- Email: [email protected]
Adjuvant Chemotherapy Uptake in Non-small Cell Lung Cancer
ORIGINAL ARTICLE
Adjuvant Chemotherapy Uptake in Non-small Cell Lung Cancer Tallal Younis, MBBCh, FRCP (UK),*† Turki Al-Fayea, MBBS,*† Kiran Virik, MBBS, FRACP,*† Wojciech Morzycki, MD, FRCPC,*† and Nathalie Saint-Jacques, MSc‡
Introduction: Adjuvant chemotherapy in non-small cell lung cancer (NSCLC) has become a new standard of care. This study examines the uptake patterns for adjuvant chemotherapy outside of clinical trials. Methods: A retrospective study of all patients diagnosed with NSCLC in the year 2005 who underwent curative-intent surgery in Nova Scotia, Canada was conducted. Logistic regression models and discriminant function analyses were employed to identify cofactors associated with referral to medical oncology and/or utilization of adjuvant chemotherapy. Results: Of 540 patients with NSCLC, 108 underwent curativeintent surgery (67% lobectomy; 15% pneumonectomy; 19% wedge resection) for NSCLC (39% IA; 24% IB; 25% II; 14% III). Referral to medical oncology was observed in 44% (47 of 108) of all patients including 73% (30 of 41) of those with stage II–III. Adjuvant chemotherapy utilization was observed in 62% (29 of 47) of those referred including 73% (22 of 30) of those with stage II–III. Overall, 27% (29 of 108) of all patients received adjuvant chemotherapy, including 54% (22 of 41) of those with stage II–III. Higher uptake was significantly associated with age (younger versus older), stage (II/III versus I), and surgery type (pneumonectomy versus wedge). Weaker associations were observed with other cofactors including surgeon, health center, mean household income, and surgery-medical oncologist consult timeline. Conclusions: The uptake of adjuvant chemotherapy in patients with resected NSCLC outside of clinical trials is low overall, but is higher among younger patients and those with more advanced stages. These uptake patterns may allow future planning of health resource utilization and/or improvement of chemotherapy utilization rates.
*Division of Medical Oncology, Queen Elizabeth II Health Sciences Centre, Halifax, Nova Scotia, Canada; †Department of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada; and ‡Surveillance and Epidemiology Unit, Cancer Care Nova Scotia, Halifax, Nova Scotia, Canada. Disclosure: The authors declare no conflicts of interest. Supported by Surveillance and Epidemiology Unit and the Lung Cancer Site Team of Cancer Care Nova Scotia. Presented as a poster presentation at the 2007 annual meeting of the American Society of Clinical Oncology in Chicago, USA. Address for correspondence: Tallal Younis, MBBCh, FRCP (UK), Queen Elizabeth II Health Sciences Centre, Division of Medical Oncology, 455 Bethune Building at 1278 Tower Road, Halifax, NS, Canada B3H 2Y9. E-mail: [email protected] Copyright © 2008 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/08/0311-1272
1272
Key Words: Non-small cell lung cancer, Adjuvant chemotherapy, Treatment uptake. (J Thorac Oncol. 2008;3: 1272–1278)
P
atients undergoing curative-intent surgical resection for early-stage breast, colorectal and lung cancers represent a potentially curable population. For those patients with higher risk disease, adjuvant chemotherapy has been shown to improve overall survival.1–5 However, the overall impact of adjuvant chemotherapy in cancer patients ultimately depends upon its uptake in unselected patient populations outside of clinical trials. The uptake of adjuvant chemotherapy in breast and colorectal cancers has been previously examined.6 –14 Guidelines currently recommend adjuvant chemotherapy for patients with node-positive or high-risk node-negative breast cancer and for patients with stage III and possibly high-risk stage II colorectal cancer.15,16 Approximately 30% and 50% of patients with non metastatic breast cancer were treated with adjuvant chemotherapy in the mid 1990s and in the year 2000, respectively.6 –9 Higher uptake was associated with younger age, more advanced stage disease, and negative hormone receptor status. As well, approximately 30% of patients with non metastatic colorectal cancer were treated with adjuvant chemotherapy.10 –14 Higher uptake was associated with younger age, more advanced stages (e.g., 70% uptake in patients with stage III disease) and fewer comorbidities. A number of recent clinical studies have revealed that adjuvant chemotherapy improves overall survival for patients with higher risk non-small cell lung cancer (NSCLC) following curative-intent surgical resection.17–24 Current guidelines recommend adjuvant platinum based doublet chemotherapy for patients with resected stage II–IIIA NSCLC, while evidence is less clear for stage IB disease.25,26 The overall survival impact of adjuvant chemotherapy, however, will depend upon its uptake outside of these clinical trials. We examined the uptake patterns of adjuvant chemotherapy in NSCLC patients including; (1) the rates of referral to medical oncology (MO) and/or utilization of adjuvant chemotherapy, and (2) the factors associated with a higher referral to MO and/or utilization of adjuvant chemotherapy.
PATIENTS AND METHODS The study cohort included all patients diagnosed with NSCLC in the year 2005 who underwent curative-intent
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
surgical resection for stage I–III disease in Nova Scotia, Canada. Patients were identified through the Nova Scotia Cancer Registry and retrospective chart reviews at the two health centers in Nova Scotia where thoracic surgeries are exclusively performed: (1) the Queen Elizabeth II Health Sciences Centre in Halifax or the (2) Cape Breton Regional Hospital (CBRH) in Sydney. A retrospective review was conducted to abstract patient demographics as well as disease and treatment characteristics from radiologic, surgical, and pathology reports contained in patient charts and the Oncology Patient Information System (a database maintained by the Nova Scotia Cancer Registry and the regional cancer centers). Data quality in Oncology Patient Information System is ensured through online system edits, routine edits, and periodic chart audits. Dates of care events abstracted included: (i) disease detection (Detection), (ii) definitive curative-intent surgery (Surgery), and (iii) first MO consultation (MO Consult). Disease detection was defined as the first abnormal imaging study that prompted a referral for surgery consultation. Referral to MO (Referral) and adjuvant chemotherapy treatment (Treatment) were noted. The study cohort was divided into three groups based on Referral and Treatment states; (A) not referred, (B) referred-not treated, and (C) referred-treated. The rates of “Referral” (referred/resected) and “Treatment” (treated/referred) as well as “Overall Uptake” of adjuvant chemotherapy (treated/resected) were computed. Any identified reasons for no referral or no treatment were abstracted. A logistic regression (LR) model with a stepwise selection (p ⬍ 0.05) was used to identify the primary cofactors influencing Referral and Treatment. Cofactors were entered into the stepwise selection analysis only if their univariate probability of nonrandom association was ⬍0.3. The examined cofactors included: patient sex (Sex), age at diagnosis (Age), marital status (Marital Status), smoking history (Smoking), lung cancer subtype (Histopathology), stage of disease (Stage), margin status (Margin), the presence or absence of ischemic heart disease or diabetes mellitus (Comorbidity), performance status (PS) based on the Eastern Cooperative Oncology Group scale, health center where surgically treated (Health Centre), type of definitive surgery (Surgery Type), presence or absence of postoperative complications (PO Complication), surgeon (Surgeon), medical oncologist (Med Onc), distance of residence to a cancer center derived from address at diagnosis (Distance), median household income (MHI) in the area of patient residence, mean level of education in the area of patient residence (Education), and the time elapsed between Detection-Surgery (Detection-Surgery elapsed time) and Surgery-Medical Oncologist Consult (Surgery-MO Consult elapsed time). Timelines between these two care intervals were adjusted for cofactors and expressed as calendar days.27 Underlying lung cancer histopathologies were squamous carcinoma, adenocarcinoma, and other (large cell and bronchio-alveolar carcinomas). Mixed NSCLC-small cell lung cancers were excluded. Aggregate census data from 1996 and 2001 were used to compute the socio-economic factors. Variations in MO prac-
Adjuvant Chemotherapy Uptake in NSCLC
tices could not be accounted within multifactorial analyses due to the small number of entries (⬍5) in each factor level. Discriminant function analysis (DFA) was used to verify the robustness of the results obtained by LR, as it is considered preferable to stepwise LR for cofactor selection in data sets with small sample size.28 DFA is a multivariate statistical method that finds the linear combination (the discriminant function) of attributes (cofactors) that best separates two groups (e.g., not referred versus referred).28 An analysis of variance was used to test for the significance of the difference between the group centroids (means). A 2 test with Yates’ continuity correction was used to determine the predictive value of the discriminant function (i.e., the capacity of the resulting combination of cofactors to predict either Referral or Treatment patterns).
