Abstract
Background
Current guidelines recommend routine imaging surveillance for non-small cell lung cancer (NSCLC) patients following treatment. Little is known about surveillance patterns for surgically resected, early-stage lung cancer patients in the community-at-large. We sought to characterize surveillance patterns in a national cohort.
Methods
We conducted a retrospective study using Surveillance, Epidemiology, and End-Results (SEER)-Medicare database (1995-2010). Patients with stage I/II NSCLC treated with surgical resection were included. Our primary outcome was receipt of imaging between 4 and 8 months following surgery. Covariates included demographics and comorbidities.
Results
Chest radiography (CXR) was the most frequent initial modality (60%) followed by chest computerized tomography (CT) (25%). Positron emission tomography (PET) was least frequent as initial imaging modality (3%). A total of 13% of patients received no imaging within the initial surveillance period. Adherence to National Comprehensive Cancer Network (NCCN) guidelines for imaging by overall prevalence was 47% for receipt of CT, however rates of CT imaging increased over time from 28% to 61% (p<0.01). Reduced rates of CT imaging were associated with stage I disease and surgery as the sole treatment modality.
Conclusions
Imaging following definitive surgery for NSCLC predominantly utilizes CXR rather than CT. Most of this imaging is likely for surveillance and in that context, CXR has inferior detection rates for recurrence and detection of new cancers. Adherence to guideline recommended CT surveillance following surgery is poor, but the reason multifactorial. Efforts to improve adherence to imaging surveillance must be coupled with greater evidence demonstrating improved long-term outcomes.
Introduction
There were an estimated 412,230 people living with lung cancer in the US in 2012.1 This number of cancer survivors is expected to increase with adoption of the US Preventative Services Task Force recommendation for screening high-risk patients for early detection of lung cancer.2 Seventy percent of patients with lung cancer detected via computerized tomography (CT) screening programs are identified at an early stage.3 Surgical resection remains the cornerstone for definitive treatment of early stage non-small cell lung cancer (NSCLC) given superior outcomes compared to non-surgical therapies. Yet, despite surgical advances, many patients will experience recurrence and remain at risk for second primary lung cancers in need of further treatment. Recognizing this risk, several national guidelines have been developed to standardize protocols for appropriate imaging surveillance following treatment.
Current guidelines are based almost exclusively on consensus opinion or selective single-center trials. Most guidelines recommend routine imaging using either CT or chest radiography (CXR) with more frequent surveillance in the first two years. The lack of evidence of efficacy is compounded by the lack of knowledge of real-world application of these guidelines. We sought to characterize the imaging surveillance patterns of resected NSCLC patients in a national cohort to determine adherence to national standards.
Material and Methods
Study Population
The Institutional Review Board of the University of Washington approved this study. Data was obtained from the national SEER registry database with Medicare linkage for patients with NSCLC diagnosed during 1996-2009. A total of 438,872 patients were identified. We excluded patients younger than age 65 years (n=58,744) or diagnosed at autopsy or by death certificate (n=12,119). We restricted the cohort to stage I and II patients (exclude n=280,476) who underwent lung resection within 6 months of diagnosis (exclude n=23,565). Patients with prior malignancy or other malignancy diagnosed within 12 months after lung cancer diagnosis were also excluded (n=5,235). Patients had to have continuous Medicare eligibility for 12 months or more after date of diagnosis (exclude n=40,327). The final study population consisted of 18,406 patients. (Appendix A)
Primary Outcome
Our primary outcome was receipt of CT during the initial surveillance period. Secondary outcomes were receipt of CXR, positron emission tomography (PET), or any of the three modalities within the defined surveillance period. National Comprehensive Cancer Network (NCCN) guidelines for lung cancer management recommend imaging surveillance with chest CT between 4-6 months following definitive therapy. Accordingly, the initial surveillance period was defined as 4-6 months following date of surgery; however the interval was increased to 4-8 months to allow for a more liberal estimate of adherence. The latter served as the basis for all analyses. Given the fact that some patients may have received more than one imaging study during the surveillance period, we examined imaging both by rates of initial imaging modality exclusive to other modalities as well as prevalence of each imaging type inclusive of all modalities.
Analysis
Unadjusted analyses for rates of imaging were compared using chi-square test and Mantel-Haenszel chi-square test for trends over time. Kaplan-Meier estimates were used to examine the cumulative proportion of patients receiving imaging for each modality. Two multivariate logistic regression models were fitted to the data. One model assessed factors associated with receipt of CT as first imaging and the other model assessed factors associated with no imaging during the initial surveillance period. All analyses were performed in SAS, v9.3 (SAS Institute Inc, Cary, NC).
