0
Original Research: Lung Cancer |

When Is a Biopsy-Proven Diagnosis Necessary Before Stereotactic Ablative Radiotherapy for Lung Cancer?Stereotactic Ablative Radiotherapy for Single Node: A Decision Analysis FREE TO VIEW

Alexander V. Louie, MD; Suresh Senan, PhD; Pretesh Patel, MD; Bart S. Ferket, MD, PhD; Frank J. Lagerwaard, MD, PhD; George B. Rodrigues, MD, PhD; Joseph K. Salama, MD; Christopher Kelsey, MD; David A. Palma, MD, PhD; Myriam G. Hunink, MD, PhD
Author and Funding Information

From the Department of Radiation Oncology (Drs Louie, Senan, and Lagerwaard), VU University Medical Center, Amsterdam, The Netherlands; the Department of Radiation Oncology (Drs Louie, Rodrigues, and Palma), London Regional Cancer Program, London, ON, Canada; the Department of Epidemiology (Drs Louie, Ferket, and Hunink), Harvard School of Public Health, Boston, MA; the Department of Radiation Oncology (Drs Patel, Salama, and Kelsey), Duke University Medical Center, Durham, NC; and the Department of Epidemiology and Biostatistics (Drs Ferket and Hunink), Erasmus MC, Rotterdam, The Netherlands.

CORRESPONDENCE TO: Alexander V. Louie, MD, Department of Radiation Oncology, VU University Medical Center, De Boelelaan 1118, 1081 HV Amsterdam, The Netherlands; e-mail: Dr.alexlouie@gmail.com


FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2014;146(4):1021-1028. doi:10.1378/chest.13-2924
Text Size: A A A
Published online

BACKGROUND:  The practice of treating a solitary pulmonary nodule (SPN) suspicious for stage I non-small cell lung cancer (NSCLC) with stereotactic ablative radiotherapy (SABR) in the absence of pathology is growing. In the absence of randomized evidence, the appropriate prior probability threshold of lung cancer of when such a strategy is warranted can be informed using decision analysis.

METHODS:  A decision tree and Markov model were constructed to evaluate the relative merits of surveillance, a PET scan-directed SABR strategy (without pathology), or a PET scan-biopsy-SABR strategy, when faced with an SPN at different prior probabilities for lung cancer. Diagnostic characteristics, as well as disease, treatment, and toxicity parameters, were extracted from the literature. Deterministic analysis and probabilistic sensitivity analyses were performed to inform the appropriate lung cancer prior probability threshold between treatment strategies.

RESULTS:  In the reference case analysis, the prior probability threshold between surveillance and PET scan-biopsy-SABR was 17.0%; between PET scan-directed SABR and PET scan-biopsy-SABR, the threshold was 85.0%. The latter finding was confirmed on probabilistic sensitivity analysis (85.2%; 95% CI, 80.0% to 87.2%). This predicted lung cancer prior probability threshold was most sensitive to the diagnostic sensitivity of transthoracic biopsy (range, 77.2% to 94.0%) and the detection rate of false negatives on CT scan surveillance (range, 82.4% to 92.3%).

CONCLUSIONS:  This model suggests that if there are concerns about morbidity related to biopsy for an SPN, a PET scan-directed SABR strategy is warranted when the prior probability of lung cancer exceeds a point estimate of 85%.

Figures in this Article

Surgery is the standard of care for patients with stage I non-small cell lung cancer (NSCLC). However, factors such as age, baseline comorbidities, and personal preference preclude up to one-third of patients from undergoing surgical resection.1,2 The standard of care for patients who are medically inoperable is stereotactic ablative radiotherapy (SABR), a technique that accurately and precisely delivers highly conformal doses of radiotherapy over a few fractions.3,4 Multicenter prospective trials using SABR have reported 3-year local control rates in excess of 90%, with low toxicity.5 In addition, population-based studies have demonstrated that an increase in the use of SABR correlated with a decrease in the proportion of untreated elderly patients, resulting in significant overall survival (OS) gains.6

For patients who are fit for surgery, current guidelines suggest that a biopsy may be unnecessary for a solitary pulmonary nodule (SPN) if the probability of malignancy is > 65%.7 Although the incidence of complications of a diagnostic transthoracic biopsy, such as pneumothorax, may be acceptable in fit patients, these risks are increased in the typical SABR patient population with severe comorbidities.8 The rates of biopsy prior to SABR for early lung cancer are variable (Table 1),912 reflecting the juxtaposition of a desire for certainty before treatment and the potential toxicity from diagnostic interventions.