RESULTS Of 540 patients who were diagnosed with NSCLC in Nova Scotia, Canada in the year 2005, a total of 108 patients (20%) underwent curative-intent surgical resection (lobectomy 67%,; pneumonectomy 15%; wedge resection 18%) for stage I–III disease (39% IA; 23% IB; 24% II; 14% III). Their median age was 66 years, and 56% were males. Other characteristics are described in Table 1. Referral and Treatment patterns are shown in Figure 1. After curative-intent surgery, 44% (47 of 108) of all patients were referred to MO for consideration of adjuvant chemotherapy including 73% (30 of 41) of those with stage II–III disease. Of all those referred, 62% (29 of 47) were treated with adjuvant chemotherapy including 73% (22 of 30) of those with stage II–III disease. Overall, adjuvant chemotherapy was prescribed in 27% (29 of 108) of all patients after curative-intent surgery including 54% (22 of 41) of those with stage II–III disease. All patients received platinumbased doublet chemotherapy including paclitaxel and carboplatin (23 of 29), vinorelbine and cisplatin (5 of 29) or vinorelbine and carboplatin (1 of 29). Physician recommendation and patient choice were the most common identified causes for “no referral” and “no treatment” (Table 2). Of 61 patients not referred to MO, surgeons did not recommend referral in 64% while 5% refused referral. Of 18 patients who were referred but not treated, medical oncologists did not recommend treatment in 39% while 39% refused treatment. Overall, patients with stage IA disease represented 59% (33 of 56) of all patients for whom chemotherapy was not recommended by surgeons and/or medical oncologists. A number of cofactors were associated with uptake patterns by univariate analysis (Table 3). Younger age (⬍65 versus 65–75 versus ⬎75 years), more advanced stage (II–III versus IB versus IA), and more extensive surgery (pneumonectomy versus lobectomy versus wedge resection) were significant predictors of higher Referral, Treatment, and Overall Uptake. Practice related variables, including Health Centre and/or Surgeon, were also associated with Referral and Overall Uptake. Socio-economic variables, including Distance and MHI, were weakly associated with Referral; patients living closer to cancer center (p ⫽ 0.049) and in
Copyright © 2008 by the International Association for the Study of Lung Cancer
1273
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
Younis et al.
TABLE 1.
Characteristics of Study Population Cofactors: No. (%)
Age (yr) ⬍65 65–75 ⬎75 Sex Male Female Marital status Married Single Smoking Current Former Never Education ⱕMedian ⬎Median MHI ($) Low (ⱕ40K) High (⬎40K) Distance (km) Near (ⱕ32) Far (⬎32)
46 (43) 40 (37) 22 (20) 61 (56) 47 (44) 75 (86) 12 (14) 36 (35) 55 (53) 13 (12) 44 (41) 64 (59) 47 (44) 61 (56)
Histopathology Adenocarcinoma Squamous Other Stage IA IB II IIIa Margin Negative Positive Comorbidity Yes No PS 0–1 II–III Health centre QEII HSC CBRH
46 (43) 35 (32) 27 (25) 42 (39) 25 (23) 26 (24) 15 (14) 99 (97) 3 (3) 34 (33) 69 (67) 77 (71) 31 (29) 97 (90) 11 (10)
Surgeon A B C D Surgery type Wedge Lobectomy Pneumonectomy PO complications No Yes Med oncologist A B C D E F G
34 (32) 26 (24) 10 (9) 38 (35) 20 (18) 72 (67) 16 (15) 69 (64) 39 (36) 7 (16) 3 (7) 2 (5) 12 (28) 5 (12) 7 (16) 7 (16)
56 (52) 52 (48)
a Two patients underwent curative-intent surgery for stage IIIB disease, and were included in this cohort. MHI, median household income; K, thousands; km, kilometer; PS, performance status; QEII HSC, Queen Elizabeth II Health Sciences Centre; CBRH, Cape Breton Regional Hospital; PO Complications, postoperative complications; Med Onc, medical oncologist.
FIGURE 1. Uptake patterns. Number of patients diagnosed with Non-small Cell Lung Cancer (NSCLC) and those who underwent surgical resection over 1 year in Nova Scotia, Canada are shown (all stages or stage II–III). The circles show numbers (n) and percentages (%) of patients according to Referral (referred/resected), Treatment (treated/referred) and Overall Uptake (treated/ resected).
areas of higher MHI (p ⫽ 0.049) were more likely to be referred. There were no statistically significant associations between uptake patterns and other cofactors including Sex, Marital Status, Smoking, Histopathology, Margin, Comorbidity, PS, PO Complication, Med Onc or Education. As well, there were no significant associations observed between “Detection-Surgery” elapsed time (median 108 days) and uptake patterns. However, an inverse association was observed between “Surgery-MO Consult” elapsed time (median 49 days) and Treatment; referred patients who experienced shorter versus longer wait times to MO consult (⬍49 versus ⱖ49 days) were more likely treated (76% versus 46%, respectively: p ⫽ 0.039). Referral and Treatment patterns were examined further using both multifactorial (LR) and multivariate (DFA) anal-
1274
yses. Based on LR, independent predictors of Referral (Table 4) were Stage (P ⬍ 0.001), Age (p ⫽ 0.005), Surgery Type (p ⫽ 0.005), Surgeon (P ⫽ 0.03), and MHI (p ⫽ 0.02). Independent predictors of Treatment were Age (OR ⫽ 2.3; 95% CI, 1.1– 4.76; p ⫽ 0.016) as well as Stage and Surgery Type: none of the patients who had stage IA disease or underwent wedge surgery were treated with adjuvant chemotherapy. DFA showed very similar Referral and Treatment patterns (Figure 2). Higher likelihood of Referral was associated with Stage (II/III versus I), Age (younger versus older), Surgery Type (pneumonectomy versus wedge), Health Centre (CBRH versus NSCC), and MHI (higher versus lower). Higher likelihood of Treatment was associated with Stage (II/III versus I), Age (younger versus older), Surgery Type (pneumonectomy versus wedge) and “Surgery-MO Consult”
Copyright © 2008 by the International Association for the Study of Lung Cancer
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
TABLE 2. Identified Causes for “No Referral” and “No Treatment” by Stage Disease Stage
TABLE 4.
Underlying Causes
IA
IB
II–III
Total No. of Patients
No referrala
Not recommended Refused by patient Unknown/other Total Not recommended Refused by patient Unknown/other Total
30 0 8 38 3 0 1 4
4 2 6 12 0 4 2 6
5 1 5 11 4 3 1 8
39 3 19 61 7 7 4 18
a b
Causes of No referral in all patients. Causes of No treatment in referred-not treated cohort.
TABLE 3. Uptake Patterns Based on Univariate Analyses Referral Cofactorsb Age (yr) ⬍65 65–75 ⬎75 Stage IA IB II–III Surgery type Wedge Lobectomy Pneumonectomy Health centre QEII HSC CBRH Surgeon A B C D MHI ($) Low (ⱕ40K) High (⬎40K) Distance (km) Near (ⱕ32) Far (⬎32)
%
pc
%
0.019 59 35 27
0 24 75 0.731
60 55 0.087
0.037 24 55
0.066 80 30 75 57
0.049 32 53
⬍0.001
0.022
0.006
44 39 80 37
⬍0.001 0 28 54
0 59 80
39 82
0.002
0.009
⬍0.001
0.254
0.516 23 30
0.749 60 65
Cofactorsa Age Per decade Stage IA IB II–III Surgery type Lobectomy Wedge Pneumonectomy Surgeon A B C D MHI Low (ⱕ40 K) High (⬎40 K)
0.276 32 21
a An inverse association between Treatment and “Surgery-Medical Oncology Consult” elapsed time was also observed (p ⫽ 0.039). b Results shown include only those cofactors that predict at least one uptake pattern at p value ⬍0.05. c p-values associated with the likelihood ratio by 2 test (2). MHI, median household income; K, thousands; km, kilometer; PS, performance status; QEII HSC, Queen Elizabeth II Health Sciences Centre; CBRH, Cape Breton Regional Hospital.
elapsed time (shorter versus longer). These DFA models correctly predicted 80% and 85% of observed Referral and Treatment patterns, respectively. Misclassifications in Refer-
Yes
No
Odds Ratio
95% CI
47
61
2.6
4 13 30
38 12 11
1.0 14.9 27.4
— (2.9–76.9) (5.8–130.7)
29 3 15
43 17 1
1.0 0.5 40.4
— (0.1–2.8) (2.3–705.9)
15 10 8 14
19 16 2 24
1.0 0.14 3.3 0.7
— (0.03–0.8) (0.3–31.7) (0.2–3.1)
15 32
32 29
1.0 4.7
— (1.3–17.8)
P-valueb 0.005
(1.3–5.5) ⬍0.001
0.005
0.028
0.015
MHI, median household income; K, thousands; km, kilometer; No., Number. a Results shown include only those cofactors that predict Referral pattern at p-value ⬍0.05. b p-values associated with the likelihood ratio by 2 test (2).
ral (Figure 2A) were mainly due to the presence of patients with early stage disease among the referred population and patients with advanced stage disease among the nonreferred population. Misclassifications in Treatment (Figure 2B) were mainly due to the presence of patients with older age and stage IB disease among treated population and a patient living in an area with the highest MHI among non treated population.