Results
The study population was 46% male and 89% white race. One half of patients had no comorbid conditions. Most patients were pathologic stage I (86%) and underwent surgery as their sole treatment (84%). The most frequent initial imaging modality performed during initial surveillance was CXR (60%), followed by CT (25%), and PET (3%). (Table 1) The prevalence of each imaging modality performed inclusive of one another is shown in Table 2. The proportion of patients receiving no imaging surveillance during the first 4-8 months following resection was 13%. On univariate analysis, increased rates of CT imaging were associated with younger age, (p<0.01), married status (p<0.01), low poverty level (p<0.01), stage II disease (p<0.01), and multimodality treatment (p<0.01). An inverse relationship was seen for receipt of no imaging which was associated with unmarried status (p<0.01), non-white race (p<0.01), fewer comorbid conditions (p<0.01) stage I disease (p<0.01), and surgery alone as primary therapy (p<0.01).
Table 1. First Imaging Surveillance 4-8 months following surgical resection.
| Number of subjects | First Surveillance Imaging in Patients with Early Stage Lung Cancer 4-8 Months Following Surgical Resection | |||||
|---|---|---|---|---|---|---|
| Number of subjects | N (%) | CXR | CT | PET/CT | None of Three Imaging | P value |
| Total N (%) | 18,406 | 10,964 (59.6) | 4,531 (24.6) | 482 (2.6) | 2,429 (13.2) | |
| Age (years) | 0.0011 | |||||
| 65∼69 | 4,902 (26.6) | 57.4 | 26.7 | 2.6 | 13.2 | |
| 70∼74 | 5,568 (30.3) | 59.6 | 24.5 | 2.9 | 13.0 | |
| 75∼79 | 4,735 (25.7) | 61.4 | 23.4 | 2.5 | 12.7 | |
| 80+ | 3,201 (17.4) | 60.0 | 23.3 | 2.3 | 14.3 | |
| Gender | 0.1432 | |||||
| Male | 8,435 (45.8) | 60.3 | 24.3 | 2.7 | 12.7 | |
| Female | 9,971 (54.2) | 58.9 | 24.9 | 2.6 | 13.6 | |
| Marital Status | < .0001 | |||||
| Married | 10,602 (57.6) | 59.8 | 25.5 | 2.6 | 12.1 | |
| Others | 7,804 (42.4) | 59.3 | 23.4 | 2.6 | 14.7 | |
| Poverty Level | < .0001 | |||||
| High | 3,570 (19.5) | 62.8 | 20.3 | 2.9 | 14.1 | |
| Middle | 9,004 (49.2) | 60.3 | 23.7 | 2.6 | 13.5 | |
| Low | 5,745 (31.4) | 56.4 | 28.9 | 2.6 | 12.1 | |
| Education Level | < .0001 | |||||
| Low | 3,555 (19.4) | 61.5 | 21.6 | 2.8 | 14.1 | |
| Middle | 9,193 (50.2) | 60.6 | 23.8 | 2.5 | 13.1 | |
| High | 5,571 (30.4) | 56.5 | 28.1 | 2.7 | 12.7 | |
| Race | 0.0062 | |||||
| White | 1,6402 (89.1) | 59.9 | 24.7 | 2.6 | 12.8 | |
| Black | 962 (5.2) | 56.3 | 24.5 | 2.3 | 16.9 | |
| Other/unknown | 1,038 (5.7) | 57.8 | 23.9 | 2.7 | 15.6 | |
| Charlson Comorbidity | 0.0111 | |||||
| No conditions | 9,199 (50.3) | 59.4 | 24.4 | 2.5 | 13.7 | |
| 1 Condition | 5,555 (30.3) | 60.1 | 25.0 | 2.5 | 12.4 | |
| 2-3 Conditions | 2,983 (16.3) | 59.6 | 24.8 | 3.1 | 12.5 | |
| ≥ 4 Conditions | 568 (3.1) | 63.9 | 23.6 | 3.0 | 8.8 | |
| Stage | < .0001 | |||||
| I | 15,863 (86.2) | 60.1 | 23.9 | 2.3 | 13.7 | |
| II | 2,543 (13.8) | 56.6 | 28.9 | 4.4 | 10.2 | |
| Initial Treatment | < .0001 | |||||
| Surgery alone | 15,510 (84.3) | 60.7 | 22.8 | 2.1 | 14.4 | |
| Surgery and XRT | 705(3.8) | 66.2 | 22.3 | 3.0 | 8.5 | |
| Surgery and Chemo | 1,907 (10.4) | 47.9 | 39.3 | 6.2 | 6.5 | |
| Surgery and XRT and Chemo | 284 (1.5) | 58.1 | 32.0 | 6.7 | 3.2 | |
| Surgical Resection | < .0001 | |||||
| Wedge | 4,851 (26.4) | 55.4 | 28.0 | 2.7 | 13.9 | |
| Segmentectomy | 839 (4.6) | 55.5 | 27.5 | 3.9 | 13.0 | |
| Lobectomy | 11,569 (62.9) | 61.0 | 23.3 | 2.6 | 13.1 | |
| Bilobectomy | 590 (3.2) | 68.5 | 18.6 | 1.9 | 11.0 | |
| Pneumonectomy | 494 (2.7) | 62.5 | 24.7 | 1.4 | 11.4 | |
| Mediastinoscopy Performed | 0.0002 | |||||
| Yes | 3,277 (17.8) | 60.0 | 24.0 | 2.6 | 13.4 | |
| No | 15,129 (82.2) | 57.7 | 27.5 | 2.6 | 12.1 | |
| Histology | < .0001 | |||||
| Adenocarcinoma | 10,346 (56.2) | 57.8 | 26.2 | 2.7 | 13.3 | |
| Squamous | 5,527 (30.0) | 62.8 | 21.9 | 2.3 | 13.0 | |
| Large Cell | 690 (3.8) | 64.9 | 21.3 | 3.3 | 10.4 | |
| NSCLC Undiff | 707 (3.8) | 56.4 | 28.0 | 2.7 | 12.9 | |
| Others | 1,136 (6.2) | 58.7 | 23.7 | 2.6 | 15.1 | |
Poverty: quartile of national poverty level by census tract; High: >/=75%, Middle: 25%-75%, Low: </= 25%
Education: quartile of national non-high school graduation rate by census tract; Low: >/=75, Middle: 25-75%, High: </=25%
XRT: Radiation; Chemo: chemotherapy
Table 2. Overall Imaging Surveillance 4-8 months following surgical resection.
