19% of patients younger than 60 years to 58% of patients aged 80 years or older. Among patients aged 60–79 years, the proportion who underwent prostatectomy also significantly decreased with co-morbidity, whereas the proportion receiving radiotherapy or endocrine treatment increased (P < 0.01). For patients with indolent non-Hodgkin’s lymphoma (NHL) the proportion receiving chemotherapy did not decrease with age. For patients with aggressive NHL, the proportion receiving chemotherapy decreased from 88% of those aged 60 years or younger to 68% of those older than 60 years (P < 0.01). Among patients with aggressive disease the proportion receiving chemotherapy also decreased with co-morbidity (P < 0.05). 3.3. Complications of treatment (Fig. 1) Thirty-seven percent of resected colorectal cancer patients between 40 and 69 years old suffered from any complication within 1 year of diagnosis compared to 44% of patients aged 70 years or older (P = 0.08) (Fig. 1a). The most frequent complications were major infections (11%), minor infections (11%), typical irradiation side
234
Table 2 Age-specific prevalence (%) of the most common serious concomitant diseasesa among newly diagnosed patients with 13 major cancers, 1995–2002 Sex
N
Any co-morbidity (%)
Previous cancers (%)
Heart and vascular disease (%)
COPD (%)
Hypertension (%)
Diabetes (%)
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
Esophagus
M F
602 258
51 43
72 53
74 70
6 11
17 15
26 15
14 5
36 17
28 13
12 14
17 10
24 10
15 13
21 19
10 23
5 5
11 12
18 12
Stomach
M F
1310 721
50 41
67 58
72 68
6 11
12 14
20 16
19 8
33 16
25 25
8 3
13 9
16 7
15 13
15 26
18 31
6 6
10 13
12 20
Colon/rectum
M F
4502 3992
40 37
64 56
71 65
7 10
15 14
22 16
13 5
28 14
32 24
6 5
15 8
15 9
16 15
21 25
15 25
7 5
10 14
13 17
Pancreas
M F
551 518
49 42
67 63
67 66
7 8
12 9
15 8
21 8
28 14
33 24
5 3
17 8
15 6
13 15
18 26
21 29
13 10
21 26
23 25
Lung
M F
6834 1816
53 49
74 67
72 61
9 9
16 16
16 14
18 11
34 22
32 25
20 21
24 24
31 16
11 12
15 21
11 22
7 6
10 12
11 11
Breast
F
7978
28
52
67
6
10
15
4
12
22
4
6
8
13
29
27
5
13
16
Cervix uteri
F
289
33
51
62
6
8
12
5
14
21
4
6
2
13
21
26
8
14
31
Corpus uteri
F
1308
37
64
77
8
10
20
3
13
24
3
6
12
22
38
33
8
22
25
Ovary
F
1011
34
58
65
12
15
15
4
14
20
5
7
6
12
25
27
4
12
11
Prostate
M
6272
36
56
59
6
9
14
12
24
27
5
12
15
12
17
12
4
8
9
Bladder
M F
1646 449
47 49
70 60
73 64
18 23
24 23
29 20
16 8
29 13
27 23
8 8
15 8
20 8
9 10
16 29
13 20
5 6
11 16
11 16
Kidney
M F
725 534
46 46
69 69
82 76
11 7
19 18
26 18
14 6
31 18
39 27
5 8
15 7
16 5
16 25
24 32
21 48
8 8
10 19
21 21
NHL
M F
792 685
39 32
67 61
76 65
5 5
14 16
19 13
13 5
27 11
39 23
4 2
15 8
11 4
13 14
17 26
16 20
5 6
9 12
11 13
Source: Eindhoven Cancer Registry. a More conditions per patient possible.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
Tumour
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
effects, such as radiation enteritis (9% of patients receiving radiotherapy) and typical chemotherapy side effects, such as neutropenia, thrombocytopenia, diarrhea and soreness of the mouth (10% of patients aged 40–69 years receiving chemotherapy). Elderly patients suffered more from cardiac complications than patients aged 40–69 years (11% versus 5%, P < 0.01), especially those with cardiovascular co-morbidity. Forty-nine percent of patients with localised NSCLC who underwent resection suffered from one or more
235
complications within 1 year of diagnosis (no difference between both age groups, Fig. 1b). The most frequent postoperative complications were major infections (16%) and minor infections (11%). Postoperative death occurred in 5% of the patients. It occurred significantly more often in patients aged 70 years or older (11%) than in those younger than 70 years (2%) (P < 0.01). Seventy-seven percent of patients with SCLC receiving chemotherapy suffered from at least one complication, mainly haematological complications (57%) or other complications due to chemotherapy
Fig. 1. Age-specific prevalence of complications during the first year after diagnosis (1995–1999), (a) postoperative complications colorectal cancer, (b) postoperative complications localised NSCLC, (c) complications after chemotherapy (±RT) SCLC, (d) postoperative complications breast cancer, (e) complications after radiotherapy prostate cancer and (f) complications after chemotherapy aggressive NHL.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
236
Table 3 Influence of age and co-morbidity on primary treatment, according to tumour type and/or stage Tumour
Stage
Influence of rising age
Influence of co-morbidity
Colon
Stage I/II Stage III
None Less adjuvant CT
None Age 65–79 years: less adjuvant CT
Rectum
Stage II/III
Less adjuvant RT
Less adjuvant RT
NSCLCa
Stage I/II Stage III/IV
Less surgery, more RT alone Less CT
Age 60–79 years: less surgery None
SCLCb
Limited Extensive
Less CT + RT, more CT alone Less CT, more abstinence
Age 70–79 years: less CT + RT None
Less surgery, more endocrine
Less adjuvant RT, more endocrine
Less surgery + CT
Less surgery + CT
Less prostatectomy, more endocrine
Age 60–79 years: less prostatectomy, more endocrine
Less CT, more RT and wait and see Less CT, more abstinence
None Age 70+ years: less CT
Breast Ovary
FIGO II/III
Prostate NHLc
Indolent Aggressive
Source: Eindhoven Cancer Registry; CT = chemotherapy; RT = radiotherapy. a Non-small cell lung cancer. b Small cell lung cancer. c Non-Hodgkin’s lymphoma.