0.013 35 12 60 21
73 56 0.049
pc
44 18 9
0 54 73
15 40 94
%
0.003
⬍0.001
54 33
pc
74 50 33
10 52 73
Overall Uptake
Treatmenta
Referral Pattern Based on Multifactorial Analysis Patients Referred (No.)
Uptake Pattern
No treatmentb
Adjuvant Chemotherapy Uptake in NSCLC
DISCUSSION Adjuvant chemotherapy has recently become a new standard of care for patients with NSCLC after clinical trials showed approximately 10% improvement in overall survival for those with higher risk disease (i.e., stage II and IIIA).3–5 In this population-based study, approximately one third of the patients who underwent curative-intent surgical resection for NSCLC were subsequently treated with adjuvant chemotherapy. These uptake rates may seem comparable to that in breast or colorectal cancers.6 –14 However, there are far fewer patients who undergo curative-intent surgical resection for NSCLC compared with other cancers as stage distribution in lung cancer is shifted towards more advanced stages.29 –32 In this study, only 20% of the patients with NSCLC underwent curative-intent surgical resection. Adjuvant chemotherapy therefore, would be expected to result in a modest less than 1% improvement in the current 5-year overall survival rate of 15% in lung cancer.30 –32 Our finding of a lower uptake of chemotherapy in older patients has also been observed in studies examining other cancers.33–35 This could reflect concerns about higher toxic-
Copyright © 2008 by the International Association for the Study of Lung Cancer
1275
Younis et al.
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
FIGURE 2. Discriminant Function Analysis (DFA) Results. A, shows discriminant function (LD1) that best separates patients referred from those not referred. Patients most likely to be referred had advanced stage (Stage: II/III versus I), were younger (Age: younger versus older), underwent pneumonectomy (Surgery Type: pneumonectomy versus wedge), treated at the CBRH (Health Centre: CBRH versus NSCC), or lived in area of higher MHI (MHI: higher versus lower). Factors loadings (relative magnitude of influence) were as follows: Stage (0.87), Age (⫺0.38), Surgery Type (0.29), Health Centre (⫺0.30), and MHI (0.29). B, shows discriminant function (LD1) that best separates patients referred and treated from those referred and not treated. Patients most likely to be treated had advanced stage (Stage: II/III versus I), were younger (Age: younger versus older), underwent pneumonectomy (Surgery Type: pneumonectomy versus wedge), and experienced shorter “Surgery-MO Consult” elapsed time (Timeline: shorter versus longer). Factors loadings (relative magnitude of influence) were as follows: Stage (0.53), Age (⫺0.40), Surgery Type (⫺0.37), and Surgery-MO Consult elapsed time (⫺0.47).
ity, paucity of data, and possible lack of clear benefit in older patients who are often excluded from clinical trials because of associated comorbidities and poorer PS. Treatment decisions concerning palliative chemotherapy in NSCLC are often influenced by PS rather than age alone.36 For adjuvant chemotherapy, a subset analysis of the NCIC BR 10 study36 revealed a significant overall survival benefit with adjuvant chemotherapy for patients over the age of 65 years (HR ⫽ 0.61; 95% CI, 0.38 – 0.98; p ⫽ 0.04), which was comparable to that observed in younger patients, and an acceptable toxicity. Patients over 75 years of age, however, had a worse overall survival compared with that of all patients enrolled in the trial (HR ⫽ 2.35; 95% CI, 0.84 – 6.58; p ⫽ 0.09) though sample size was too small to draw any firm conclusions in this elderly age subgroup. The higher uptake of adjuvant chemotherapy in stage II–III NSCLC reflects the current evidence which strongly supports treatment provision in stages II–IIIA disease.25 In our study, approximately three fourths of patients who underwent surgery for stage II–III NSCLC were referred to MO and three fourths of those referred were treated (i.e., approximately 50% Overall Uptake). For patients with stage IB disease, approximately half of patients who underwent sur-
1276
gery were referred and half of those referred were treated (i.e., approximately 25% Overall Uptake). Currently, the role of chemotherapy in stage IB disease is less clear,25,26 but a subset analysis of the CALGB 9633 study suggest that adjuvant chemotherapy could be considered for stage IB disease with larger size tumors.23,24 Conversely, there is no evidence to support platinum based chemotherapy in stage IA disease and the LACE meta-analysis suggest a potential detrimental effect in this patient subgroup (HR ⫽ 1.41; 95% CI, 0.96 –2.09).5 In our study, only 10% of patients with stage IA disease were referred to MO but none were treated with adjuvant chemotherapy. Decision making such as that concerning referral and/or treatment is complex and involves weighing a number of variables that are unique to every health care system-providerpatient encounter. As this study illustrates, uptake patterns were influenced by a number of variables that are related to patient (e.g., Age), disease (e.g., Stage or Surgery Type), practice (e.g., Surgeon or Health Centre), or socioeconomic status (e.g., MHI). The DFA models we present illustrate the combined impact of these variables in predicting referral and/or treatment but the interaction among these variables may be too complex to allow for identifying exact causes for
Copyright © 2008 by the International Association for the Study of Lung Cancer
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
no referral and/or no treatment. Moreover, studies that assess uptake of newer interventions should be considered to provide general predictors of outcomes and should not be simply regarded as performance measures. The uptake of adjuvant chemotherapy in NSCLC was reported in another retrospective study conducted at a single tertiary center in Ontario, Canada.38 Referral to MO and/or utilization of chemotherapy during the later period of that study (June 2004 –May 2005) seems comparable to that observed in our study (January 2005–December 2005); 63% of all patients were referred to MO and 54% of those referred were treated (i.e., 34% Overall Uptake). Possible reasons for not being referred and/or not being treated were retrospectively abstracted from patients’ charts, but statistical analysis to identify predictors of referral and/or treatment was not conducted. Reasons for not being referred to MO included; stage IA disease (37%), patient choice (18%), surgeon recommendation (15%), significant comorbidities (10%), PO complications (5%), disease progression (3%), and advanced age (2%). Reasons for not prescribing adjuvant chemotherapy included patient refusal (50%), comorbidities (14%), stage IA disease (10%), disease progression (6%), advanced age (6%), and unknown (14%). Our study has limitations. Firstly, the number of cases meeting the inclusion criteria was relatively small reducing the overall statistical power of the study and limiting the stability of weak associations. However, we employed both multifactorial and discriminant function analyses and the results were almost identical. Secondly and perhaps more importantly, the retrospective nature of our study did not allow us to determine the exact causes of no referral and/or no treatment. However, the predictors of referral and/or treatment identified may help improve future chemotherapy uptake rates. Thirdly, our study examined the uptake patterns in the first year (i.e., 2005) after adjuvant chemotherapy for NSCLC was accepted as a new standard of care. It is possible that uptake patterns may have changed since then, as further updated data from clinical trials were reported, and/or mature studies were fully published. In summary, approximately one third of patients who underwent curative-intent surgery for NSCLC were treated with adjuvant chemotherapy. Consistent with other reports in breast and colorectal cancers, higher uptake of adjuvant chemotherapy was observed in younger patients and those with more advanced disease stages. Improved referral to MO may increase the overall uptake of adjuvant chemotherapy as a high proportion of those referred were treated. Prospective assessment, however, is still required to better determine the exact causes of no referral and/or no treatment in clinical decision making that is unique to every health care systemprovider-patient encounter. In the interim, the uptake patterns observed in this study may help future planning of health resource utilization and/or improvement of chemotherapy uptake rates.