| Number of subjects | N (%) | CXR | P-value | CT | P-value | PET/CT | P-value | None | P-value |
|---|---|---|---|---|---|---|---|---|---|
| Size N | Total 18,406 | 12,608 (68.5) | 8,758 (47.6) | 1,952 (10.6) | 2,429 (13.2) | ||||
| Age | 0.0078 | < .0001 | < .0001 | 0.1684 | |||||
| 65∼69 | 4,902 (26.6) | 66.8 | 49.7 | 11.9 | 13.2 | ||||
| 70∼74 | 5,568 (30.3) | 68.8 | 48.1 | 11.6 | 13.0 | ||||
| 75∼79 | 4,735 (25.7) | 70.0 | 46.8 | 9.3 | 12.7 | ||||
| 80+ | 3,201 (17.4) | 68.5 | 44.5 | 8.8 | 14.3 | ||||
| Gender | 0.0008 | 0.2805 | 0.1571 | 0.0593 | |||||
| Male | 8,435 (45.8) | 69.8 | 48.0 | 11.0 | 12.7 | ||||
| Female | 9,971 (54.2) | 67.5 | 47.2 | 10.3 | 13.6 | ||||
| Marital Status | 0.0476 | 0.0013 | 0.2729 | < .0001 | |||||
| Married | 10,602 (57.6) | 69.1 | 48.6 | 10.8 | 12.1 | ||||
| Others | 7,804 (42.4) | 67.7 | 46.2 | 10.3 | 14.7 | ||||
| Poverty Level | < .0001 | < .0001 | 0.0947 | 0.0093 | |||||
| High | 3,570 (19.5) | 71.8 | 44.9 | 11.6 | 14.1 | ||||
| Middle | 9,004 (49.2) | 69.2 | 45.7 | 10.5 | 13.5 | ||||
| Low | 5,745 (31.4) | 65.3 | 52.3 | 10.2 | 12.1 | ||||
| Education Level | < .0001 | 0.0007 | 0.0076 | 0.1709 | |||||
| High | 5,571 (30.4) | 65.3 | 49.7 | 10.4 | 12.7 | ||||
| Middle | 9,193 (50.2) | 69.8 | 46.8 | 10.3 | 13.1 | ||||
| Low | 3,555 (19.4) | 70.2 | 46.4 | 12.1 | 14.1 | ||||
| Race | 0.0567 | 0.0816 | 0.3269 | 0.0002 | |||||
| White | 1,6402 (89.1) | 68.8 | 47.9 | 10.6 | 12.8 | ||||
| Black | 962 (5.2) | 66.1 | 45.4 | 9.4 | 16.9 | ||||
| Other/unknown | 1,038 (5.7) | 65.7 | 44.4 | 11.9 | 15.6 | ||||
| Charlson Comorbidity | < .0001 | < .0001 | < .0001 | 0.0017 | |||||
| No conditions | 9,199 (50.3) | 67.3 | 44.9 | 9.7 | 13.7 | ||||
| 1 Condition | 5,555 (30.3) | 69.5 | 49.1 | 11.0 | 12.4 | ||||
| 2-3 Conditions | 2,983 (16.3) | 70.4 | 51.9 | 12.4 | 12.5 | ||||
| ≥ 4 Conditions | 568 (3.1) | 74.8 | 56.9 | 13.6 | 8.8 | ||||
| Stage | 0.1404 | < .0001 | < .0001 | < .0001 | |||||
| I | 15,863 (86.2) | 68.3 | 45.7 | 9.4 | 13.7 | ||||
| II | 2,543 (13.8) | 69.8 | 59.6 | 18.0 | 10.2 | ||||
| Initial Treatment | < .0001 | < .0001 | < .0001 | < .0001 | |||||
| Surgery alone | 15,510 (84.3) | 68.6 | 43.5 | 8.4 | 14.4 | ||||
| Surgery and XRT | 705(3.8) | 77.3 | 58.0 | 13.8 | 8.5 | ||||
| Surgery and Chemo | 1,907 (10.4) | 62.6 | 72.5 | 24.1 | 6.5 | ||||
| Surgery and XRT and Chemo | 284 (1.5) | 78.8 | 77.1 | 30.6 | 3.2 | ||||
| Surgical Resection | < .0001 | < .0001 | < .0001 | 0.3308 | |||||
| Wedge | 4,851 (26.4) | 64.9 | 50.8 | 11.9 | 13.9 | ||||