(42%) (Fig. 1c). The proportion of complications did not differ between the two age groups. The number of breast cancer patients with complications after surgery was similar for both age groups (27%) (Fig. 1d). Minor infection (11%) was the most common complication, followed by lymphoedema (7% of those who underwent axillary dissection). In prostate cancer patients undergoing radical prostatectomy, 38% suffered a complication during the first year after diagnosis. The most common complication was infection (14% in those younger than 70 years and 25% in those aged 70 years or older). In patients younger than 70 years,
26% had other complications like strictures. Among those receiving radiotherapy the proportion of patients with a complication was about 20% (Fig. 1e). Most of these complications were radiotherapy-related complications, such as proctitis. The proportion of patients with aggressive NHL who developed any complication after chemotherapy was 63% for patients aged 40–60 years and 69% for those aged 61 years or older (P = 0.4, Fig. 1f). The most common complications were haematological (33%). Cardiovascular complications were more common among the elderly (21% versus 11%, P = 0.06). Three percent of patients died due to treatment
Table 4 Relative 2- and 5-year survival rates (% and 95% CI), according to age and tumour site Tumour type
Relative survival (% (95% CI)) 2-year <70 years
5-year <70 years
70+ years
Colon
70 (68–72)
70+ years 66 (64–68)
57 (55–60)
59 (56–62)
Rectum
76 (73–78)
63 (60–67)
62 (59–65)
51 (47–55) 29 (24–34)
NSCLCa
localised
64 (61–68)
41 (37–46)
47 (43–51)
NSCLCa
non-localised
13 (11–14)
8 (7–10)
4 (3–6)
–
SCLCb
12 (10–14)
6 (4–8)
5 (4–7)
–
Breast
95 (94–96)
91 (90–93)
86 (85–87)
81 (78–84)
Ovary
67 (62–71)
39 (34–45)
45 (40–50)
25 (20–31)
Prostate
100 (98–100)
99 (97–101)
95 (93–97)
95 (91–98)
NHLc indolent
89 (84–93)
71 (61–80)
73 (66–80)
65 (51–78)
NHLc aggressive
63 (58–68)
43 (37–50)
51 (45–57)
36 (28–45)
Source: Eindhoven Cancer Registry. a Non-small cell lung cancer. b Small cell lung cancer. c Non-Hodgkin’s lymphoma.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
(1% of those younger than 61 years and 4% of the elderly, P = 0.4). 3.4. Survival (Table 4) In general, the relative 5-year survival rate was lower for the elderly, except for patients with colon cancer, prostate cancer or indolent NHL (Table 4). For elderly patients with non-localised NSCLC or SCLC 5-year survival could not be estimated, because of the poor prognosis. For these patients relative 2-year survival was also lower for the elderly. In multivariable survival analyses, the independent prognostic effect of the number of co-morbid conditions was evaluated (no co-morbidity as reference), after adjustment for age, gender, subsite (only for colorectal cancer), histology (only for NSCLC), stage and treatment. The hazard ratio for death for one and two or more co-morbid conditions for patients with colon cancer was 1.2 (95% CI 1.1–1.3) and 1.4 (95% CI 1.2–1.5), respectively. For rectal cancer, these hazard ratios were 1.3 (95% CI 1.1–1.5) and 1.6 (95% CI 1.4–1.9), for breast cancer patients 1.3 (95% 1.2–1.5) and 1.4 (95% CI 1.3–1.5) and for localised prostate cancer 1.2 (95% CI 1.03–1.4) and 1.9 (95% CI 1.6–2.2). The hazard ratio for death for at least one co-morbid condition was 1.7 (95% CI 1.2–2.5) for patients with indolent NHL and 1.2 (95% CI 0.99–1.5) for patients with aggressive NHL. For patients with localised NSCLC, non-localised NSCLC, SCLC or ovarian cancer co-morbidity had no independent prognostic effect. 4. Discussion 4.1. Validity of data Between 2001 and 2003 the completeness and accuracy of our data on co-morbidity in the Eindhoven Cancer registry were validated in a random sample of 2607 patients with colorectal, lung, breast and prostate cancer and non-Hodgkin’s lymphoma aged 40 years and older and diagnosed between 1995 and 1999. Co-morbidity scored by the registry team was compared with that scored by a team of a retired surgeon and an epidemiologist. Some underregistration occurred, especially of cardiovascular conditions (internal reports, 2002). This appeared to be mainly due to the use of unknown terminology, unknown abbreviations or illegible handwriting of the specialist. This implies that the real effects of co-morbidity on treatment and survival might be stronger than those presented in this article. 4.2. Prevalence of co-morbidity The higher prevalence of co-morbidity among older patients was expected, because the prevalence of chronic diseases generally increases with age. The prevalence of comorbidity among older patients may even be underestimated due to ascertainment bias. Younger patients underwent
237
surgery more often and received chemotherapy more often. The prevalence of co-morbidity reported by the treating physician might then be more elevated among younger patients, due to the required screening examinations before treatment. The high risk of cardiovascular diseases and chronic obstructive pulmonary diseases for patients with cancer of the esophagus, stomach, lung, bladder and kidney can be explained by the high proportion of smokers among these patients, especially men [8,9]. It is not surprising that diabetes mellitus occurred in a high proportion of patients with cancer of the pancreas [10,11]. A history of diabetes has been consistently associated with a two-fold increased risk for endometrial cancer [12], probably because both are related with obesity. The risk of renal cancer is also elevated among patients with diabetes [13]. Renal cancer has also been associated with hypertension, although it is unknown whether this results from the hypertension itself or from the antihypertensives [14]. The prevalence of cardiovascular diseases and pulmonary diseases was higher among men compared to women, which can be largely explained by a higher prevalence of smoking among men in the past. By contrast, the prevalence of hypertension (a less serious condition) and diabetes was higher among women. 4.3. Treatment and complications If alternative treatment strategies were available, older patients (with co-morbidity) were often treated less aggressively than younger patients. When surgery is inevitable like in patients with colorectal cancer, higher age or the prevalence of co-morbidity did not have any influence on the resection rate. On the other hand, older patients with localised non-small cell lung cancer (with serious co-morbidity) more often received radiotherapy instead of surgery [15]. Surgical mortality increases markedly with age and is especially high for pneumonectomy [16,17]. The resection rate also declined with co-morbidity, probably because of the expected higher incidence of postoperative complications and mortality [18]. In accordance with recommendations in treatment guidelines for elderly patients with prostate cancer, elderly received more hormonal treatment and less prostatectomy or radiotherapy [19–21]. The frequency of administering adjuvant chemotherapy for patients with stage III colon cancer markedly decreased with rising age and co-morbidity [22], probably because of the higher rate of haematological complications [23–25]. For the same reason, administration of primary chemotherapy also decreased with age and co-morbidity for patients with non-Hodgkin’s lymphoma [26]. The lower proportion of elderly patients undergoing surgery or receiving chemotherapy also appeared in another area of the Netherlands [27]. In our study, administration of adjuvant radiotherapy decreased with age and co-morbidity in patients with
238
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
rectal cancer, limited small cell lung cancer or breast cancer [22,28]. However, we did not find a higher rate of complications due to radiotherapy for older patients with co-morbidity [20]. Therefore, the reluctance of accepting adjuvant radiotherapy might be related to practical reasons like the distance to a radiotherapy institute or the burden of the 20–30 visits to the radiotherapy institute or the availability of alternative treatment (such as breast amputation instead of breast conserving surgery with radiotherapy). Age seemed to have more influence on treatment chosen than co-morbidity. Apparently, co-morbidity alone does not entirely explain why elderly non-small cell lung cancer patients and prostate cancer patients underwent surgery less often and why those with colon cancer, rectal cancer, small cell lung cancer and breast cancer received less (adjuvant) chemotherapy or adjuvant radiotherapy. For patients with breast cancer, current guidelines recommend less chemotherapy for elderly breast cancer patients, because postmenopausal women with receptor positive breast cancer will get the highest benefit from hormonal adjuvant therapy and current guidelines for prostate cancer recommend radiotherapy or hormonal treatment for elderly patients [21,29]. However, performance status, the psychological condition of the patient, social factors and patient’s decision may also play a role, especially for the other tumours [3,30]. The fact that elderly often did not have more complications than younger patients might be explained by (appropriate) selection for treatment. This was not a clinical trial, but a population-based study in which physicians selected patients for treatment. Among patients with colorectal cancer or NHL, cardiac complications occurred more among the elderly, especially those with cardiovascular co-morbidity. This might also partly be explained by the fact that the occurrence of cardiac events in the general population increases with age. 4.4. Survival For most tumour types relative survival decreased with age, except for patients with colon cancer, prostate cancer or indolent NHL. Relative survival reflects disease-specific survival. Therefore, the decrease in relative survival with age for the other tumour types cannot be explained by a higher rate of death in the elderly due to other causes. Less intense treatment of either the tumour or the co-morbid condition might play a role. For patients with lung cancer co-morbidity had no independent prognostic effect [15]. This contradicts some other studies [31–34], which however were not populationbased. In the other studies also other scales for measuring co-morbidity were used: the Kaplan–Feinstein Index [35] and the Cumulative Illness Rating Scale-Geriatric (CIRS-G) [36]. In one of the studies, co-morbidity affected overall survival in surgically resected stage I NSCLC patients, when co-morbidity was rated according to CIRS-G, but