ACKNOWLEDGMENTS The authors would like to thank Ron Dewar and Maureen MacIntyre for their helpful suggestions throughout this study.
Adjuvant Chemotherapy Uptake in NSCLC
REFERENCES 1. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG): effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomized trials. Lancet 2005;365:1687–1717. 2. Wolpin BM, Meyerhardt JA, Mamon HJ, et al. Adjuvant treatment of colorectal cancer. CA Cancer J Clin 2007;57:168 –185. 3. Hotta K, Matsuo K, Ueoka H, et al. Role of adjuvant chemotherapy in patients with resected non-small-cell lung cancer: reappraisal with a meta-analysis of randomized controlled trials. J Clin Oncol 2004;22: 3860 –3867. 4. Sedrakyan A, Van Der Meulen J, O’Byrne K, et al. Postoperative chemotherapy for non-small cell lung cancer: a systematic review and meta-analysis. J Thorac Cardiovasc Surg 2004;128:414 – 419. 5. Pignon JP, Triboder GV, Scagliotti G, et al. Lung Adjuvant Cisplatin Evaluation (LACE): a pooled analysis of five randomized clinical trials including 4,584 patients. J Clin Oncol 2006;24(18 Suppl):366S (abstr. #7008). 6. Harlan LC, Abrams J, Warren JL, et al. Adjuvant therapy for breast cancer: practice patterns of community physicians. J Clin Oncol 2002; 20:1809 –1817. 7. Du X, Goodwin JS. Patterns of use of chemotherapy for breast cancer in older women: findings from Medicare claims data. J Clin Oncol 2001; 19:1455–1461. 8. Du X, Key C, Osborne C, et al. Discrepancy between consensus recommendations and actual community use of adjuvant chemotherapy in women with breast cancer. Ann Intern Med 2003;138:90 –97. 9. Edwards BK, Brown ML, Wingo PA, et al. Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst 2005;97:1407–1427. 10. Schrag D, Gelfand SE, Bach PB, et al. Who gets adjuvant treatment for stage II and III rectal cancer? Insight from surveillance, epidemiology, and end results–medicare. J Clin Oncol 2001;19:3712–3718. 11. Ong S, Watters JM, Grunfeld E, et al. Predictors of uptake of adjuvant therapy in CRC. Can J Surg 2005;48:225–229. 12. Jessup JM, Stewart A, Greene FL, et al. Adjuvant chemotherapy for stage iii colon cancer: implications of race/ethnicity, age, and differentiation. JAMA 2005;294:2703–2711. 13. Potosky AL, Harlan LC, Kaplan RS, et al. Age, sex, and racial differences in the use of standard adjuvant therapy for colorectal cancer. J Clin Oncol 2002;20:1192–1202. 14. Ayanian JZ, Zaslavsky AM, Fuchs CS, et al. Use of adjuvant therapy for CRC in a population-based cohort. J Clin Oncol 2003;21:1293–1300. 15. American Society of Clinical Oncology. Clinical Practice Guidelines (http://www.asco.org). 16. National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology (http://www.ncccn.org). 17. Keller SM, Adak S, Wanger H, et al. A randomized trial of postoperative adjuvant therapy in patients with completely resected Stage II or IIIA non-small cell lung cancer (Eastern Cooperative Oncology Group). N Engl J Med 2000;343:1217–1222. 18. Scagliotti GV, Fossati R, Torri V, et al. Randomized study of adjuvant chemotherapy for completely resected stage I, II, or IIIA non-small-cell Lung cancer. J Natl Cancer Inst 2003;95:1453–1461. 19. Waller D, Peake MD, Stephens RJ, et al. Chemotherapy for patients with non-small cell lung cancer: the surgical setting of the Big Lung Trial. Eur J Cardiothorac Surg 2004;26:173–182. 20. Arriagada R, Bergman B, Dunant A, et al. (for The International Adjuvant Lung Cancer Trial Collaborative Group). Cisplatin based adjuvant chemotherapy in patients with completely resected non-smallcell lung cancer. N Engl J Med 2004;350:351–360. 21. Winton TL, Livingstan R, Johnson D, et al. A prospective randomized trial of adjuvant vinorelbine (VNR) and cisplatin (CIS) in completely resected stage Ib and II non small cell lung cancer (NSCLC) Intergroup JBR 10. N Engl J Med 2005;352:2589 –2597. 22. Douillard JY, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation patients with completely resected stage IB–IIIA non-small-cell lung cancer (Adjuvant Navelbine International Trialist Association [ANITA]): a randomised controlled trial. Lancet Oncol 2006;7:719 –727. 23. Strauss GM, Hernden J, Maddaus MA, et al. Randomized clinical trial of
Copyright © 2008 by the International Association for the Study of Lung Cancer
1277
Younis et al.
24.
25.
26. 27. 28.
29. 30.
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
adjuvant chemotherapy with paclitaxel and carboplatin following resection in Stage Ib Non-Small Cell Lung Cancer (NSCLC): report of cancer and leukemia group B (CALGB) protocol 9633. J Clin Oncol 2004;22: 621S (abstr. #7019). Strauss GM, Hernden J, Maddaus MA, et al. Adjuvant chemotherapy in Stage IB non-small cell lung cancer (NSCLC): update on Cancer and Leukemia Group B (CALGB) protocol 9633. J Clin Oncol 2006;24(18 Suppl):365S (abstr. #7007). Pisters KM, Evans WK, Azzoli CG, et al. Cancer Care Ontario and American Society of Clinical Oncology adjuvant chemotherapy and adjuvant radiation therapy for stages I–IIIA resectable non small-cell lung cancer guideline. J Clin Oncol 2007;25:5506 –5518. Wakelee H, Dubey S, Gandara D. Optimal adjuvant therapy for nonsmall cell lung cancer– how to handle stage I disease. Oncologist 2007;12:331–337. Saint-Jacques N, Rayson D, Al-Fayea T, et al. Waiting times in early-stage non-small cell lung cancer (NSCLC). J Thorac Oncol 2008;3:865– 870. O’Gorman TW, Woolson RF. Variable Selection to Discriminate Between Two Groups: Stepwise Logistic Regression or Stepwise Discriminant Analysis? In The American Statistician: American Statistical Association, 1991. Pp. 187–193. Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710 –1717. SEER: Surveillance, Epidemiology, and End Results Program. National Cancer Institute. Cancer Statistics 2005 (www.seer.cancer.gov). Accessed October 10, 2008.
1278
31. American Cancer Society: Cancer Facts and Figures 2006. Atlanta, GA, American Cancer Society, 2006 (http://americancancersociety.org). Accessed October 10, 2008. 32. Canadian Cancer Society, National Cancer Institute of Canada. Canadian Cancer Statistics, 2007. Toronto, Canada 2007 (www.cancer.ca). Accessed October 10, 2008. 33. Fisher B. Highlights from recent National Surgical Adjuvant Breast and Bowel Project studies in the treatment and prevention of breast cancer. CA Cancer J Clin 1999;49:159 –177. 34. Du XL, Jones DV, Zhang D. Effectiveness of adjuvant chemotherapy fornode-positive operable breast cancer in older women. J Gerontol A Biol Sci Med Sci 2005;60:1137–1144. 35. Giordano SH, Duan Z, Kuo Y-F, et al. Use and outcomes of adjuvant chemotherapy in older women with breast cancer. J Clin Oncol 2006; 24:2750 –2756. 36. Gridelli C. The ELVIS trial: a phase III study of single-agent vinorelbine as first-line treatment in elderly patients with advanced non-small cell lung cancer. Elderly Lung Cancer Vinorelbine Italian Study. Oncologist 2001;1:(6 Suppl) 4 –7. 37. Pepe C, Hasan B, Winton T, et al. Adjuvant vinorelbine and cisplatin in elderly patients: National Cancer Institute of Canada and Intergroup Study JBR. 10. J Clin Oncol 2007;25:1553–1561. 38. Kassam F, Shepherd FA, Johnston M, et al. Referral patterns for adjuvant chemotherapy in patients with completely resected non-small cell lung cancer. J Thorac Oncol 2007;2:39 – 43.