| Segmentectomy | 839 (4.6) | 66.0 | 51.7 | 14.2 | 13.0 | ||||
| Lobectomy | 11,569 (62.9) | 69.5 | 46.2 | 10.0 | 13.1 | ||||
| Bilobectomy | 590 (3.2) | 78.0 | 41.2 | 8.6 | 11.0 | ||||
| Pneumonectomy | 494 (2.7) | 71.8 | 49.3 | 9.2 | 11.4 | ||||
| Others | 63 (0.3) | 73.0 | 46.0 | 6.4 | 11.1 | ||||
| Mediastinoscopy | 0.9100 | < .0001 | 0.0003 | 0.0435 | |||||
| Yes | 3,277 (17.8) | 68.4 | 46.4 | 10.2 | 13.4 | ||||
| No | 15,129 (82.2) | 68.5 | 53.1 | 12.4 | 12.1 | ||||
| Histology | < .0001 | 0.0189 | 0.3296 | 0.0744 | |||||
| Adenocarcinoma | 10,346 (56.2) | 66.6 | 47.9 | 10.6 | 13.3 | ||||
| Squamous | 5,527 (30.0) | 71.9 | 46.8 | 10.5 | 13.0 | ||||
| Large Cell | 690 (3.8) | 74.1 | 47.1 | 10.7 | 10.4 | ||||
| NSCLC Undiff | 707 (3.8) | 66.6 | 52.8 | 13.0 | 12.9 | ||||
| Others | 1,136 (6.2) | 67.1 | 45.3 | 10.3 | 15.1 |
Examination of imaging trends over time demonstrated an increase from 10% to 36% (p<0.01) for CT comparing the earliest (1996-2000) to latest study periods (2006-2009) while CXR decreased from 75% to 47% (p<0.01). (Table 3) Receipt of no imaging declined from 15% to 13% (p<0.01). Time to event analysis estimated 75% of all patients would receive a CT by 538 days following surgery compared to 217 days for CXR. (Figure 1)
Table 3. Imaging by Study Period.
| 1996-2000 | 2001-2005 | 2006-2009 | *p-value | |
|---|---|---|---|---|
| Total (n=18,406) | 4144 (22%) | 7690 (42%) | 6572 (36%) | |
| First Surveillance Imaging | ||||
| CXR % | 75 | 62 | 47 | <0.0001 |
| CT % | 10 | 23 | 36 | |
| PET/CT % | 0 | 2 | 5 | |
| None % | 15 | 13 | 12 | |
| Prevalence of Surveillance Imaging | ||||
| CXR % | 79 | 71 | 59 | <0.0001 |
| CT % | 28 | 47 | 61 | <0.0001 |
| PET/CT % | 2 | 9 | 18 | <0.0001 |
Figure 1. Kaplan-Meier estimates for Imaging over time.
On multivariate analysis, receipt of CT as the first imaging modality was associated with older age (75-79 years, OR 1.08 [95% CI 1.00-1.18], Stage II disease (OR 1.32 [95% CI 1.15-1.41]) and multimodality therapy compared to surgery alone. (Table 4) Lower odds of CT during the surveillance period were associated with high poverty (middle, OR 0.75 [95% CI 0.70-0.80], high, OR 0.72 [95% CI 0.66-0.79]) compared to low, anatomic resection compared to wedge resection (lobectomy OR 0.81 {95% CI 0.76-0.87], bilobectomy OR 0.64 [95% CI 0.53-0.77], pneumonectomy OR 0.71 [95% CI 0.58-0.87] and more comorbidities. Lack of any imaging during the surveillance period was associated with surgery as the sole treatment rendered.
Table 4. Predictors of Imaging.