not according to the Charlson scale [31]. In another American study, co-morbidity count and the Charlson index were significant predictors for lung cancer survival, but only explained 2.5 and 2.0% of the survival variation, respectively [37]. We also found no independent prognostic effect of co-morbidity for patients with ovarian cancer. Probably, the influence of co-morbidity on survival is of less importance in the case of a lethal disease, such as lung cancer or ovarian cancer. Most of these patients die of the cancer, before they become at risk of dying from the co-morbid condition. This was confirmed in two recent American studies, in which the prognostic effect of co-morbidity was found to be smallest for tumours with a poor prognosis [38,39]. For the other tumours, co-morbidity had an independent prognostic effect. This negative influence of co-morbidity on survival of cancer might be due to several mechanisms: the increased risk of death due to the co-morbid condition itself, more contra-indications for anti-cancer treatment, more indications for dose reduction and a higher rate of treatment-related complications, such as infections and cardiovascular events. In several of our recent studies, the adverse effects of co-morbidity on survival appeared to be independent of treatment, so less aggressive treatment could not (fully) account for the observed differences in survival between patients with and without co-morbidity [22,26,28,40]. Some studies have shown that performance status and co-morbidity are both independent prognostic factors [2,3], which therefore may both need to be included in future prognostic studies, supplemented by the psychological or mental condition of the patient and the patient’s and/or family decision or even doctor’s decision. 4.5. Conclusions The prevalence of co-morbidity among older cancer patients is high and older patients (with co-morbidity) are often treated less aggressively, which often seems to have a negative influence on survival. However, the question remains whether outcomes would really improve if more patients are treated, according to the guidelines that were developed on the basis of results in groups of younger patients without co-morbidity. It also remains relevant to study the quality of life after treatment, besides toxicity from treatment and prognosis. Reviewers Prof. Vesa V. Kataja, University Hospital of Kuopio, Department of Oncology, P.O. Box 1777, FIN-70211 Kuopio, Finland. Lenny Verkooijen, M.D., Ph.D., Institut de M´edecine Sociale et Pr´eventive, Registre Genevois des Tumeurs, Universit´e de Gen´eve, Boulevard de la Cluse 55, CH-1205 Geneva, Switzerland.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
References [1] Yancik R, Ganz PA, Varricchio CG, Conley B. Perspectives on comorbidity and cancer in older patients: approaches to expand the knowledge base. J Clin Oncol 2001;19:1147–51. [2] Extermann M, Overcash J, Lyman GH, Parr J, Balducci L. Comorbidity and functional status are independent in older cancer patients. J Clin Oncol 1998;16:1582–7. [3] Repetto L, Fratino L, Audisio RA, et al. Comprehensive geriatric assessment adds information to Eastern Cooperative Oncology Group performance status in elderly cancer patients: an Italian Group for Geriatric Oncology Study. J Clin Oncol 2002;20:494– 502. [4] Coebergh JWW, Janssen-Heijnen MLG, Louwman WJ, Voogd AC, editors. Cancer incidence and survival in the south of the Netherlands, 1955–1999 & incidence in the north of Belgium, 1996–1998 [CD-ROM]. Eindhoven: Comprehensive Cancer Centre South (IKZ); 2001. [5] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83. [6] Kieszak SM, Flanders WD, Kosinski AS, Shipp CC, Karp H. A comparison of the Charlson comorbidity index derived from medical record data and administrative billing data. J Clin Epidemiol 1999;52:137–42. [7] Hakulinen T, Abeywickrama KH. A computer program package for relative survival analysis. Comput Programs Biomed 1985;19:197–207. [8] Janssen-Heijnen ML, Schipper RM, Razenberg PP, Crommelin MA, Coebergh JW. Prevalence of co-morbidity in lung cancer patients and its relationship with treatment: a population-based study. Lung Cancer 1998;21:105–13. [9] Koppert LB, Janssen-Heijnen ML, Louwman MW, et al. Comparison of comorbidity prevalence in oesophageal and gastric carcinoma patients: a population-based study. Eur J Gastroenterol Hepatol 2004;16:681–8. [10] Jain M, Howe GR, St Louis P, Miller AB. Coffee and alcohol as determinants of risk of pancreas cancer: a case-control study from Toronto. Int J Cancer 1991;47:384–9. [11] Kalapothaki V, Tzonou A, Hsieh CC, et al. Ethanol, coffee, pancreatitis, diabetes mellitus, and cholelithiasis as risk factors for pancreatic carcinoma. Cancer Causes Control 1993;4:375–82. [12] Parazzini F, La Vecchia C, Bocciolone L, Franceschi S. The epidemiology of endometrial cancer. Gynecol Oncol 1991;41:1–16. [13] Lindblad P, Chow WH, Chan J, et al. The role of diabetes mellitus in the aetiology of renal cell cancer. Diabetologia 1999;42: 107–12. [14] McLaughlin JK, Chow WH, Mandel JS, et al. International renalcell cancer study. VIII. Role of diuretics, other anti-hypertensive medications and hypertension. Int J Cancer 1995;63:216–21. [15] Janssen-Heijnen ML, Smulders S, Lemmens VE, Smeenk FW, van Geffen HJ, Coebergh JW. Effect of comorbidity on the treatment and prognosis of elderly patients with non-small cell lung cancer. Thorax 2004;59:602–7. [16] Finlayson EV, Birkmeyer JD. Operative mortality with elective surgery in older adults. Eff Clin Pract 2001;4:172–7. [17] Damhuis RA, Schutte PR. Resection rates and postoperative mortality in 7,899 patients with lung cancer. Eur Respir J 1996;9:7–10. [18] Bolliger CT, Jordan P, Soler M, et al. Exercise capacity as a predictor of postoperative complications in lung resection candidates. Am J Respir Crit Care Med 1995;151:1472–80. [19] Post PN, Kil PJ, Hendrikx AJ, Janssen-Heijnen ML, Crommelin MA, Coebergh JW. Comorbidity in patients with prostate cancer and its relevance to treatment choice. BJU Int 1999;84:652–6. [20] Houterman S, Janssen-Heijnen ML, Verheij CD, et al. Comorbidity has negligible impact on treatment and complications but influences survival in breast cancer patients. Br J Cancer 2004;90:2332–7.
239
[21] Aus G, Abbou CC, Pacik D, et al. EAU guidelines on prostate cancer. Eur Urol 2001;40:97–101. [22] Lemmens VE, Janssen-Heijnen ML, Verheij CD, Houterman S, Repelaer van Driel OJ, Coebergh JW. Comorbidity leads to altered treatment and worse survival of elderly patients with colorectal cancer. Br J Surg 2005;92:615–23. [23] Forman WB. The role of chemotherapy and adjuvant therapy in the management of colorectal cancer. Cancer 1994;74:2151–3. [24] Stein BN, Petrelli NJ, Douglass HO, Driscoll DL, Arcangeli G, Meropol NJ. Age and sex are independent predictors of 5fluorouracil toxicity, analysis of a large scale phase III trial. Cancer 1995;75:11–7. [25] Zalcberg J, Kerr D, Seymour L, Palmer M, Tomudex International Study Group. Haematological and non-haematological toxicity after 5-fluorouracil and leucovorin in patients with advanced colorectal cancer is significantly associated with gender, increasing age and cycle number. Eur J Cancer 1998;34:1871–5. [26] van Spronsen DJ, Janssen-Heijnen ML, Lemmens VE, Peters WG, Coebergh JW. Independent prognostic effect of co-morbidity in lymphoma patients: results of the population-based Eindhoven Cancer Registry. Eur J Cancer 2005;41:1051–7. [27] de Rijke JM, Schouten LJ, Schouten HC, Jager JJ, Koppejan AG, van den Brandt PA. Age-specific differences in the diagnostics and treatment of cancer patients aged 50 years and older in the province of Limburg, The Netherlands. Ann Oncol 1996;7:677– 85. [28] Louwman WJ, Janssen-Heijnen ML, Houterman S, et al. Less extensive treatment and inferior prognosis for breast cancer patient with comorbidity: a population-based study. Eur J Cancer 2005;41:779–85. [29] (NABON) NBON, editor. Richtlijn Behandeling van het mammacarcinoom. Alphen aan den Rijn: Van Zuiden Communications B.V.; 2002. [30] Ayanian JZ, Zaslavsky AM, Fuchs CS, et al. Use of adjuvant chemotherapy and radiation therapy for colorectal cancer in a population-based cohort. J Clin Oncol 2003;21:1293–300. [31] Firat S, Bousamra M, Gore E, Byhardt RW. Comorbidity and KPS are independent prognostic factors in stage I non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2002;52:1047–57. [32] Firat S, Byhardt RW, Gore E, Radiation Therapy Oncology Group. Comorbidity and karnofksy performance score are independent prognostic factors in stage III non-small-cell lung cancer: an institutional analysis of patients treated on four RTOG studies. Int J Radiat Oncol Biol Phys 2002;54:357–64. [33] Battafarano RJ, Piccirillo JF, Meyers BF, et al. Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 2002;123:280–7. [34] Piccirillo JF. Importance of comorbidity in head and neck cancer. Laryngoscope 2000;110:593–602. [35] Kaplan MH, Feinstein AR. The importance of classifying initial comorbidity in evaluatin the outcome of diabetes mellitus. J Chronic Dis 1974;27:387–404. [36] Miller MD, Paradis CF, Houck PR, et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the cumulative illness rating scale. Psychiatry Res 1992;41:237– 48. [37] Tammemagi CM, Neslund-Dudas C, Simoff M, Kvale P. Impact of comorbidity on lung cancer survival. Int J Cancer 2003;103:792–802. [38] Piccirillo JF, Tierney RM, Costas I, Grove L, Spitznagel Jr EL. Prognostic importance of comorbidity in a hospital-based cancer registry. JAMA 2004;291:2441–7. [39] Read WL, Tierney RM, Page NC, et al. Differential prognostic impact of comorbidity. J Clin Oncol 2004;22:3099–103. [40] Post PN, Hansen BE, Kil PJ, Janssen-Heijnen ML, Coebergh JW. The independent prognostic value of comorbidity among men aged <75 years with localized prostate cancer: a population-based study. BJU Int 2001;87:821–6.