Copyright © 2008 by the International Association for the Study of Lung Cancer
Adjuvant Chemotherapy Uptake in Non-small Cell Lung Cancer Tallal Younis, MBBCh, FRCP (UK),*† Turki Al-Fayea, MBBS,*† Kiran Virik, MBBS, FRACP,*† Wojciech Morzycki, MD, FRCPC,*† and Nathalie Saint-Jacques, MSc‡
Introduction: Adjuvant chemotherapy in non-small cell lung cancer (NSCLC) has become a new standard of care. This study examines the uptake patterns for adjuvant chemotherapy outside of clinical trials. Methods: A retrospective study of all patients diagnosed with NSCLC in the year 2005 who underwent curative-intent surgery in Nova Scotia, Canada was conducted. Logistic regression models and discriminant function analyses were employed to identify cofactors associated with referral to medical oncology and/or utilization of adjuvant chemotherapy. Results: Of 540 patients with NSCLC, 108 underwent curativeintent surgery (67% lobectomy; 15% pneumonectomy; 19% wedge resection) for NSCLC (39% IA; 24% IB; 25% II; 14% III). Referral to medical oncology was observed in 44% (47 of 108) of all patients including 73% (30 of 41) of those with stage II–III. Adjuvant chemotherapy utilization was observed in 62% (29 of 47) of those referred including 73% (22 of 30) of those with stage II–III. Overall, 27% (29 of 108) of all patients received adjuvant chemotherapy, including 54% (22 of 41) of those with stage II–III. Higher uptake was significantly associated with age (younger versus older), stage (II/III versus I), and surgery type (pneumonectomy versus wedge). Weaker associations were observed with other cofactors including surgeon, health center, mean household income, and surgery-medical oncologist consult timeline. Conclusions: The uptake of adjuvant chemotherapy in patients with resected NSCLC outside of clinical trials is low overall, but is higher among younger patients and those with more advanced stages. These uptake patterns may allow future planning of health resource utilization and/or improvement of chemotherapy utilization rates.
*Division of Medical Oncology, Queen Elizabeth II Health Sciences Centre, Halifax, Nova Scotia, Canada; †Department of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada; and ‡Surveillance and Epidemiology Unit, Cancer Care Nova Scotia, Halifax, Nova Scotia, Canada. Disclosure: The authors declare no conflicts of interest. Supported by Surveillance and Epidemiology Unit and the Lung Cancer Site Team of Cancer Care Nova Scotia. Presented as a poster presentation at the 2007 annual meeting of the American Society of Clinical Oncology in Chicago, USA. Address for correspondence: Tallal Younis, MBBCh, FRCP (UK), Queen Elizabeth II Health Sciences Centre, Division of Medical Oncology, 455 Bethune Building at 1278 Tower Road, Halifax, NS, Canada B3H 2Y9. E-mail: [email protected] Copyright © 2008 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/08/0311-1272
1272
Key Words: Non-small cell lung cancer, Adjuvant chemotherapy, Treatment uptake. (J Thorac Oncol. 2008;3: 1272–1278)
P
atients undergoing curative-intent surgical resection for early-stage breast, colorectal and lung cancers represent a potentially curable population. For those patients with higher risk disease, adjuvant chemotherapy has been shown to improve overall survival.1–5 However, the overall impact of adjuvant chemotherapy in cancer patients ultimately depends upon its uptake in unselected patient populations outside of clinical trials. The uptake of adjuvant chemotherapy in breast and colorectal cancers has been previously examined.6 –14 Guidelines currently recommend adjuvant chemotherapy for patients with node-positive or high-risk node-negative breast cancer and for patients with stage III and possibly high-risk stage II colorectal cancer.15,16 Approximately 30% and 50% of patients with non metastatic breast cancer were treated with adjuvant chemotherapy in the mid 1990s and in the year 2000, respectively.6 –9 Higher uptake was associated with younger age, more advanced stage disease, and negative hormone receptor status. As well, approximately 30% of patients with non metastatic colorectal cancer were treated with adjuvant chemotherapy.10 –14 Higher uptake was associated with younger age, more advanced stages (e.g., 70% uptake in patients with stage III disease) and fewer comorbidities. A number of recent clinical studies have revealed that adjuvant chemotherapy improves overall survival for patients with higher risk non-small cell lung cancer (NSCLC) following curative-intent surgical resection.17–24 Current guidelines recommend adjuvant platinum based doublet chemotherapy for patients with resected stage II–IIIA NSCLC, while evidence is less clear for stage IB disease.25,26 The overall survival impact of adjuvant chemotherapy, however, will depend upon its uptake outside of these clinical trials. We examined the uptake patterns of adjuvant chemotherapy in NSCLC patients including; (1) the rates of referral to medical oncology (MO) and/or utilization of adjuvant chemotherapy, and (2) the factors associated with a higher referral to MO and/or utilization of adjuvant chemotherapy.
PATIENTS AND METHODS The study cohort included all patients diagnosed with NSCLC in the year 2005 who underwent curative-intent
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
surgical resection for stage I–III disease in Nova Scotia, Canada. Patients were identified through the Nova Scotia Cancer Registry and retrospective chart reviews at the two health centers in Nova Scotia where thoracic surgeries are exclusively performed: (1) the Queen Elizabeth II Health Sciences Centre in Halifax or the (2) Cape Breton Regional Hospital (CBRH) in Sydney. A retrospective review was conducted to abstract patient demographics as well as disease and treatment characteristics from radiologic, surgical, and pathology reports contained in patient charts and the Oncology Patient Information System (a database maintained by the Nova Scotia Cancer Registry and the regional cancer centers). Data quality in Oncology Patient Information System is ensured through online system edits, routine edits, and periodic chart audits. Dates of care events abstracted included: (i) disease detection (Detection), (ii) definitive curative-intent surgery (Surgery), and (iii) first MO consultation (MO Consult). Disease detection was defined as the first abnormal imaging study that prompted a referral for surgery consultation. Referral to MO (Referral) and adjuvant chemotherapy treatment (Treatment) were noted. The study cohort was divided into three groups based on Referral and Treatment states; (A) not referred, (B) referred-not treated, and (C) referred-treated. The rates of “Referral” (referred/resected) and “Treatment” (treated/referred) as well as “Overall Uptake” of adjuvant chemotherapy (treated/resected) were computed. Any identified reasons for no referral or no treatment were abstracted. A logistic regression (LR) model with a stepwise selection (p ⬍ 0.05) was used to identify the primary cofactors influencing Referral and Treatment. Cofactors were entered into the stepwise selection analysis only if their univariate probability of nonrandom association was ⬍0.3. The examined cofactors included: patient sex (Sex), age at diagnosis (Age), marital status (Marital Status), smoking history (Smoking), lung cancer subtype (Histopathology), stage of disease (Stage), margin status (Margin), the presence or absence of ischemic heart disease or diabetes mellitus (Comorbidity), performance status (PS) based on the Eastern Cooperative Oncology Group scale, health center where surgically treated (Health Centre), type of definitive surgery (Surgery Type), presence or absence of postoperative complications (PO Complication), surgeon (Surgeon), medical oncologist (Med Onc), distance of residence to a cancer center derived from address at diagnosis (Distance), median household income (MHI) in the area of patient residence, mean level of education in the area of patient residence (Education), and the time elapsed between Detection-Surgery (Detection-Surgery elapsed time) and Surgery-Medical Oncologist Consult (Surgery-MO Consult elapsed time). Timelines between these two care intervals were adjusted for cofactors and expressed as calendar days.27 Underlying lung cancer histopathologies were squamous carcinoma, adenocarcinoma, and other (large cell and bronchio-alveolar carcinomas). Mixed NSCLC-small cell lung cancers were excluded. Aggregate census data from 1996 and 2001 were used to compute the socio-economic factors. Variations in MO prac-
Adjuvant Chemotherapy Uptake in NSCLC
tices could not be accounted within multifactorial analyses due to the small number of entries (⬍5) in each factor level. Discriminant function analysis (DFA) was used to verify the robustness of the results obtained by LR, as it is considered preferable to stepwise LR for cofactor selection in data sets with small sample size.28 DFA is a multivariate statistical method that finds the linear combination (the discriminant function) of attributes (cofactors) that best separates two groups (e.g., not referred versus referred).28 An analysis of variance was used to test for the significance of the difference between the group centroids (means). A 2 test with Yates’ continuity correction was used to determine the predictive value of the discriminant function (i.e., the capacity of the resulting combination of cofactors to predict either Referral or Treatment patterns).