| Initial CT Chest @ 4-8 months | No Imaging @ 4-8 months | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |||
| Age | ||||||
| 65∼69 | REF | |||||
| 70∼74 | 1.06 | 0.98 | 1.15 | 0.98 | 0.87 | 1.10 |
| 75∼79 | 1.08 | 1.00 | 1.18 | 1.04 | 0.92 | 1.18 |
| 80+ | 1.13 | 1.03 | 1.24 | 0.95 | 0.83 | 1.09 |
| Gender | ||||||
| Female | REF | |||||
| Male | 0.97 | 0.91 | 1.03 | 1.03 | 0.94 | 1.13 |
| Marital Status | ||||||
| Not Married | REF | |||||
| Married | 1.06 | 1.00 | 1.13 | 0.82 | 0.75 | 0.90 |
| Poverty Level | ||||||
| Low | REF | |||||
| Middle | 0.76 | 0.70 | 0.81 | 1.13 | 1.01 | 1.27 |
| High | 0.71 | 0.63 | 0.79 | 1.11 | 0.93 | 1.31 |
| Education Level | ||||||
| Low | REF | |||||
| Middle | 1.08 | 0.98 | 1.20 | 1.04 | 0.90 | 1.19 |
| High | 1.06 | 0.94 | 1.19 | 1.04 | 0.88 | 1.23 |
| Race | ||||||
| White vs Black | 1.08 | 0.94 | 1.24 | 0.77 | 0.64 | 0.93 |
| White vs Other/unknown | 1.12 | 0.98 | 1.28 | 0.79 | 0.66 | 0.95 |
| Black vs Other/unknown | 1.04 | 0.87 | 1.25 | 1.03 | 0.81 | 1.32 |
| Charlson Comorbidity Index | ||||||
| No conditions | REF | |||||
| 1 condition | 0.83 | 0.77 | 0.89 | 1.13 | 1.02 | 1.25 |
| 2-3 Conditions | 0.72 | 0.66 | 0.78 | 1.16 | 1.02 | 1.31 |
| ≥ 4 Conditions | 0.55 | 0.46 | 0.65 | 1.82 | 1.35 | 2.47 |
| Stage | ||||||
| I | REF | |||||
| II | 1.32 | 1.15 | 1.41 | 0.96 | 0.83 | 1.12 |
| Treatment | ||||||
| Surgery Alone | REF | |||||
| Surg+Rads | 1.69 | 1.43 | 1.96 | 0.56 | 0.43 | 0.74 |
| Surg+Chemo | 3.23 | 2.86 | 3.57 | 0.43 | 0.36 | 0.53 |
| Surg+Rads+Chemo | 4.35 | 3.23 | 5.88 | 0.20 | 0.10 | 0.30 |
| Surgical Resection | ||||||
| Wedge | REF | |||||
| Segmentectomy | 0.99 | 0.85 | 1.15 | 0.93 | 0.74 | 1.16 |
| Lobectomy | 0.81 | 0.76 | 0.87 | 0.93 | 0.85 | 1.03 |
| Bilobectomy | 0.64 | 0.53 | 0.77 | 0.78 | 0.59 | 1.02 |
| Pneumonectomy | 0.71 | 0.58 | 0.87 | 0.94 | 0.69 | 1.28 |
| Mediastinoscopy | ||||||
| No | REF | |||||
| Yes | 0.80 | 0.73 | 0.86 | 1.06 | 0.94 | 1.19 |
Comment
Guidelines for follow up of early stage lung cancer patients are widely reported, but the evidence underlying these guidelines is not strong. Agencies advocating for surveillance include NCCN4, American College of Radiology (ACR)5, American College of Chest Physicians6, American Society of Clinical Oncology7 and the European Society of Medical Oncology8. All recommend more frequent imaging in the first 2 years consistent with peak risk of recurrence.9-11 However, they differ in recommended modality and intervals for follow-up. Both NCCN and ACR recommend CT as the preferred modality.
Imaging surveillance for lung cancer patients is supported by reported rates of recurrence between 4-41%.12-15 Recurrence detection is even higher within standardized imaging surveillance programs.14,16-18 Many patients with surveillance detected recurrent cancer will be eligible for secondary therapy with curative intent.10,12,16,17,19-22 Data regarding the impact of surveillance on improved survival have been mixed however. At least one study has demonstrated improved 3-year survival following treatment of surveillance-detected recurrence of 31% compared to 13% for those detected by symptoms.23 Many have reported 5-year survival rates comparable to those undergoing surgical resection of initial lung cancers.10,12,21,24-27 Still others report no difference in overall survival for surveillance programs.17,28
There is also limited data examining real-world practice of imaging surveillance in a large population. The largest claims based study of guideline recommended follow-up care studied 100 lung cancer patients and found that most underwent surveillance with CXR (74%) and only 37% had surveillance CT which is consistent our finding of less than 50% of patients receiving guideline recommended imaging.29 We found low rates of surveillance associated with high poverty which raises the issue of access to care as a major determinant of health status and ongoing care.30,31 We also found that CT surveillance increased over time, potentially due to increase in imaging overall for reasons independent of cancer. Changes in the guidelines themselves over time may also have influenced this finding. In 1996, version 1 of the NCCN guidelines recommended surveillance with history and physical examination and CXR every 4 months for the first 2 years.32 Our earliest cohort (1996-2000) showed a 75% adherence rate for CXR. Following that, CXR use declined and CT use increased. Nonetheless, the rate of CT imaging only reached 36% by the latest study time period. Lack of imaging surveillance demonstrated only modest decline over time and still represented 12% of the population.