240
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
Biography Dr. Maryska L.G. Janssen-Heijnen is working as a senior epidemiologist at the Eindhoven Cancer Registry, The Netherlands. She focuses at prognostic factors for cancer patients, with a special interest in elderly patients
with co-morbidity. Since 2000, she is co-ordinating a large population-based project in which the prevalence of co-morbidity in cancer patients is studied, as well as the influence of increasing age and co-morbidity on choice of treatment, complications of treatment and prognosis.
Prognostic impact of increasing age and co-morbidity in cancer patients: A population-based approach Maryska L.G. Janssen-Heijnen a,∗ , Saskia Houterman a , Valery E.P.P. Lemmens a , Marieke W.J. Louwman a , Huub A.A.M. Maas b , Jan Willem W. Coebergh a,c a
Eindhoven Cancer Registry, Comprehensive Cancer Centre South, P.O. Box 231, 5600 AE Eindhoven, The Netherlands b Department of Clinical Geriatrics, Tweesteden Hospital, Tilburg, The Netherlands c Department of Public Health, University Medical Centre Rotterdam, Rotterdam, The Netherlands Accepted 22 April 2005
Contents 1. 2. 3.
4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Prevalence of co-morbidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Treatment (see overview in Table 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Complications of treatment (Fig. 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Survival (Table 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Validity of data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Prevalence of co-morbidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Treatment and complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reviewers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
232 232 233 233 233 233 237 237 237 237 237 238 238 238 239 240
Abstract This large population-based study focuses on the prognostic role of increasing age and co-morbidity in cancer patients diagnosed in the southern Netherlands. Data of patients diagnosed between 1995 and 2002 and recorded in the population-based Eindhoven Cancer Registry were used. Older patients (with serious co-morbidity) with non-small cell lung cancer or prostate cancer underwent surgery less often than younger patients. Elderly with stage III colon cancer, small cell lung cancer, FIGO II or III ovarian cancer or non-Hodgkin’s lymphoma (NHL) received (adjuvant) chemotherapy less often, probably because of the higher rate of haematological complications. Administration of adjuvant radiotherapy decreased with age and co-morbidity in patients with rectal cancer, limited small cell lung cancer or breast cancer. In general, elderly did not suffer from more complications than younger patients, except for cardiac complications (colorectal cancer and NHL) and postoperative death (non-small cell lung cancer). For most tumours relative survival was lower for the elderly, except for patients with colon cancer, prostate cancer or indolent NHL. Co-morbidity had an independent prognostic effect, except for tumours with a very poor prognosis. Future prospective studies should investigate whether the guidelines for cancer treatment should be adjusted for elderly with serious co-morbidity. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Elderly; Co-morbidity; Cancer; Treatment; Complications of treatment; Survival
∗
Corresponding author. Tel.: +31 40 2971616; fax: +31 40 2971610. E-mail address: [email protected] (M.L.G. Janssen-Heijnen).
1040-8428/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.critrevonc.2005.04.008
232
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
1. Introduction Due to ageing of the population and the rising incidence rates of most cancers with age, the mean age of patients diagnosed with cancer is increasing in western countries. This implies that they increasingly suffer from one or more other serious (chronic) diseases. Besides affecting the life expectancy, co-morbid conditions may complicate the treatment of cancer patients, especially when they are frail [1–3]. Since elderly patients are often excluded from clinical trials, little is known about treatment outcome, such as complications, quality of life and survival. Although the number of trials that include elderly is increasing, generally only relatively healthy elderly are included. This means that trial results may not be valid for many older cancer patients. This study focuses on the prognostic role of increasing age and co-morbidity in cancer patients diagnosed in the registration area of the population-based Eindhoven Cancer Registry.
2. Methods The Eindhoven Cancer Registry records data on all patients newly diagnosed with cancer in the southern part of the Netherlands, an area with 2.3 million inhabitants and only community hospitals [4]. Since 1993, serious co-morbidity with prognostic impact has been recorded for all patients. The Charlson co-morbidity index is most widely used for recording co-morbidity and was validated in various studies [5]. We used a slightly modified version of this index for recording co-morbidity (Table 1). Co-morbidity was defined as life-shortening diseases that were present at the time of cancer diagnosis. Table 1 Classification of co-morbidity, according to a modified version of the list of Charlson et al. [5] Chronic obstructive pulmonary disease (COPD) Cardiovascular disease: myocardial infarction, cardiac insufficiency, angina pectoris, coronary artery bypass graft (CABG) Peripheral arterial disease: intermittent claudication, abdominal aneurysm, surgical intervention Cerebrovascular diseases (cerebrovascular accident, hemiplegia) Other malignancies (except basal cell skin carcinoma) Hypertension Diabetes mellitus Other Autoimmune diseases: sarcoidosis, Wegener’s disease, systemic lupus erythematosis (SLE) Rheumatoid arthritis (only severe) Kidney diseases: glomerulonephritis, pyelonephritis Gastrointestinal: stomach ulcer and resection, colitis Liver diseases: cirrhosis, hepatitis Dementia Chronic infections
Trained registry personnel extracted all data from the medical records between 6 and 18 months after diagnosis. The medical record is generally regarded as the most complete source of information on the patient’s past and current health status [6]. Patients with cancer of the esophagus, stomach, colon or rectum, pancreas, lung, breast, cervix uteri, corpus uteri, ovary, prostate, bladder, kidney and non-Hodgkin’s lymphoma, newly diagnosed between 1995 and 2002 (N = 43,111), were included for this overview. Patients with cancer diagnosed at autopsy (N = 661 or 1.5%) were excluded. Treatment was classified as surgery (resection), radiotherapy, chemotherapy, endocrine therapy (or combinations of the aforementioned treatment options) and ‘other or none’. Surgery did not comprise diagnostic operations. For a random sample of 1686 patients with cancer of the colon/rectum, lung, breast, prostate and non-Hodgkin’s lymphoma additional information about grade 3 or 4 complications after treatment (according to the NCI CTC-toxicity criteria) within 1 year after diagnosis were gathered from the medical records. Survival analyses were restricted to patients with cancer of the colon or rectum, lung, breast, ovary, prostate or nonHodgkin’s lymphoma newly diagnosed between 1995 and 2001 (N = 29,249). Follow-up was completed until 1 January 2004. In addition to passive follow-up via the hospitals, this information was also obtained from the municipal registries in the area of the Eindhoven Cancer Registry and the Central Bureau for Genealogy. The latter is an institution that collects data on all deceased Dutch citizens via the civil municipal registries. In this way, information on patients who had moved outside the registry area was also obtained. Patients who died outside the Netherlands were wrongly considered as ‘being alive’. However, the estimated proportion of these patients was less than 0.3%. Survival time was defined as the time from diagnosis to death or the end of the study. Survival generally decreases with age, because other causes of death also take their share. Therefore, for describing the prognostic effect of age, we calculated relative survival rates for each age group, which are an estimation of disease-specific survival. In this way, survival of cancer patients is adjusted for mortality from all causes of death in the background population with the same age structure [7]. Mortality in the background population was estimated from life tables for regional male and female populations. The independent prognostic effect of the number of co-morbid conditions was estimated with a multivariable Cox regression model. The hazard rates for death were adjusted for age, gender, subsite (only for colorectal cancer), histology (only for NSCLC), stage and treatment. The SAS computer package (Version 8.2) was used for all statistical analyses (SAS Institute Inc., Cary, North Carolina, USA, 1999).