Table Graphic Jump Location
TABLE 1 ]  Biopsy Rates in a Sample of Studies Using SABR in Stage I NSCLC

There appears to be a trend toward increased rates of biopsy in North America compared with other parts of the world. SABR = stereotactic ablative radiotherapy.

The practice of treating an SPN suspicious for NSCLC with SABR in the absence of pathology is growing.1316 The merit of this approach in areas where 2-(fluorine-18) fluoro-2-deoxy-d-glucose (18F-FDG) PET scan avidity may be confounded by high rates of benign disease is less clear, with only a retrospective series describing outcomes.17 As evidence-based recommendations to inform the appropriate threshold in using this treatment strategy are lacking, integrating information on the pretest probability of cancer based on diagnostic characteristics with fundamental concepts from decision analysis can be helpful. The goal of this study was to determine the appropriate lung cancer prior probability threshold for treating an SPN suspicious for NSCLC with SABR in the absence of pathology, using a decision modeling approach.

Decision Tree

We developed a model to simulate the clinical history of a cohort of 75-year-old patients with a noncalcified SPN > 1 cm in size that is suspicious for stage I NSCLC. Due to severe COPD, these patients were considered medically inoperable but eligible for SABR. Patients begin with an SPN on CT scan that is further evaluated by a PET-CT scan. As depicted in the decision tree in Figure 1, three treatment strategies are compared: (1) PET scan-biopsy-SABR, (2) surveillance, and (3) PET scan-directed SABR. Institutional review board approval was not obtained because the model used data from published studies.

Figure Jump LinkFigure 1 –  Decision tree depicting treatment options for a solitary pulmonary nodule (SPN) on CT scan. Sensitivities and specificities of diagnostic tests inform initial probabilities of health states at the terminal branches, where the Markov model begins. Bx = biopsy; M = Markov model; SABR = stereotactic ablative radiotherapy.Grahic Jump Location

Based on diagnostic outcomes, patients are classified as a true positive (TP), false positive (FP), false negative (FN), or true negative (TN). The sensitivities and specificities for transthoracic needle biopsy and PET-CT imaging to calculate these probabilities were informed by a review performed in a guideline by the American College of Chest Physicians (CHEST).7 Patients testing positive for malignancy (TPs and FPs) receive SABR, whereas those testing negative for disease (FNs and TNs) are observed.

Markov Model

A state transition diagram describing the possible transitions between health states is shown in Figure 2. Although TP patients treated with SABR enter into a health state of no evidence of disease immediately after treatment (although they may experience recurrence thereafter), FP patients continue to cycle in a well-health state unless they experience a treatment-related toxicity. For TP patients, additional health states were defined to represent local recurrence, regional recurrence, and distant metastasis. To simplify the model, we assumed that patients with recurrence were ineligible for salvage treatment due to comorbidity. Transition probabilities for recurrences were calculated from individual patient data from a study previously comparing patients undergoing SABR with and without pathology.15 Patients with recurrent cancer will die of cancer, according to probabilities calculated from a meta-analysis.18

Figure Jump LinkFigure 2 –  State transition diagram of health states. Note that DEATH includes death from other causes, biopsy-related death, and cancer-related death. DM = distant metastasis; FN = false negative; FP = false positive; LR = local recurrence; NED = no evidence of disease (true positive); RR = regional recurrence; TN = true negative; TP = true positive.Grahic Jump Location

For FN patients, disease could be detected based on growth on serial CT scans at 3 and 6 months. As there is a paucity of data to inform the probability of this, our multidisciplinary group arrived at a consensus for the reference value. As the prior probability of malignancy in lesions with serial growth is very high, these were treated with SABR without considering additional testing (PET scan-directed SABR or PET scan-biopsy-SABR). This assumption is supported by findings from the Dutch-Belgian lung cancer screening trial (NELSON), in which additional testing is rarely performed for indeterminate SPNs that are found to be positive on recall CT scan.19

Patients in any health state could also die of other causes calculated using United States standard life tables.20 As patients with severe COPD have an increased rate of death from other causes, a hazard rate ratio on standard death risks was applied. Hazard rate ratios were calculated from a meta-analysis of risk stratified by smoking habit.21 The model was constructed with a 5-year time horizon using a monthly cycle length, and quality adjusted life years (QALYs) were discounted at a 3% rate.