Two other findings that stand out were patients with stage I disease and those undergoing surgery as sole therapy are at increased risk for lack of imaging surveillance. This subset of patients stands to gain the most benefit from surveillance as they have the best overall survival and potential time-at-risk for recurrent disease. However, patients undergoing sub-lobar lung resections are also at higher risk for local recurrence, and had higher odds of receiving a CT than those undergoing more extensive surgical resection. Thus the patterns observed demonstrate appropriate patient selection for some and inappropriate selection for others. Such heterogeneity in adherence is also found in post-treatment surveillance programs for other cancer types.33-38
We specifically targeted early stage lung cancer patients undergoing surgical resection for their favorable outcomes. Our findings of stage I patients or those receiving surgery alone associated with the lowest odds of surveillance might be explained by heterogeneity in their follow up. These patients are potentially less likely to be referred to medical oncology for adjuvant therapy and may bypass the usual home for cancer surveillance. The task of surveillance may default to the thoracic surgeon, pulmonologist, or primary care provider. Exactly which of these providers will take the lead for surveillance may be ill defined. Thus, while Stage I patients undergoing surgical resection may have the best long-term survival, they may also paradoxically have the worst chance for meaningful surveillance. Further barriers to surveillance may include lack of reimbursement incentives, poor communication between providers and specialists, heterogeneity in knowledge, and inconsistent dissemination of current practice guidelines. Surveillance strategies reported by thoracic surgeons vary widely with only 44% believing in survival benefit. This heterogeneity may not be specialty specific.39,40
Our study has several limitations. We utilized administrative claims with inherent challenges, however claims data have demonstrated ability to characterize guideline adherence comparable to medical chart review for other types of cancer surveillance.41 We do not have data on indication for the studies being performed thus there is potential for misclassification bias. Additionally, patient selection may impact a provider's decision to request surveillance imaging. Optimal outcomes from surveillance programs are seen in patients with better performance status, female gender, and younger age.16 Careful patient selection may actually be the key to a successful surveillance program. A selective approach rather than universal surveillance might lead to improved outcomes and efficiency in healthcare expenditures.42 More prospective studies, including randomized trials of surveillance effectiveness in influencing lung cancer specific survival, are needed to determine this definitively.
Acknowledgments
Funded by NIH/NCATS National Center for Research Resources 2 KL2 TR000421-06 KL2 Scholar ITHS Multidisciplinary Clinical Research Training Program. Views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States Government.
Appendix A: Study cohort and exclusion criteria
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Siegel R, DeSantis C, Virgo K, et al. Cancer treatment and survivorship statistics, 2012. CA Cancer J Clin. 2012 Jul-Aug;62(4):220–241. doi: 10.3322/caac.21149. [DOI] [PubMed] [Google Scholar]
- 2.Moyer VA. Screening for Lung Cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med. 2013 Dec 31; doi: 10.7326/M13-2771. [DOI] [PubMed] [Google Scholar]
- 3.Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. The New England journal of medicine. 2011 Aug 4;365(5):395–409. doi: 10.1056/NEJMoa1102873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Network NCC. [Accessed 12/29/2013];NCCN Clinical Practice Guidelines in Oncology: Non-Small Cell Lung Cancer 2013;V.2.2014. [Google Scholar]
- 5.Sause WT, Byhardt RW, Curran WJ, Jr, et al. Follow-up of non-small cell lung cancer. American College of Radiology. ACR Appropriateness Criteria. Radiology. 2000 Jun;215(Suppl):1363–1372. [PubMed] [Google Scholar]
- 6.Rubins J, Unger M, Colice GL. Follow-up and surveillance of the lung cancer patient following curative intent therapy: ACCP evidence-based clinical practice guideline (2nd edition) Chest. 2007 Sep;132(3 Suppl):355S–367S. doi: 10.1378/chest.07-1390. [DOI] [PubMed] [Google Scholar]