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
3. Results 3.1. Prevalence of co-morbidity The prevalence of co-morbidity usually increased with age (Table 2), but remained stable or decreased above age 80 for registered patients with pancreatic or lung cancer. About 60% of all new cancer patients older than 65 suffered from at least one other serious disease. The most frequent concomitant diseases were previous cancers, heart disease, hypertension, COPD and diabetes mellitus, with prevalence rates up to 20, 23, 26, 17 and 16%, respectively. The prevalence of cardiovascular diseases was higher among men compared to women, and was highest for patients with cancer of the digestive tract, lung and kidney and non-Hodgkin’s lymphoma. The prevalence of COPD was relatively high among patients with lung cancer, and also among men with esophageal or bladder cancer. The prevalence of hypertension was highest among women with cancer of the corpus uteri or kidney. High prevalence rates of diabetes in older patients were observed for cancer of the pancreas, cervix uteri, corpus uteri and kidney. 3.2. Treatment (see overview in Table 3) Patients with colon cancer underwent surgery regardless of age or the number of co-morbid conditions: more than 95% with stages I–III disease and about 75% of patients with stage IV disease (Table 3). However, patients with stage III disease received adjuvant chemotherapy less frequently with the rise of age: from 82% of patients in age group 50–64 years to 42% of patients aged 65–79 years (P < 0.01, data 1997–2002). The proportion of patients with stage III disease aged 50–79 years receiving adjuvant chemotherapy also decreased from 69% of patients without co-morbidity to 53% of those with co-morbidity (P < 0.01, data 1997–2002). The proportion of rectal cancer patients stages I–III receiving adjuvant radiotherapy decreased from 57% of patients younger than 65 years to 36% of patients aged 65 years or older (P < 0.01), and also decreased with co-morbidity from 50% of patients without co-morbidity to 39% in case of co-morbidity (P < 0.01). The proportion of patients with localised non-small cell lung cancer who underwent surgery with or without radiotherapy was only 11% of those aged 80 years or older versus 91%, 79% and 59% of the age groups <60, 60–69 and 70–79 years, respectively (P < 0.01). Patients aged 60–69 and 70–79 years also underwent surgery less often when co-morbidity was present (P < 0.01 and P = 0.01, respectively). Most patients with non-localised non-small cell lung cancer received only radiotherapy (36%). The proportion receiving chemotherapy (with or without radiotherapy) was considerably higher among patients younger than 60 years (28%) than among those aged 80 years or older (2%) (P < 0.01). The number of co-morbid conditions had no substantial influence on treatment chosen for patients with
233
non-localised disease. Elderly patients with limited small cell lung cancer received combined chemoradiation less often and chemotherapy alone more often (P < 0.01). Among patients with limited small cell lung cancer aged 70–79 years the proportion receiving adjuvant radiotherapy also decreased in the presence of co-morbidity (P = 0.05). Among patients with breast cancer younger than 80 years over 95% underwent surgery, compared with only 76% of those aged 80 years or older (P < 0.01). Among those undergoing surgery, the proportion of breast conserving surgery decreased from 54% of those younger than 60 years to 29% of those aged 80 years or older (P < 0.01). Axilliary dissection for those undergoing breast conserving surgery decreased in the presence of co-morbidity (78% of patients with at least two co-morbid conditions compared to 97% of those without co-morbidity, P < 0.01), as did adjuvant radiotherapy (from 94% of patients without co-morbidity to 78% of those with at least two co-morbid conditions, P < 0.01). Eighty-three percent of patients younger than 70 years with FIGO II or III ovarian cancer received the recommended treatment (combination of surgery and chemotherapy): 87% of those without co-morbidity and 73% of those with at least one co-morbid condition (P < 0.01). In contrast, only 45% of the patients aged 70 years or older received the recommended treatment (54% of those without co-morbidity and only 38% of the patients with co-morbidity (P < 0.05). The proportion of prostate cancer patients undergoing prostatectomy decreased with increasing age, from 41% of patients younger than 60 years to 1% of patients aged 80 years or older (P < 0.01). The proportion of patients receiving curative radiotherapy also decreased from 30% of those younger than 60 years to 5% of those aged 80 years or older (P < 0.01). Endocrine treatment increased from 19% of patients younger than 60 years to 58% of patients aged 80 years or older. Among patients aged 60–79 years, the proportion who underwent prostatectomy also significantly decreased with co-morbidity, whereas the proportion receiving radiotherapy or endocrine treatment increased (P < 0.01). For patients with indolent non-Hodgkin’s lymphoma (NHL) the proportion receiving chemotherapy did not decrease with age. For patients with aggressive NHL, the proportion receiving chemotherapy decreased from 88% of those aged 60 years or younger to 68% of those older than 60 years (P < 0.01). Among patients with aggressive disease the proportion receiving chemotherapy also decreased with co-morbidity (P < 0.05). 3.3. Complications of treatment (Fig. 1) Thirty-seven percent of resected colorectal cancer patients between 40 and 69 years old suffered from any complication within 1 year of diagnosis compared to 44% of patients aged 70 years or older (P = 0.08) (Fig. 1a). The most frequent complications were major infections (11%), minor infections (11%), typical irradiation side
234
Table 2 Age-specific prevalence (%) of the most common serious concomitant diseasesa among newly diagnosed patients with 13 major cancers, 1995–2002 Sex
N
Any co-morbidity (%)
Previous cancers (%)
Heart and vascular disease (%)
COPD (%)
Hypertension (%)
Diabetes (%)
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
50–64 years
65–79 years
80+ years
Esophagus
M F
602 258
51 43
72 53
74 70
6 11
17 15
26 15
14 5
36 17
28 13
12 14
17 10
24 10
15 13
21 19
10 23
5 5
11 12
18 12
Stomach
M F
1310 721
50 41
67 58
72 68
6 11
12 14
20 16
19 8
33 16
25 25
8 3
13 9
16 7
15 13
15 26
18 31
6 6
10 13
12 20
Colon/rectum
M F
4502 3992
40 37
64 56
71 65
7 10
15 14
22 16
13 5
28 14
32 24
6 5
15 8
15 9
16 15
21 25
15 25
7 5
10 14
13 17
Pancreas
M F
551 518
49 42
67 63
67 66
7 8
12 9
15 8
21 8
28 14
33 24
5 3
17 8
15 6
13 15
18 26
21 29
13 10
21 26
23 25
Lung
M F
6834 1816
53 49
74 67
72 61
9 9
16 16
16 14
18 11
34 22
32 25
20 21
24 24
31 16
11 12
15 21
11 22
7 6
10 12
11 11
Breast
F
7978
28
52
67
6
10
15
4
12
22
4
6
8
13
29
27
5
13
16
Cervix uteri
F
289
33
51
62
6
8
12
5
14
21
4
6
2
13
21
26
8
14
31
Corpus uteri
F
1308
37
64
77
8
10
20
3
13
24
3
6
12
22
38
33
8
22
25
Ovary
F
1011
34
58
65
12
15
15
4
14
20
5
7
6
12
25
27
4
12
11
Prostate
M
6272
36
56
59
6
9
14
12
24
27
5
12
15
12
17
12
4
8
9
Bladder
M F
1646 449
47 49
70 60
73 64
18 23
24 23
29 20
16 8
29 13
27 23
8 8
15 8
20 8
9 10
16 29
13 20
5 6
11 16
11 16
Kidney
M F
725 534
46 46
69 69
82 76
11 7
19 18
26 18
14 6
31 18
39 27
5 8
15 7
16 5
16 25
24 32
21 48
8 8
10 19
21 21
NHL
M F
792 685
39 32
67 61
76 65
5 5
14 16
19 13
13 5
27 11
39 23
4 2
15 8
11 4
13 14
17 26
16 20
5 6
9 12
11 13
Source: Eindhoven Cancer Registry. a More conditions per patient possible.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
Tumour
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