Toxicities

Patients who undergo a biopsy were at risk for hemorrhage and pneumothorax. We considered only patients with a major pneumothorax requiring a chest tube as a clinically significant event; the probability of this event was informed by an American population-based analysis.22 Although rare, patients could also die of the biopsy procedure itself.7 In addition to treatment-related death,23 patients receiving SABR were also at risk for both radiation pneumonitis and chest wall injury/fracture. Both were assumed to last for 6 months, consistent with a previous Markov model.24

Utilities

The EuroQol five-dimension (EQ-5D) instrument is the preferred measure to obtain health state utilities to calculate QALYs.25 Although utilities of patients with early-stage NSCLC treated with SABR have not been well studied, quality-of-life reports using the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 have described minimal clinically relevant deterioration.26 We used the methods described by Jang et al27 in a process called mapping to derive a time-weighted EQ-5D-based utility after SABR from 382 prospective individually collected EORTC QLQ-C30 scores.26

The absorbing health state of death was assigned a utility of 0. Utilities of patients with recurrence were defined using pooled estimates from a meta-analysis.28 For toxicity, the presence of chest wall pain and/or radiation pneumonitis from SABR can be calculated as a disutility subtracted from the reference value for the first year,24 an assumption that is derived from a linear regression in a prior study and robustly tested in sensitivity analyses.29 Patients who experience a pneumothorax from biopsy requiring a chest tube experience a one-time disutility.30 All health states in the model were also adjusted for COPD severity using the product of COPD utility as described by Fryback and Lawrence31 and used in another SABR Markov model.32

Model Validation

The model was internally validated by comparing OS outcomes at 1, 3, and 5 years for treatment with and without pathologic confirmation of malignancy. To facilitate external validation of our model with outcomes of untreated patients with early lung cancer in the California Cancer Registry,33 the surveillance strategy was modified to create an observation strategy, whereby all FNs are instead left untreated. Although PET scan staging is considered standard in the work-up of an SPN (to characterize the lesion and rule out metastases),34 two additional strategies (1) CT scan-SABR, and (2) CT scan-biopsy-SABR were evaluated to ensure that predicted thresholds were internally consistent (e-Fig 1). Finally, a probabilistic sensitivity analysis (PSA) of the PET scan-directed SABR and PET scan-biopsy-SABR prior probability threshold was undertaken, with parameter reference values and ranges summarized in Table 235,36 (e-Appendix 1).

Table Graphic Jump Location
TABLE 2 ]  Model Parameters

DM = distant metastasis; FN = false negative; GOLD = Global Initiative for Chronic Obstructive Lung Disease; HR = hazard ratio; LR = local recurrence; NED = no evidence of disease; RR = regional recurrence; SA = sensitivity analysis. See Table 1 legend for expansion of other abbreviation.

a 

Range studied in deterministic sensitivity analysis.

b 

Lung cancer prior probability threshold range based on two-way SA, varying prior probability with each parameter in the model.

c 

As sensitivity and specificity are related, sensitivity analysis was performed on these using a constant diagnostic OR, calculated from reference values.

d 

Transitions are listed as annual rates to allow for easier interpretation. In the model, these are converted into monthly probabilities to coincide with the 1-mo cycle length.

Model Validity

The constructed model was calibrated for each treatment modality and internally validated by showing close correlation with OS outcomes of the source data for patients treated with SABR both with and without pathology.15 The predicted OS was within 2% annually for the SABR treatment strategies. Observation yielded OS rates of 56.2%, 25.0%, and 9.5% at 1, 3, and 5 years, respectively, consistent with outcomes from the California Cancer Registry, providing evidence of external validity.33 Comparison of each treatment strategy with its validation comparator is summarized in Table 3.