RESULTS Of 540 patients who were diagnosed with NSCLC in Nova Scotia, Canada in the year 2005, a total of 108 patients (20%) underwent curative-intent surgical resection (lobectomy 67%,; pneumonectomy 15%; wedge resection 18%) for stage I–III disease (39% IA; 23% IB; 24% II; 14% III). Their median age was 66 years, and 56% were males. Other characteristics are described in Table 1. Referral and Treatment patterns are shown in Figure 1. After curative-intent surgery, 44% (47 of 108) of all patients were referred to MO for consideration of adjuvant chemotherapy including 73% (30 of 41) of those with stage II–III disease. Of all those referred, 62% (29 of 47) were treated with adjuvant chemotherapy including 73% (22 of 30) of those with stage II–III disease. Overall, adjuvant chemotherapy was prescribed in 27% (29 of 108) of all patients after curative-intent surgery including 54% (22 of 41) of those with stage II–III disease. All patients received platinumbased doublet chemotherapy including paclitaxel and carboplatin (23 of 29), vinorelbine and cisplatin (5 of 29) or vinorelbine and carboplatin (1 of 29). Physician recommendation and patient choice were the most common identified causes for “no referral” and “no treatment” (Table 2). Of 61 patients not referred to MO, surgeons did not recommend referral in 64% while 5% refused referral. Of 18 patients who were referred but not treated, medical oncologists did not recommend treatment in 39% while 39% refused treatment. Overall, patients with stage IA disease represented 59% (33 of 56) of all patients for whom chemotherapy was not recommended by surgeons and/or medical oncologists. A number of cofactors were associated with uptake patterns by univariate analysis (Table 3). Younger age (⬍65 versus 65–75 versus ⬎75 years), more advanced stage (II–III versus IB versus IA), and more extensive surgery (pneumonectomy versus lobectomy versus wedge resection) were significant predictors of higher Referral, Treatment, and Overall Uptake. Practice related variables, including Health Centre and/or Surgeon, were also associated with Referral and Overall Uptake. Socio-economic variables, including Distance and MHI, were weakly associated with Referral; patients living closer to cancer center (p ⫽ 0.049) and in
Copyright © 2008 by the International Association for the Study of Lung Cancer
1273
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
Younis et al.
TABLE 1.
Characteristics of Study Population Cofactors: No. (%)
Age (yr) ⬍65 65–75 ⬎75 Sex Male Female Marital status Married Single Smoking Current Former Never Education ⱕMedian ⬎Median MHI ($) Low (ⱕ40K) High (⬎40K) Distance (km) Near (ⱕ32) Far (⬎32)
46 (43) 40 (37) 22 (20) 61 (56) 47 (44) 75 (86) 12 (14) 36 (35) 55 (53) 13 (12) 44 (41) 64 (59) 47 (44) 61 (56)
Histopathology Adenocarcinoma Squamous Other Stage IA IB II IIIa Margin Negative Positive Comorbidity Yes No PS 0–1 II–III Health centre QEII HSC CBRH
46 (43) 35 (32) 27 (25) 42 (39) 25 (23) 26 (24) 15 (14) 99 (97) 3 (3) 34 (33) 69 (67) 77 (71) 31 (29) 97 (90) 11 (10)
Surgeon A B C D Surgery type Wedge Lobectomy Pneumonectomy PO complications No Yes Med oncologist A B C D E F G
34 (32) 26 (24) 10 (9) 38 (35) 20 (18) 72 (67) 16 (15) 69 (64) 39 (36) 7 (16) 3 (7) 2 (5) 12 (28) 5 (12) 7 (16) 7 (16)
56 (52) 52 (48)
a Two patients underwent curative-intent surgery for stage IIIB disease, and were included in this cohort. MHI, median household income; K, thousands; km, kilometer; PS, performance status; QEII HSC, Queen Elizabeth II Health Sciences Centre; CBRH, Cape Breton Regional Hospital; PO Complications, postoperative complications; Med Onc, medical oncologist.
FIGURE 1. Uptake patterns. Number of patients diagnosed with Non-small Cell Lung Cancer (NSCLC) and those who underwent surgical resection over 1 year in Nova Scotia, Canada are shown (all stages or stage II–III). The circles show numbers (n) and percentages (%) of patients according to Referral (referred/resected), Treatment (treated/referred) and Overall Uptake (treated/ resected).
areas of higher MHI (p ⫽ 0.049) were more likely to be referred. There were no statistically significant associations between uptake patterns and other cofactors including Sex, Marital Status, Smoking, Histopathology, Margin, Comorbidity, PS, PO Complication, Med Onc or Education. As well, there were no significant associations observed between “Detection-Surgery” elapsed time (median 108 days) and uptake patterns. However, an inverse association was observed between “Surgery-MO Consult” elapsed time (median 49 days) and Treatment; referred patients who experienced shorter versus longer wait times to MO consult (⬍49 versus ⱖ49 days) were more likely treated (76% versus 46%, respectively: p ⫽ 0.039). Referral and Treatment patterns were examined further using both multifactorial (LR) and multivariate (DFA) anal-
1274
yses. Based on LR, independent predictors of Referral (Table 4) were Stage (P ⬍ 0.001), Age (p ⫽ 0.005), Surgery Type (p ⫽ 0.005), Surgeon (P ⫽ 0.03), and MHI (p ⫽ 0.02). Independent predictors of Treatment were Age (OR ⫽ 2.3; 95% CI, 1.1– 4.76; p ⫽ 0.016) as well as Stage and Surgery Type: none of the patients who had stage IA disease or underwent wedge surgery were treated with adjuvant chemotherapy. DFA showed very similar Referral and Treatment patterns (Figure 2). Higher likelihood of Referral was associated with Stage (II/III versus I), Age (younger versus older), Surgery Type (pneumonectomy versus wedge), Health Centre (CBRH versus NSCC), and MHI (higher versus lower). Higher likelihood of Treatment was associated with Stage (II/III versus I), Age (younger versus older), Surgery Type (pneumonectomy versus wedge) and “Surgery-MO Consult”
Copyright © 2008 by the International Association for the Study of Lung Cancer
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
TABLE 2. Identified Causes for “No Referral” and “No Treatment” by Stage Disease Stage
TABLE 4.
Underlying Causes
IA
IB
II–III
Total No. of Patients
No referrala
Not recommended Refused by patient Unknown/other Total Not recommended Refused by patient Unknown/other Total
30 0 8 38 3 0 1 4
4 2 6 12 0 4 2 6
5 1 5 11 4 3 1 8
39 3 19 61 7 7 4 18
a b
Causes of No referral in all patients. Causes of No treatment in referred-not treated cohort.
TABLE 3. Uptake Patterns Based on Univariate Analyses Referral Cofactorsb Age (yr) ⬍65 65–75 ⬎75 Stage IA IB II–III Surgery type Wedge Lobectomy Pneumonectomy Health centre QEII HSC CBRH Surgeon A B C D MHI ($) Low (ⱕ40K) High (⬎40K) Distance (km) Near (ⱕ32) Far (⬎32)
%
pc
%
0.019 59 35 27
0 24 75 0.731
60 55 0.087
0.037 24 55
0.066 80 30 75 57
0.049 32 53
⬍0.001
0.022
0.006
44 39 80 37
⬍0.001 0 28 54
0 59 80
39 82
0.002
0.009
⬍0.001
0.254
0.516 23 30
0.749 60 65
Cofactorsa Age Per decade Stage IA IB II–III Surgery type Lobectomy Wedge Pneumonectomy Surgeon A B C D MHI Low (ⱕ40 K) High (⬎40 K)
0.276 32 21
a An inverse association between Treatment and “Surgery-Medical Oncology Consult” elapsed time was also observed (p ⫽ 0.039). b Results shown include only those cofactors that predict at least one uptake pattern at p value ⬍0.05. c p-values associated with the likelihood ratio by 2 test (2). MHI, median household income; K, thousands; km, kilometer; PS, performance status; QEII HSC, Queen Elizabeth II Health Sciences Centre; CBRH, Cape Breton Regional Hospital.
elapsed time (shorter versus longer). These DFA models correctly predicted 80% and 85% of observed Referral and Treatment patterns, respectively. Misclassifications in Refer-
Yes
No
Odds Ratio
95% CI
47
61
2.6
4 13 30
38 12 11
1.0 14.9 27.4
— (2.9–76.9) (5.8–130.7)
29 3 15
43 17 1
1.0 0.5 40.4
— (0.1–2.8) (2.3–705.9)
15 10 8 14
19 16 2 24
1.0 0.14 3.3 0.7
— (0.03–0.8) (0.3–31.7) (0.2–3.1)
15 32
32 29
1.0 4.7
— (1.3–17.8)
P-valueb 0.005
(1.3–5.5) ⬍0.001
0.005
0.028
0.015
MHI, median household income; K, thousands; km, kilometer; No., Number. a Results shown include only those cofactors that predict Referral pattern at p-value ⬍0.05. b p-values associated with the likelihood ratio by 2 test (2).
ral (Figure 2A) were mainly due to the presence of patients with early stage disease among the referred population and patients with advanced stage disease among the nonreferred population. Misclassifications in Treatment (Figure 2B) were mainly due to the presence of patients with older age and stage IB disease among treated population and a patient living in an area with the highest MHI among non treated population.