effects, such as radiation enteritis (9% of patients receiving radiotherapy) and typical chemotherapy side effects, such as neutropenia, thrombocytopenia, diarrhea and soreness of the mouth (10% of patients aged 40–69 years receiving chemotherapy). Elderly patients suffered more from cardiac complications than patients aged 40–69 years (11% versus 5%, P < 0.01), especially those with cardiovascular co-morbidity. Forty-nine percent of patients with localised NSCLC who underwent resection suffered from one or more
235
complications within 1 year of diagnosis (no difference between both age groups, Fig. 1b). The most frequent postoperative complications were major infections (16%) and minor infections (11%). Postoperative death occurred in 5% of the patients. It occurred significantly more often in patients aged 70 years or older (11%) than in those younger than 70 years (2%) (P < 0.01). Seventy-seven percent of patients with SCLC receiving chemotherapy suffered from at least one complication, mainly haematological complications (57%) or other complications due to chemotherapy
Fig. 1. Age-specific prevalence of complications during the first year after diagnosis (1995–1999), (a) postoperative complications colorectal cancer, (b) postoperative complications localised NSCLC, (c) complications after chemotherapy (±RT) SCLC, (d) postoperative complications breast cancer, (e) complications after radiotherapy prostate cancer and (f) complications after chemotherapy aggressive NHL.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
236
Table 3 Influence of age and co-morbidity on primary treatment, according to tumour type and/or stage Tumour
Stage
Influence of rising age
Influence of co-morbidity
Colon
Stage I/II Stage III
None Less adjuvant CT
None Age 65–79 years: less adjuvant CT
Rectum
Stage II/III
Less adjuvant RT
Less adjuvant RT
NSCLCa
Stage I/II Stage III/IV
Less surgery, more RT alone Less CT
Age 60–79 years: less surgery None
SCLCb
Limited Extensive
Less CT + RT, more CT alone Less CT, more abstinence
Age 70–79 years: less CT + RT None
Less surgery, more endocrine
Less adjuvant RT, more endocrine
Less surgery + CT
Less surgery + CT
Less prostatectomy, more endocrine
Age 60–79 years: less prostatectomy, more endocrine
Less CT, more RT and wait and see Less CT, more abstinence
None Age 70+ years: less CT
Breast Ovary
FIGO II/III
Prostate NHLc
Indolent Aggressive
Source: Eindhoven Cancer Registry; CT = chemotherapy; RT = radiotherapy. a Non-small cell lung cancer. b Small cell lung cancer. c Non-Hodgkin’s lymphoma.
(42%) (Fig. 1c). The proportion of complications did not differ between the two age groups. The number of breast cancer patients with complications after surgery was similar for both age groups (27%) (Fig. 1d). Minor infection (11%) was the most common complication, followed by lymphoedema (7% of those who underwent axillary dissection). In prostate cancer patients undergoing radical prostatectomy, 38% suffered a complication during the first year after diagnosis. The most common complication was infection (14% in those younger than 70 years and 25% in those aged 70 years or older). In patients younger than 70 years,
26% had other complications like strictures. Among those receiving radiotherapy the proportion of patients with a complication was about 20% (Fig. 1e). Most of these complications were radiotherapy-related complications, such as proctitis. The proportion of patients with aggressive NHL who developed any complication after chemotherapy was 63% for patients aged 40–60 years and 69% for those aged 61 years or older (P = 0.4, Fig. 1f). The most common complications were haematological (33%). Cardiovascular complications were more common among the elderly (21% versus 11%, P = 0.06). Three percent of patients died due to treatment
Table 4 Relative 2- and 5-year survival rates (% and 95% CI), according to age and tumour site Tumour type
Relative survival (% (95% CI)) 2-year <70 years
5-year <70 years
70+ years
Colon
70 (68–72)
70+ years 66 (64–68)
57 (55–60)
59 (56–62)
Rectum
76 (73–78)
63 (60–67)
62 (59–65)
51 (47–55) 29 (24–34)
NSCLCa
localised
64 (61–68)
41 (37–46)
47 (43–51)
NSCLCa
non-localised
13 (11–14)
8 (7–10)
4 (3–6)
–
SCLCb
12 (10–14)
6 (4–8)
5 (4–7)
–
Breast
95 (94–96)
91 (90–93)
86 (85–87)
81 (78–84)
Ovary
67 (62–71)
39 (34–45)
45 (40–50)
25 (20–31)
Prostate
100 (98–100)
99 (97–101)
95 (93–97)
95 (91–98)
NHLc indolent
89 (84–93)
71 (61–80)
73 (66–80)
65 (51–78)
NHLc aggressive
63 (58–68)
43 (37–50)
51 (45–57)
36 (28–45)
Source: Eindhoven Cancer Registry. a Non-small cell lung cancer. b Small cell lung cancer. c Non-Hodgkin’s lymphoma.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
(1% of those younger than 61 years and 4% of the elderly, P = 0.4). 3.4. Survival (Table 4) In general, the relative 5-year survival rate was lower for the elderly, except for patients with colon cancer, prostate cancer or indolent NHL (Table 4). For elderly patients with non-localised NSCLC or SCLC 5-year survival could not be estimated, because of the poor prognosis. For these patients relative 2-year survival was also lower for the elderly. In multivariable survival analyses, the independent prognostic effect of the number of co-morbid conditions was evaluated (no co-morbidity as reference), after adjustment for age, gender, subsite (only for colorectal cancer), histology (only for NSCLC), stage and treatment. The hazard ratio for death for one and two or more co-morbid conditions for patients with colon cancer was 1.2 (95% CI 1.1–1.3) and 1.4 (95% CI 1.2–1.5), respectively. For rectal cancer, these hazard ratios were 1.3 (95% CI 1.1–1.5) and 1.6 (95% CI 1.4–1.9), for breast cancer patients 1.3 (95% 1.2–1.5) and 1.4 (95% CI 1.3–1.5) and for localised prostate cancer 1.2 (95% CI 1.03–1.4) and 1.9 (95% CI 1.6–2.2). The hazard ratio for death for at least one co-morbid condition was 1.7 (95% CI 1.2–2.5) for patients with indolent NHL and 1.2 (95% CI 0.99–1.5) for patients with aggressive NHL. For patients with localised NSCLC, non-localised NSCLC, SCLC or ovarian cancer co-morbidity had no independent prognostic effect. 4. Discussion 4.1. Validity of data Between 2001 and 2003 the completeness and accuracy of our data on co-morbidity in the Eindhoven Cancer registry were validated in a random sample of 2607 patients with colorectal, lung, breast and prostate cancer and non-Hodgkin’s lymphoma aged 40 years and older and diagnosed between 1995 and 1999. Co-morbidity scored by the registry team was compared with that scored by a team of a retired surgeon and an epidemiologist. Some underregistration occurred, especially of cardiovascular conditions (internal reports, 2002). This appeared to be mainly due to the use of unknown terminology, unknown abbreviations or illegible handwriting of the specialist. This implies that the real effects of co-morbidity on treatment and survival might be stronger than those presented in this article. 4.2. Prevalence of co-morbidity The higher prevalence of co-morbidity among older patients was expected, because the prevalence of chronic diseases generally increases with age. The prevalence of comorbidity among older patients may even be underestimated due to ascertainment bias. Younger patients underwent
237