Table Graphic Jump Location
TABLE 3 ]  Model Validation

The model was validated internally for survival outcomes in patients treated with and without pathology. External validation of observation strategy performed with data from the California Cancer Registry. See Table 1 legend for expansion of abbreviation.

Reference Case Scenario

At a lung cancer prior probability of 65%, PET scan-biopsy-SABR was the preferred treatment strategy. This yielded 2.640 QALYs, compared with 2.563 and 2.086 for the PET scan-directed SABR and surveillance strategies, respectively.

Sensitivity Analyses on Prior Probability Threshold

Varying the lung cancer prior probability (Fig 1, disease + vs disease −) from 0% to 100% in one-way sensitivity analysis revealed that the threshold for the decision between the PET scan-biopsy-SABR and PET scan-directed SABR strategies was 85.0%; between PET scan-biopsy-SABR and surveillance, the prior probability threshold was 17.0%. For each parameter listed in Table 2, a two-way sensitivity analysis was performed while varying the lung cancer prior probability threshold. The uncertainty of biopsy sensitivity had the largest influence on the first threshold, ranging from 77.2% to 94.0%. The uncertainty of the FN detection rate on surveillance CT scan also had a significant influence, ranging from 82.4% to 92.3%. Most other variables had limited effect on the prior probability threshold (< 5% variation); results of these two-way sensitivity analyses are summarized in Table 2. Additional clinically relevant two-way sensitivity analyses beyond the meta-analytic and consensus-based ranges tested are available in e-Appendix 2. PSA recapitulated the findings on deterministic sensitivity analyses, predicting a prior probability threshold of 85.2%, with a 95% CI between 80.0% and 87.2%.

Rates of obtaining pre-SABR pathologic diagnoses are variable, reflecting clinical uncertainty as to the necessity of biopsy vs the risks of complications.37 To address these concerns, we developed a model that demonstrates that treating an SPN without pathology is justified when the likelihood of malignancy is > 85%, an estimate robust over a wide range of sensitivity analyses. The model was internally and externally validated and further predicted that if the probability of malignancy of an SPN was > 17.0%, a biopsy should be performed rather than surveillance, consistent with the findings of another decision analysis.38Figure 3 depicts a suggested algorithm for patients with an SPN suspicious on CT scan for NSCLC being considered for SABR.

Figure Jump LinkFigure 3 –  Suggested algorithm for a patient with an SPN, suspicious for lung cancer on CT scan. Horizontal arrows indicate factors that may increase or decrease the lung cancer prior probability threshold between treatment strategies. See Figure 1 and 2 legends for expansion of abbreviations.Grahic Jump Location

To our knowledge, our model is the first to quantitatively assess the threshold for performing SABR in the absence of pathology, and it is consistent with recently published expert opinion. The CHEST guideline states that in patients being considered for nonsurgical treatments like SABR, the diagnosis of lung cancer ideally should be confirmed by biopsy and is a reasonable treatment modality when the pretest probability of malignancy exceeds 65%.7 As this proposed threshold has been informed using surgical data,30 others have argued that 85% is more appropriate, despite the low morbidity of SABR, as these patients typically have increased comorbidities.39 This latter estimate is consistent with the recommendations of the International Association for the Study of Lung Cancer, which states that a pathologic diagnosis of benign disease should not exceed 15% in centers performing CT scan screening.40 An important distinction of our proposed prior probability threshold is that SABR without biopsy is warranted only in a PET scan-avid lesion; in the CHEST guidelines, surgery is recommended independent of PET scan avidity above a 65% threshold, with the use of PET scan for staging purposes. Although our prior probability threshold of 85% exceeds this surgical threshold, concluding that surgery is the preferred modality in patients who are candidates for either surgery or SABR is a flawed comparison. Although obtaining histologic information during surgery may be acceptable in patients fit for surgery, this may not be the case in a 75-year-old patient with severe COPD.