- 7.Pfister DG, Benson AB, 3rd, Somerfield MR. Clinical practice. Surveillance strategies after curative treatment of colorectal cancer. N Engl J Med. 2004 Jun 3;350(23):2375–2382. doi: 10.1056/NEJMcp010529. [DOI] [PubMed] [Google Scholar]
- 8.Vansteenkiste J, De Ruysscher D, Eberhardt WE, et al. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013 Oct;24(Suppl 6):vi89–98. doi: 10.1093/annonc/mdt241. [DOI] [PubMed] [Google Scholar]
- 9.Demicheli R, Fornili M, Ambrogi F, et al. Recurrence dynamics for non-small-cell lung cancer: effect of surgery on the development of metastases. J Thorac Oncol. 2012 Apr;7(4):723–730. doi: 10.1097/JTO.0b013e31824a9022. [DOI] [PubMed] [Google Scholar]
- 10.Voltolini L, Paladini P, Luzzi L, Ghiribelli C, Di Bisceglie M, Gotti G. Iterative surgical resections for local recurrent and second primary bronchogenic carcinoma. Eur J Cardiothorac Surg. 2000 Nov;18(5):529–534. doi: 10.1016/s1010-7940(00)00572-8. [DOI] [PubMed] [Google Scholar]
- 11.Egermann U, Jaeggi K, Habicht JM, Perruchoud AP, Dalquen P, Soler M. Regular follow-up after curative resection of nonsmall cell lung cancer: a real benefit for patients? Eur Respir J. 2002 Mar;19(3):464–468. doi: 10.1183/09031936.02.00231802. [DOI] [PubMed] [Google Scholar]
- 12.Endo C, Sakurada A, Notsuda H, et al. Results of long-term follow-up of patients with completely resected non-small cell lung cancer. Ann Thorac Surg. 2012 Apr;93(4):1061–1068. doi: 10.1016/j.athoracsur.2012.01.004. [DOI] [PubMed] [Google Scholar]
- 13.Walsh GL, O'Connor M, Willis KM, et al. Is follow-up of lung cancer patients after resection medically indicated and cost-effective? Ann Thorac Surg. 1995 Dec;60(6):1563–1570. doi: 10.1016/0003-4975(95)00893-4. discussion 1570-1562. [DOI] [PubMed] [Google Scholar]
- 14.Lou F, Huang J, Sima CS, Dycoco J, Rusch V, Bach PB. Patterns of recurrence and second primary lung cancer in early-stage lung cancer survivors followed with routine computed tomography surveillance. J Thorac Cardiovasc Surg. 2013 Jan;145(1):75–81. doi: 10.1016/j.jtcvs.2012.09.030. discussion 81-72. [DOI] [PubMed] [Google Scholar]
- 15.Sugimura H, Nichols FC, Yang P, et al. Survival after recurrent nonsmall-cell lung cancer after complete pulmonary resection. Ann Thorac Surg. 2007 Feb;83(2):409–417. doi: 10.1016/j.athoracsur.2006.08.046. discussioin 417-408. [DOI] [PubMed] [Google Scholar]
- 16.Westeel V, Choma D, Clement F, et al. Relevance of an intensive postoperative follow-up after surgery for non-small cell lung cancer. Ann Thorac Surg. 2000 Oct;70(4):1185–1190. doi: 10.1016/s0003-4975(00)01731-8. [DOI] [PubMed] [Google Scholar]
- 17.Benamore R, Shepherd FA, Leighl N, et al. Does intensive follow-up alter outcome in patients with advanced lung cancer? J Thorac Oncol. 2007 Apr;2(4):273–281. doi: 10.1097/01.JTO.0000263708.08332.76. [DOI] [PubMed] [Google Scholar]
- 18.Lamont JP, Kakuda JT, Smith D, Wagman LD, Grannis FW., Jr Systematic postoperative radiologic follow-up in patients with non-small cell lung cancer for detecting second primary lung cancer in stage IA. Arch Surg. 2002 Aug;137(8):935–938. doi: 10.1001/archsurg.137.8.935. discussion 938-940. [DOI] [PubMed] [Google Scholar]
- 19.Kelsey CR, Clough RW, Marks LB. Local recurrence following initial resection of NSCLC: salvage is possible with radiation therapy. Cancer J. 2006 Jul-Aug;12(4):283–288. doi: 10.1097/00130404-200607000-00006. [DOI] [PubMed] [Google Scholar]
- 20.Sugimura H, Nichols FC, Yang P, et al. Survival after recurrent nonsmall-cell lung cancer after complete pulmonary resection. The Annals of Thoracic Surgery. 2007 Feb;83(2):409–417. doi: 10.1016/j.athoracsur.2006.08.046. discussioin 417-408. [DOI] [PubMed] [Google Scholar]
- 21.Watanabe Y, Shimizu J, Oda M, Tatsuzawa Y, Hayashi Y, Iwa T. Second surgical intervention for recurrent and second primary bronchogenic carcinomas. Scand J Thorac Cardiovasc Surg. 1992;26(1):73–78. doi: 10.3109/14017439209099057. [DOI] [PubMed] [Google Scholar]
- 22.Hung JJ, Hsu WH, Hsieh CC, et al. Post-recurrence survival in completely resected stage I non-small cell lung cancer with local recurrence. Thorax. 2009 Mar;64(3):192–196. doi: 10.1136/thx.2007.094912. [DOI] [PubMed] [Google Scholar]
- 23.Westeel V, Choma D, Clement F, et al. Relevance of an intensive postoperative follow-up after surgery for non-small cell lung cancer. Ann Thorac Surg. 2000 Oct;70(4):1185–1190. doi: 10.1016/s0003-4975(00)01731-8. [DOI] [PubMed] [Google Scholar]