surgery more often and received chemotherapy more often. The prevalence of co-morbidity reported by the treating physician might then be more elevated among younger patients, due to the required screening examinations before treatment. The high risk of cardiovascular diseases and chronic obstructive pulmonary diseases for patients with cancer of the esophagus, stomach, lung, bladder and kidney can be explained by the high proportion of smokers among these patients, especially men [8,9]. It is not surprising that diabetes mellitus occurred in a high proportion of patients with cancer of the pancreas [10,11]. A history of diabetes has been consistently associated with a two-fold increased risk for endometrial cancer [12], probably because both are related with obesity. The risk of renal cancer is also elevated among patients with diabetes [13]. Renal cancer has also been associated with hypertension, although it is unknown whether this results from the hypertension itself or from the antihypertensives [14]. The prevalence of cardiovascular diseases and pulmonary diseases was higher among men compared to women, which can be largely explained by a higher prevalence of smoking among men in the past. By contrast, the prevalence of hypertension (a less serious condition) and diabetes was higher among women. 4.3. Treatment and complications If alternative treatment strategies were available, older patients (with co-morbidity) were often treated less aggressively than younger patients. When surgery is inevitable like in patients with colorectal cancer, higher age or the prevalence of co-morbidity did not have any influence on the resection rate. On the other hand, older patients with localised non-small cell lung cancer (with serious co-morbidity) more often received radiotherapy instead of surgery [15]. Surgical mortality increases markedly with age and is especially high for pneumonectomy [16,17]. The resection rate also declined with co-morbidity, probably because of the expected higher incidence of postoperative complications and mortality [18]. In accordance with recommendations in treatment guidelines for elderly patients with prostate cancer, elderly received more hormonal treatment and less prostatectomy or radiotherapy [19–21]. The frequency of administering adjuvant chemotherapy for patients with stage III colon cancer markedly decreased with rising age and co-morbidity [22], probably because of the higher rate of haematological complications [23–25]. For the same reason, administration of primary chemotherapy also decreased with age and co-morbidity for patients with non-Hodgkin’s lymphoma [26]. The lower proportion of elderly patients undergoing surgery or receiving chemotherapy also appeared in another area of the Netherlands [27]. In our study, administration of adjuvant radiotherapy decreased with age and co-morbidity in patients with
238
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
rectal cancer, limited small cell lung cancer or breast cancer [22,28]. However, we did not find a higher rate of complications due to radiotherapy for older patients with co-morbidity [20]. Therefore, the reluctance of accepting adjuvant radiotherapy might be related to practical reasons like the distance to a radiotherapy institute or the burden of the 20–30 visits to the radiotherapy institute or the availability of alternative treatment (such as breast amputation instead of breast conserving surgery with radiotherapy). Age seemed to have more influence on treatment chosen than co-morbidity. Apparently, co-morbidity alone does not entirely explain why elderly non-small cell lung cancer patients and prostate cancer patients underwent surgery less often and why those with colon cancer, rectal cancer, small cell lung cancer and breast cancer received less (adjuvant) chemotherapy or adjuvant radiotherapy. For patients with breast cancer, current guidelines recommend less chemotherapy for elderly breast cancer patients, because postmenopausal women with receptor positive breast cancer will get the highest benefit from hormonal adjuvant therapy and current guidelines for prostate cancer recommend radiotherapy or hormonal treatment for elderly patients [21,29]. However, performance status, the psychological condition of the patient, social factors and patient’s decision may also play a role, especially for the other tumours [3,30]. The fact that elderly often did not have more complications than younger patients might be explained by (appropriate) selection for treatment. This was not a clinical trial, but a population-based study in which physicians selected patients for treatment. Among patients with colorectal cancer or NHL, cardiac complications occurred more among the elderly, especially those with cardiovascular co-morbidity. This might also partly be explained by the fact that the occurrence of cardiac events in the general population increases with age. 4.4. Survival For most tumour types relative survival decreased with age, except for patients with colon cancer, prostate cancer or indolent NHL. Relative survival reflects disease-specific survival. Therefore, the decrease in relative survival with age for the other tumour types cannot be explained by a higher rate of death in the elderly due to other causes. Less intense treatment of either the tumour or the co-morbid condition might play a role. For patients with lung cancer co-morbidity had no independent prognostic effect [15]. This contradicts some other studies [31–34], which however were not populationbased. In the other studies also other scales for measuring co-morbidity were used: the Kaplan–Feinstein Index [35] and the Cumulative Illness Rating Scale-Geriatric (CIRS-G) [36]. In one of the studies, co-morbidity affected overall survival in surgically resected stage I NSCLC patients, when co-morbidity was rated according to CIRS-G, but
not according to the Charlson scale [31]. In another American study, co-morbidity count and the Charlson index were significant predictors for lung cancer survival, but only explained 2.5 and 2.0% of the survival variation, respectively [37]. We also found no independent prognostic effect of co-morbidity for patients with ovarian cancer. Probably, the influence of co-morbidity on survival is of less importance in the case of a lethal disease, such as lung cancer or ovarian cancer. Most of these patients die of the cancer, before they become at risk of dying from the co-morbid condition. This was confirmed in two recent American studies, in which the prognostic effect of co-morbidity was found to be smallest for tumours with a poor prognosis [38,39]. For the other tumours, co-morbidity had an independent prognostic effect. This negative influence of co-morbidity on survival of cancer might be due to several mechanisms: the increased risk of death due to the co-morbid condition itself, more contra-indications for anti-cancer treatment, more indications for dose reduction and a higher rate of treatment-related complications, such as infections and cardiovascular events. In several of our recent studies, the adverse effects of co-morbidity on survival appeared to be independent of treatment, so less aggressive treatment could not (fully) account for the observed differences in survival between patients with and without co-morbidity [22,26,28,40]. Some studies have shown that performance status and co-morbidity are both independent prognostic factors [2,3], which therefore may both need to be included in future prognostic studies, supplemented by the psychological or mental condition of the patient and the patient’s and/or family decision or even doctor’s decision. 4.5. Conclusions The prevalence of co-morbidity among older cancer patients is high and older patients (with co-morbidity) are often treated less aggressively, which often seems to have a negative influence on survival. However, the question remains whether outcomes would really improve if more patients are treated, according to the guidelines that were developed on the basis of results in groups of younger patients without co-morbidity. It also remains relevant to study the quality of life after treatment, besides toxicity from treatment and prognosis. Reviewers Prof. Vesa V. Kataja, University Hospital of Kuopio, Department of Oncology, P.O. Box 1777, FIN-70211 Kuopio, Finland. Lenny Verkooijen, M.D., Ph.D., Institut de M´edecine Sociale et Pr´eventive, Registre Genevois des Tumeurs, Universit´e de Gen´eve, Boulevard de la Cluse 55, CH-1205 Geneva, Switzerland.