Our threshold recommendation has important implications, given the reported variability in the diagnostic performance of biopsy and the potential risk associated with FPs in suspected lung cancer. We found that transthoracic needle biopsy sensitivity has the greatest impact on the proposed lung cancer prior probability threshold. Although biopsy has a sensitivity of > 90% in most studies, the frequency of nondiagnostic results is highly variable, with a median of 6%.7,41 In terms of FPs, in an American population-based analysis, the rate of benign diagnosis after surgery for known or suspected lung cancer was 9.1% (ranging from 1.3% in Vermont to 25.0% in Hawaii), with an in-hospitality mortality rate of 2.3%.42

The prior probability threshold was also highly sensitive to our assumed detection rate of FNs on surveillance CT scan. Although population data indicate that the survival of untreated stage I NSCLC is poor,33 this was in the setting of biopsy-proven disease. The prognosis of undetected malignancy is unclear, although studies indicate that the volume-doubling time of lung cancer ranges from approximately 150 to 480 days.43 Quantifying the hazards of these uncertainties will be the subject of future modeling research from our group.

Our model can be applied at the patient level by calculating likelihood of malignancy as an estimate of prior probability. Although clinicians may estimate the pretest probability of malignancy of an SPN, prediction models based on patient and CT scan characteristics have been validated in the Dutch14 and American populations.16,44 Although these models typically did not include lesions < 1 cm in diameter, incorporating newer features, such as nodule volume-doubling time, appears to improve accuracy for smaller lesions.44 This is of particular importance for CT scan-screened patients in whom new lesions are smaller; indeed, the NELSON CT scan screening trial allowed for treatment in patients with a growing SPN without a histologic diagnosis for high-dose radiotherapy.19 Calculators to predict the likelihood of malignancy have also been developed for patients screened with low-dose CT scan.45 Ultimately, although studies indicate that the accuracy of models for predicting malignancy is similar to that of experts, correlation between the two is poor, suggesting that models provide complementary information to a multidisciplinary discussion.46

The results of our model should be considered in the context of both its strengths and limitations. Model inputs were validated internally and were derived from meta-analyses, systematic reviews, and individual-level patient data, allowing for robust statistical calculations. The reference values chosen were agreed upon by all authors a priori so as to not bias the results. Findings on deterministic sensitivity analysis were confirmed in PSA, allowing for generalizable conclusions. In terms of limitations, like all models, key assumptions were made. Although the outputs of an observation strategy were validated externally, we recognize that few data inform this particular analysis. Although the generation of SABR-specific utility values is a novel contribution to the medical literature, the use of the EQ-5D mapping process has been called into question by some investigators.47 Nonetheless, to our knowledge, SABR-specific utilities currently do not exist in the medical literature, and our model can be updated to incorporate new evidence as it becomes available.

In conclusion, our model provides a practical framework for physicians to determine the appropriateness of SABR for an SPN suspicious for lung cancer, where comorbidities may heighten concerns for complications related to biopsy. From a health policy point of view, our results argue for the continued use of SABR in the absence of pathologic confirmation when the prior probability of lung cancer is > 85%. The authors encourage referral of such patients to the multidisciplinary team, so that SABR may be discussed as a potential option.

Author contributions: A. V. L. is the guarantor of the content of the manuscript, including data and analysis and is the primary author. M. G. H. is the principal investigator of the study. A. V. L. contributed to leading study inception, study design, writing, analysis, review, and final approval; S. S., B. S. F., and D. A. P. contributed to study design, writing, analysis, review, and final approval; P. P., F. J. L., G. B. R., J. K. S., C.K., and M. G. H. contributed to writing, analysis, review, revision, and final approval; and D. A. P. contributed to study inception.

Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Louie is the 2013 recipient of the CARO-Elekta Research Fellowship, the Royal College of Physicians and Surgeons of Canada Detweiler Travelling Fellowship, and the Resident Research Career Development Program Award from Western University. Drs Senan and Lagerwaard have received speakers’ honoraria from Varian Medical Systems, Inc. The VU University Medical Center has a research agreement with Varian Medical Systems, Inc. Drs Patel, Ferket, Rodrigues, Salama, Kelsey, Palma, and Hunink have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Additional information: The e-Figure and e-Appendixes can be found in the Supplemental Materials section of the online article.