- 24.Aziz TM, Saad RA, Glasser J, Jilaihawi AN, Prakash D. The management of second primary lung cancers. A single centre experience in 15 years. Eur J Cardiothorac Surg. 2002 Mar;21(3):527–533. doi: 10.1016/s1010-7940(02)00024-6. [DOI] [PubMed] [Google Scholar]
- 25.Angeletti CA, Mussi A, Janni A, et al. Second primary lung cancer and relapse: treatment and follow-up. Eur J Cardiothorac Surg. 1995;9(11):607–611. doi: 10.1016/s1010-7940(05)80104-6. [DOI] [PubMed] [Google Scholar]
- 26.Virgo KS, McKirgan LW, Caputo MC, et al. Post-treatment management options for patients with lung cancer. Ann Surg. 1995 Dec;222(6):700–710. doi: 10.1097/00000658-199512000-00003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Haraguchi S, Koizumi K, Hirata T, et al. Surgical treatment of metachronous nonsmall cell lung cancer. Ann Thorac Cardiovasc Surg. 2010 Oct;16(5):319–325. [PubMed] [Google Scholar]
- 28.Calman L, Beaver K, Hind D, Lorigan P, Roberts C, Lloyd-Jones M. Survival benefits from follow-up of patients with lung cancer: a systematic review and meta-analysis. J Thorac Oncol. 2011 Dec;6(12):1993–2004. doi: 10.1097/JTO.0b013e31822b01a1. [DOI] [PubMed] [Google Scholar]
- 29.Cooper GS, Johnson CC, Lamerato L, et al. Use of guideline recommended follow-up care in cancer survivors: routine or diagnostic indications? Med Care. 2006 Jun;44(6):590–594. doi: 10.1097/01.mlr.0000215902.50543.77. [DOI] [PubMed] [Google Scholar]
- 30.Backhus LM, Hayanga AJ, Au D, Zeliadt SB. The effect of provider density on lung cancer survival among blacks and whites in the United States. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2013 May;8(5):549–553. doi: 10.1097/JTO.0b013e318287c24c. [DOI] [PubMed] [Google Scholar]
- 31.Hayanga AJ, Zeliadt SB, Backhus LM. Residential segregation and lung cancer mortality in the United States. JAMA surgery. 2013 Jan;148(1):37–42. doi: 10.1001/jamasurgery.2013.408. [DOI] [PubMed] [Google Scholar]
- 32.Ettinger DS, Cox JD, Ginsberg RJ, et al. NCCN Non-Small-Cell Lung Cancer Practice Guidelines. The National Comprehensive Cancer Network. Oncology (Williston Park) 1996 Nov;10(11 Suppl):81–111. [PubMed] [Google Scholar]
- 33.Investigators G. Impact of follow-up testing on survival and health-related quality of life in breast cancer patients. A multicenter randomized controlled trial. JAMA. 1994 May 25;271(20):1587–1592. doi: 10.1001/jama.1994.03510440047031. [DOI] [PubMed] [Google Scholar]
- 34.Hu CY, Delclos GL, Chan W, Du XL. Post-treatment surveillance in a large cohort of patients with colon cancer. Am J Manag Care. 2011 May;17(5):329–336. [PubMed] [Google Scholar]
- 35.Salz T, Weinberger M, Ayanian JZ, et al. Variation in use of surveillance colonoscopy among colorectal cancer survivors in the United States. BMC Health Serv Res. 2010;10:256. doi: 10.1186/1472-6963-10-256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Field TS, Doubeni C, Fox MP, et al. Under utilization of surveillance mammography among older breast cancer survivors. J Gen Intern Med. 2008 Feb;23(2):158–163. doi: 10.1007/s11606-007-0471-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Merkx MA, van Gulick JJ, Marres HA, et al. Effectiveness of routine follow-up of patients treated for T1-2N0 oral squamous cell carcinomas of the floor of mouth and tongue. Head Neck. 2006 Jan;28(1):1–7. doi: 10.1002/hed.20296. [DOI] [PubMed] [Google Scholar]
- 38.Elston Lafata J, Simpkins J, Schultz L, et al. Routine surveillance care after cancer treatment with curative intent. Med Care. 2005 Jun;43(6):592–599. doi: 10.1097/01.mlr.0000163656.62562.c4. [DOI] [PubMed] [Google Scholar]
- 39.Naunheim KS, Virgo KS, Coplin MA, Johnson FE. Clinical surveillance testing after lung cancer operations. Ann Thorac Surg. 1995 Dec;60(6):1612–1616. doi: 10.1016/0003-4975(95)00762-8. [DOI] [PubMed] [Google Scholar]
- 40.Johnson FE, Naunheim KS, Coplin MA, Virgo KS. Geographic variation in the conduct of patient surveillance after lung cancer surgery. J Clin Oncol. 1996 Nov;14(11):2940–2949. doi: 10.1200/JCO.1996.14.11.2940. [DOI] [PubMed] [Google Scholar]
- 41.Cooper GS, Schultz L, Simpkins J, Lafata JE. The utility of administrative data for measuring adherence to cancer surveillance care guidelines. Med Care. 2007 Jan;45(1):66–72. doi: 10.1097/01.mlr.0000241107.15133.54. [DOI] [PubMed] [Google Scholar]
- 42.Mollberg NM, Ferguson MK. Postoperative surveillance for non-small cell lung cancer resected with curative intent: developing a patient-centered approach. Ann Thorac Surg. 2013 Mar;95(3):1112–1121. doi: 10.1016/j.athoracsur.2012.09.075. [DOI] [PubMed] [Google Scholar]