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
References [1] Yancik R, Ganz PA, Varricchio CG, Conley B. Perspectives on comorbidity and cancer in older patients: approaches to expand the knowledge base. J Clin Oncol 2001;19:1147–51. [2] Extermann M, Overcash J, Lyman GH, Parr J, Balducci L. Comorbidity and functional status are independent in older cancer patients. J Clin Oncol 1998;16:1582–7. [3] Repetto L, Fratino L, Audisio RA, et al. Comprehensive geriatric assessment adds information to Eastern Cooperative Oncology Group performance status in elderly cancer patients: an Italian Group for Geriatric Oncology Study. J Clin Oncol 2002;20:494– 502. [4] Coebergh JWW, Janssen-Heijnen MLG, Louwman WJ, Voogd AC, editors. Cancer incidence and survival in the south of the Netherlands, 1955–1999 & incidence in the north of Belgium, 1996–1998 [CD-ROM]. Eindhoven: Comprehensive Cancer Centre South (IKZ); 2001. [5] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83. [6] Kieszak SM, Flanders WD, Kosinski AS, Shipp CC, Karp H. A comparison of the Charlson comorbidity index derived from medical record data and administrative billing data. J Clin Epidemiol 1999;52:137–42. [7] Hakulinen T, Abeywickrama KH. A computer program package for relative survival analysis. Comput Programs Biomed 1985;19:197–207. [8] Janssen-Heijnen ML, Schipper RM, Razenberg PP, Crommelin MA, Coebergh JW. Prevalence of co-morbidity in lung cancer patients and its relationship with treatment: a population-based study. Lung Cancer 1998;21:105–13. [9] Koppert LB, Janssen-Heijnen ML, Louwman MW, et al. Comparison of comorbidity prevalence in oesophageal and gastric carcinoma patients: a population-based study. Eur J Gastroenterol Hepatol 2004;16:681–8. [10] Jain M, Howe GR, St Louis P, Miller AB. Coffee and alcohol as determinants of risk of pancreas cancer: a case-control study from Toronto. Int J Cancer 1991;47:384–9. [11] Kalapothaki V, Tzonou A, Hsieh CC, et al. Ethanol, coffee, pancreatitis, diabetes mellitus, and cholelithiasis as risk factors for pancreatic carcinoma. Cancer Causes Control 1993;4:375–82. [12] Parazzini F, La Vecchia C, Bocciolone L, Franceschi S. The epidemiology of endometrial cancer. Gynecol Oncol 1991;41:1–16. [13] Lindblad P, Chow WH, Chan J, et al. The role of diabetes mellitus in the aetiology of renal cell cancer. Diabetologia 1999;42: 107–12. [14] McLaughlin JK, Chow WH, Mandel JS, et al. International renalcell cancer study. VIII. Role of diuretics, other anti-hypertensive medications and hypertension. Int J Cancer 1995;63:216–21. [15] Janssen-Heijnen ML, Smulders S, Lemmens VE, Smeenk FW, van Geffen HJ, Coebergh JW. Effect of comorbidity on the treatment and prognosis of elderly patients with non-small cell lung cancer. Thorax 2004;59:602–7. [16] Finlayson EV, Birkmeyer JD. Operative mortality with elective surgery in older adults. Eff Clin Pract 2001;4:172–7. [17] Damhuis RA, Schutte PR. Resection rates and postoperative mortality in 7,899 patients with lung cancer. Eur Respir J 1996;9:7–10. [18] Bolliger CT, Jordan P, Soler M, et al. Exercise capacity as a predictor of postoperative complications in lung resection candidates. Am J Respir Crit Care Med 1995;151:1472–80. [19] Post PN, Kil PJ, Hendrikx AJ, Janssen-Heijnen ML, Crommelin MA, Coebergh JW. Comorbidity in patients with prostate cancer and its relevance to treatment choice. BJU Int 1999;84:652–6. [20] Houterman S, Janssen-Heijnen ML, Verheij CD, et al. Comorbidity has negligible impact on treatment and complications but influences survival in breast cancer patients. Br J Cancer 2004;90:2332–7.
239
[21] Aus G, Abbou CC, Pacik D, et al. EAU guidelines on prostate cancer. Eur Urol 2001;40:97–101. [22] Lemmens VE, Janssen-Heijnen ML, Verheij CD, Houterman S, Repelaer van Driel OJ, Coebergh JW. Comorbidity leads to altered treatment and worse survival of elderly patients with colorectal cancer. Br J Surg 2005;92:615–23. [23] Forman WB. The role of chemotherapy and adjuvant therapy in the management of colorectal cancer. Cancer 1994;74:2151–3. [24] Stein BN, Petrelli NJ, Douglass HO, Driscoll DL, Arcangeli G, Meropol NJ. Age and sex are independent predictors of 5fluorouracil toxicity, analysis of a large scale phase III trial. Cancer 1995;75:11–7. [25] Zalcberg J, Kerr D, Seymour L, Palmer M, Tomudex International Study Group. Haematological and non-haematological toxicity after 5-fluorouracil and leucovorin in patients with advanced colorectal cancer is significantly associated with gender, increasing age and cycle number. Eur J Cancer 1998;34:1871–5. [26] van Spronsen DJ, Janssen-Heijnen ML, Lemmens VE, Peters WG, Coebergh JW. Independent prognostic effect of co-morbidity in lymphoma patients: results of the population-based Eindhoven Cancer Registry. Eur J Cancer 2005;41:1051–7. [27] de Rijke JM, Schouten LJ, Schouten HC, Jager JJ, Koppejan AG, van den Brandt PA. Age-specific differences in the diagnostics and treatment of cancer patients aged 50 years and older in the province of Limburg, The Netherlands. Ann Oncol 1996;7:677– 85. [28] Louwman WJ, Janssen-Heijnen ML, Houterman S, et al. Less extensive treatment and inferior prognosis for breast cancer patient with comorbidity: a population-based study. Eur J Cancer 2005;41:779–85. [29] (NABON) NBON, editor. Richtlijn Behandeling van het mammacarcinoom. Alphen aan den Rijn: Van Zuiden Communications B.V.; 2002. [30] Ayanian JZ, Zaslavsky AM, Fuchs CS, et al. Use of adjuvant chemotherapy and radiation therapy for colorectal cancer in a population-based cohort. J Clin Oncol 2003;21:1293–300. [31] Firat S, Bousamra M, Gore E, Byhardt RW. Comorbidity and KPS are independent prognostic factors in stage I non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2002;52:1047–57. [32] Firat S, Byhardt RW, Gore E, Radiation Therapy Oncology Group. Comorbidity and karnofksy performance score are independent prognostic factors in stage III non-small-cell lung cancer: an institutional analysis of patients treated on four RTOG studies. Int J Radiat Oncol Biol Phys 2002;54:357–64. [33] Battafarano RJ, Piccirillo JF, Meyers BF, et al. Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 2002;123:280–7. [34] Piccirillo JF. Importance of comorbidity in head and neck cancer. Laryngoscope 2000;110:593–602. [35] Kaplan MH, Feinstein AR. The importance of classifying initial comorbidity in evaluatin the outcome of diabetes mellitus. J Chronic Dis 1974;27:387–404. [36] Miller MD, Paradis CF, Houck PR, et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the cumulative illness rating scale. Psychiatry Res 1992;41:237– 48. [37] Tammemagi CM, Neslund-Dudas C, Simoff M, Kvale P. Impact of comorbidity on lung cancer survival. Int J Cancer 2003;103:792–802. [38] Piccirillo JF, Tierney RM, Costas I, Grove L, Spitznagel Jr EL. Prognostic importance of comorbidity in a hospital-based cancer registry. JAMA 2004;291:2441–7. [39] Read WL, Tierney RM, Page NC, et al. Differential prognostic impact of comorbidity. J Clin Oncol 2004;22:3099–103. [40] Post PN, Hansen BE, Kil PJ, Janssen-Heijnen ML, Coebergh JW. The independent prognostic value of comorbidity among men aged <75 years with localized prostate cancer: a population-based study. BJU Int 2001;87:821–6.
240
M.L.G. Janssen-Heijnen et al. / Critical Reviews in Oncology/Hematology 55 (2005) 231–240
Biography Dr. Maryska L.G. Janssen-Heijnen is working as a senior epidemiologist at the Eindhoven Cancer Registry, The Netherlands. She focuses at prognostic factors for cancer patients, with a special interest in elderly patients
with co-morbidity. Since 2000, she is co-ordinating a large population-based project in which the prevalence of co-morbidity in cancer patients is studied, as well as the influence of increasing age and co-morbidity on choice of treatment, complications of treatment and prognosis.