0.013 35 12 60 21
73 56 0.049
pc
44 18 9
0 54 73
15 40 94
%
0.003
⬍0.001
54 33
pc
74 50 33
10 52 73
Overall Uptake
Treatmenta
Referral Pattern Based on Multifactorial Analysis Patients Referred (No.)
Uptake Pattern
No treatmentb
Adjuvant Chemotherapy Uptake in NSCLC
DISCUSSION Adjuvant chemotherapy has recently become a new standard of care for patients with NSCLC after clinical trials showed approximately 10% improvement in overall survival for those with higher risk disease (i.e., stage II and IIIA).3–5 In this population-based study, approximately one third of the patients who underwent curative-intent surgical resection for NSCLC were subsequently treated with adjuvant chemotherapy. These uptake rates may seem comparable to that in breast or colorectal cancers.6 –14 However, there are far fewer patients who undergo curative-intent surgical resection for NSCLC compared with other cancers as stage distribution in lung cancer is shifted towards more advanced stages.29 –32 In this study, only 20% of the patients with NSCLC underwent curative-intent surgical resection. Adjuvant chemotherapy therefore, would be expected to result in a modest less than 1% improvement in the current 5-year overall survival rate of 15% in lung cancer.30 –32 Our finding of a lower uptake of chemotherapy in older patients has also been observed in studies examining other cancers.33–35 This could reflect concerns about higher toxic-
Copyright © 2008 by the International Association for the Study of Lung Cancer
1275
Younis et al.
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
FIGURE 2. Discriminant Function Analysis (DFA) Results. A, shows discriminant function (LD1) that best separates patients referred from those not referred. Patients most likely to be referred had advanced stage (Stage: II/III versus I), were younger (Age: younger versus older), underwent pneumonectomy (Surgery Type: pneumonectomy versus wedge), treated at the CBRH (Health Centre: CBRH versus NSCC), or lived in area of higher MHI (MHI: higher versus lower). Factors loadings (relative magnitude of influence) were as follows: Stage (0.87), Age (⫺0.38), Surgery Type (0.29), Health Centre (⫺0.30), and MHI (0.29). B, shows discriminant function (LD1) that best separates patients referred and treated from those referred and not treated. Patients most likely to be treated had advanced stage (Stage: II/III versus I), were younger (Age: younger versus older), underwent pneumonectomy (Surgery Type: pneumonectomy versus wedge), and experienced shorter “Surgery-MO Consult” elapsed time (Timeline: shorter versus longer). Factors loadings (relative magnitude of influence) were as follows: Stage (0.53), Age (⫺0.40), Surgery Type (⫺0.37), and Surgery-MO Consult elapsed time (⫺0.47).
ity, paucity of data, and possible lack of clear benefit in older patients who are often excluded from clinical trials because of associated comorbidities and poorer PS. Treatment decisions concerning palliative chemotherapy in NSCLC are often influenced by PS rather than age alone.36 For adjuvant chemotherapy, a subset analysis of the NCIC BR 10 study36 revealed a significant overall survival benefit with adjuvant chemotherapy for patients over the age of 65 years (HR ⫽ 0.61; 95% CI, 0.38 – 0.98; p ⫽ 0.04), which was comparable to that observed in younger patients, and an acceptable toxicity. Patients over 75 years of age, however, had a worse overall survival compared with that of all patients enrolled in the trial (HR ⫽ 2.35; 95% CI, 0.84 – 6.58; p ⫽ 0.09) though sample size was too small to draw any firm conclusions in this elderly age subgroup. The higher uptake of adjuvant chemotherapy in stage II–III NSCLC reflects the current evidence which strongly supports treatment provision in stages II–IIIA disease.25 In our study, approximately three fourths of patients who underwent surgery for stage II–III NSCLC were referred to MO and three fourths of those referred were treated (i.e., approximately 50% Overall Uptake). For patients with stage IB disease, approximately half of patients who underwent sur-
1276
gery were referred and half of those referred were treated (i.e., approximately 25% Overall Uptake). Currently, the role of chemotherapy in stage IB disease is less clear,25,26 but a subset analysis of the CALGB 9633 study suggest that adjuvant chemotherapy could be considered for stage IB disease with larger size tumors.23,24 Conversely, there is no evidence to support platinum based chemotherapy in stage IA disease and the LACE meta-analysis suggest a potential detrimental effect in this patient subgroup (HR ⫽ 1.41; 95% CI, 0.96 –2.09).5 In our study, only 10% of patients with stage IA disease were referred to MO but none were treated with adjuvant chemotherapy. Decision making such as that concerning referral and/or treatment is complex and involves weighing a number of variables that are unique to every health care system-providerpatient encounter. As this study illustrates, uptake patterns were influenced by a number of variables that are related to patient (e.g., Age), disease (e.g., Stage or Surgery Type), practice (e.g., Surgeon or Health Centre), or socioeconomic status (e.g., MHI). The DFA models we present illustrate the combined impact of these variables in predicting referral and/or treatment but the interaction among these variables may be too complex to allow for identifying exact causes for
Copyright © 2008 by the International Association for the Study of Lung Cancer
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
no referral and/or no treatment. Moreover, studies that assess uptake of newer interventions should be considered to provide general predictors of outcomes and should not be simply regarded as performance measures. The uptake of adjuvant chemotherapy in NSCLC was reported in another retrospective study conducted at a single tertiary center in Ontario, Canada.38 Referral to MO and/or utilization of chemotherapy during the later period of that study (June 2004 –May 2005) seems comparable to that observed in our study (January 2005–December 2005); 63% of all patients were referred to MO and 54% of those referred were treated (i.e., 34% Overall Uptake). Possible reasons for not being referred and/or not being treated were retrospectively abstracted from patients’ charts, but statistical analysis to identify predictors of referral and/or treatment was not conducted. Reasons for not being referred to MO included; stage IA disease (37%), patient choice (18%), surgeon recommendation (15%), significant comorbidities (10%), PO complications (5%), disease progression (3%), and advanced age (2%). Reasons for not prescribing adjuvant chemotherapy included patient refusal (50%), comorbidities (14%), stage IA disease (10%), disease progression (6%), advanced age (6%), and unknown (14%). Our study has limitations. Firstly, the number of cases meeting the inclusion criteria was relatively small reducing the overall statistical power of the study and limiting the stability of weak associations. However, we employed both multifactorial and discriminant function analyses and the results were almost identical. Secondly and perhaps more importantly, the retrospective nature of our study did not allow us to determine the exact causes of no referral and/or no treatment. However, the predictors of referral and/or treatment identified may help improve future chemotherapy uptake rates. Thirdly, our study examined the uptake patterns in the first year (i.e., 2005) after adjuvant chemotherapy for NSCLC was accepted as a new standard of care. It is possible that uptake patterns may have changed since then, as further updated data from clinical trials were reported, and/or mature studies were fully published. In summary, approximately one third of patients who underwent curative-intent surgery for NSCLC were treated with adjuvant chemotherapy. Consistent with other reports in breast and colorectal cancers, higher uptake of adjuvant chemotherapy was observed in younger patients and those with more advanced disease stages. Improved referral to MO may increase the overall uptake of adjuvant chemotherapy as a high proportion of those referred were treated. Prospective assessment, however, is still required to better determine the exact causes of no referral and/or no treatment in clinical decision making that is unique to every health care systemprovider-patient encounter. In the interim, the uptake patterns observed in this study may help future planning of health resource utilization and/or improvement of chemotherapy uptake rates.
ACKNOWLEDGMENTS The authors would like to thank Ron Dewar and Maureen MacIntyre for their helpful suggestions throughout this study.
Adjuvant Chemotherapy Uptake in NSCLC
REFERENCES 1. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG): effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomized trials. Lancet 2005;365:1687–1717. 2. Wolpin BM, Meyerhardt JA, Mamon HJ, et al. Adjuvant treatment of colorectal cancer. CA Cancer J Clin 2007;57:168 –185. 3. Hotta K, Matsuo K, Ueoka H, et al. Role of adjuvant chemotherapy in patients with resected non-small-cell lung cancer: reappraisal with a meta-analysis of randomized controlled trials. J Clin Oncol 2004;22: 3860 –3867. 4. Sedrakyan A, Van Der Meulen J, O’Byrne K, et al. Postoperative chemotherapy for non-small cell lung cancer: a systematic review and meta-analysis. J Thorac Cardiovasc Surg 2004;128:414 – 419. 5. Pignon JP, Triboder GV, Scagliotti G, et al. Lung Adjuvant Cisplatin Evaluation (LACE): a pooled analysis of five randomized clinical trials including 4,584 patients. J Clin Oncol 2006;24(18 Suppl):366S (abstr. #7008). 6. Harlan LC, Abrams J, Warren JL, et al. Adjuvant therapy for breast cancer: practice patterns of community physicians. J Clin Oncol 2002; 20:1809 –1817. 7. Du X, Goodwin JS. Patterns of use of chemotherapy for breast cancer in older women: findings from Medicare claims data. J Clin Oncol 2001; 19:1455–1461. 8. Du X, Key C, Osborne C, et al. Discrepancy between consensus recommendations and actual community use of adjuvant chemotherapy in women with breast cancer. Ann Intern Med 2003;138:90 –97. 9. Edwards BK, Brown ML, Wingo PA, et al. Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst 2005;97:1407–1427. 10. Schrag D, Gelfand SE, Bach PB, et al. Who gets adjuvant treatment for stage II and III rectal cancer? Insight from surveillance, epidemiology, and end results–medicare. J Clin Oncol 2001;19:3712–3718. 11. Ong S, Watters JM, Grunfeld E, et al. Predictors of uptake of adjuvant therapy in CRC. Can J Surg 2005;48:225–229. 12. Jessup JM, Stewart A, Greene FL, et al. Adjuvant chemotherapy for stage iii colon cancer: implications of race/ethnicity, age, and differentiation. JAMA 2005;294:2703–2711. 13. Potosky AL, Harlan LC, Kaplan RS, et al. Age, sex, and racial differences in the use of standard adjuvant therapy for colorectal cancer. J Clin Oncol 2002;20:1192–1202. 14. Ayanian JZ, Zaslavsky AM, Fuchs CS, et al. Use of adjuvant therapy for CRC in a population-based cohort. J Clin Oncol 2003;21:1293–1300. 15. American Society of Clinical Oncology. Clinical Practice Guidelines (http://www.asco.org). 16. National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology (http://www.ncccn.org). 17. Keller SM, Adak S, Wanger H, et al. A randomized trial of postoperative adjuvant therapy in patients with completely resected Stage II or IIIA non-small cell lung cancer (Eastern Cooperative Oncology Group). N Engl J Med 2000;343:1217–1222. 18. Scagliotti GV, Fossati R, Torri V, et al. Randomized study of adjuvant chemotherapy for completely resected stage I, II, or IIIA non-small-cell Lung cancer. J Natl Cancer Inst 2003;95:1453–1461. 19. Waller D, Peake MD, Stephens RJ, et al. Chemotherapy for patients with non-small cell lung cancer: the surgical setting of the Big Lung Trial. Eur J Cardiothorac Surg 2004;26:173–182. 20. Arriagada R, Bergman B, Dunant A, et al. (for The International Adjuvant Lung Cancer Trial Collaborative Group). Cisplatin based adjuvant chemotherapy in patients with completely resected non-smallcell lung cancer. N Engl J Med 2004;350:351–360. 21. Winton TL, Livingstan R, Johnson D, et al. A prospective randomized trial of adjuvant vinorelbine (VNR) and cisplatin (CIS) in completely resected stage Ib and II non small cell lung cancer (NSCLC) Intergroup JBR 10. N Engl J Med 2005;352:2589 –2597. 22. Douillard JY, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation patients with completely resected stage IB–IIIA non-small-cell lung cancer (Adjuvant Navelbine International Trialist Association [ANITA]): a randomised controlled trial. Lancet Oncol 2006;7:719 –727. 23. Strauss GM, Hernden J, Maddaus MA, et al. Randomized clinical trial of
Copyright © 2008 by the International Association for the Study of Lung Cancer
1277
Younis et al.
24.
25.
26. 27. 28.
29. 30.
Journal of Thoracic Oncology • Volume 3, Number 11, November 2008
adjuvant chemotherapy with paclitaxel and carboplatin following resection in Stage Ib Non-Small Cell Lung Cancer (NSCLC): report of cancer and leukemia group B (CALGB) protocol 9633. J Clin Oncol 2004;22: 621S (abstr. #7019). Strauss GM, Hernden J, Maddaus MA, et al. Adjuvant chemotherapy in Stage IB non-small cell lung cancer (NSCLC): update on Cancer and Leukemia Group B (CALGB) protocol 9633. J Clin Oncol 2006;24(18 Suppl):365S (abstr. #7007). Pisters KM, Evans WK, Azzoli CG, et al. Cancer Care Ontario and American Society of Clinical Oncology adjuvant chemotherapy and adjuvant radiation therapy for stages I–IIIA resectable non small-cell lung cancer guideline. J Clin Oncol 2007;25:5506 –5518. Wakelee H, Dubey S, Gandara D. Optimal adjuvant therapy for nonsmall cell lung cancer– how to handle stage I disease. Oncologist 2007;12:331–337. Saint-Jacques N, Rayson D, Al-Fayea T, et al. Waiting times in early-stage non-small cell lung cancer (NSCLC). J Thorac Oncol 2008;3:865– 870. O’Gorman TW, Woolson RF. Variable Selection to Discriminate Between Two Groups: Stepwise Logistic Regression or Stepwise Discriminant Analysis? In The American Statistician: American Statistical Association, 1991. Pp. 187–193. Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710 –1717. SEER: Surveillance, Epidemiology, and End Results Program. National Cancer Institute. Cancer Statistics 2005 (www.seer.cancer.gov). Accessed October 10, 2008.
1278
31. American Cancer Society: Cancer Facts and Figures 2006. Atlanta, GA, American Cancer Society, 2006 (http://americancancersociety.org). Accessed October 10, 2008. 32. Canadian Cancer Society, National Cancer Institute of Canada. Canadian Cancer Statistics, 2007. Toronto, Canada 2007 (www.cancer.ca). Accessed October 10, 2008. 33. Fisher B. Highlights from recent National Surgical Adjuvant Breast and Bowel Project studies in the treatment and prevention of breast cancer. CA Cancer J Clin 1999;49:159 –177. 34. Du XL, Jones DV, Zhang D. Effectiveness of adjuvant chemotherapy fornode-positive operable breast cancer in older women. J Gerontol A Biol Sci Med Sci 2005;60:1137–1144. 35. Giordano SH, Duan Z, Kuo Y-F, et al. Use and outcomes of adjuvant chemotherapy in older women with breast cancer. J Clin Oncol 2006; 24:2750 –2756. 36. Gridelli C. The ELVIS trial: a phase III study of single-agent vinorelbine as first-line treatment in elderly patients with advanced non-small cell lung cancer. Elderly Lung Cancer Vinorelbine Italian Study. Oncologist 2001;1:(6 Suppl) 4 –7. 37. Pepe C, Hasan B, Winton T, et al. Adjuvant vinorelbine and cisplatin in elderly patients: National Cancer Institute of Canada and Intergroup Study JBR. 10. J Clin Oncol 2007;25:1553–1561. 38. Kassam F, Shepherd FA, Johnston M, et al. Referral patterns for adjuvant chemotherapy in patients with completely resected non-small cell lung cancer. J Thorac Oncol 2007;2:39 – 43.
Copyright © 2008 by the International Association for the Study of Lung Cancer