CHEST

American College of Chest Physicians

EQ-5D

EuroQol five-dimension

FN

false negative

FP

false positive

NSCLC

non-small cell lung cancer

OS

overall survival

PSA

probabilistic sensitivity analysis

QALY

quality-adjusted life year

SABR

stereotactic ablative radiotherapy

SPN

solitary pulmonary nodule

TN

true negative

TP

true positive

Cykert S, Dilworth-Anderson P, Monroe MH, et al. Factors associated with decisions to undergo surgery among patients with newly diagnosed early-stage lung cancer. JAMA. 2010;303(23):2368-2376. [CrossRef] [PubMed]
 
Howington JA, Blum MG, Chang AC, Balekian AA, Murthy SC. Treatment of stage I and II non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143(5_suppl):e278S-e313S. [CrossRef] [PubMed]
 
Donington J, Ferguson M, Mazzone P, et al; Thoracic Oncology Network of American College of Chest Physicians; Workforce on Evidence-Based Surgery of Society of Thoracic Surgeons. American College of Chest Physicians and Society of Thoracic Surgeons consensus statement for evaluation and management for high-risk patients with stage I non-small cell lung cancer. Chest. 2012;142(6):1620-1635. [CrossRef] [PubMed]
 
Simone CB II, Wildt B, Haas AR, Pope G, Rengan R, Hahn SM. Stereotactic body radiation therapy for lung cancer. Chest. 2013;143(6):1784-1790. [CrossRef] [PubMed]
 
Timmerman R, Paulus R, Galvin J, et al. Stereotactic body radiation therapy for inoperable early stage lung cancer. JAMA. 2010;303(11):1070-1076. [CrossRef] [PubMed]
 
Haasbeek CJ, Palma D, Visser O, Lagerwaard FJ, Slotman B, Senan S. Early-stage lung cancer in elderly patients: a population-based study of changes in treatment patterns and survival in the Netherlands. Ann Oncol. 2012;23(10):2743-2747. [CrossRef] [PubMed]
 
Gould MK, Donington J, Lynch WR, et al. Evaluation of individuals with pulmonary nodules: when is it lung cancer? Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143(5_suppl):e93S-e120S. [CrossRef] [PubMed]
 
Hiraki T, Mimura H, Gobara H, et al. CT fluoroscopy-guided biopsy of 1,000 pulmonary lesions performed with 20-gauge coaxial cutting needles: diagnostic yield and risk factors for diagnostic failure. Chest. 2009;136(6):1612-1617. [CrossRef] [PubMed]
 
Guckenberger M, Allgäuer M, Appold S, et al. Safety and efficacy of stereotactic body radiotherapy for stage 1 non-small-cell lung cancer in routine clinical practice: a patterns-of-care and outcome analysis. J Thorac Oncol. 2013;8(8):1050-1058. [CrossRef] [PubMed]
 
Grills IS, Hope AJ, Guckenberger M, et al. A collaborative analysis of stereotactic lung radiotherapy outcomes for early-stage non-small-cell lung cancer using daily online cone-beam computed tomography image-guided radiotherapy. J Thorac Oncol. 2012;7(9):1382-1393. [CrossRef] [PubMed]
 
Onishi H, Yoshiyuki S, Yasuo M, et al. Japanese multi-institutional study of stereotactic body radiation therapy for more than 2000 patients with stage I non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2013;87(suppl 2):S9-S10. [CrossRef]
 
Baumann P, Nyman J, Hoyer M, et al. Outcome in a prospective phase II trial of medically inoperable stage I non-small-cell lung cancer patients treated with stereotactic body radiotherapy. J Clin Oncol. 2009;27(20):3290-3296. [CrossRef] [PubMed]
 
Vansteenkiste J, De Ruysscher D, Eberhardt WE, et al; ESMO Guidelines Working Group. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24(suppl 6):vi89-vi98. [CrossRef] [PubMed]
 
Herder GJ, van Tinteren H, Golding RP, et al. Clinical prediction model to characterize pulmonary nodules: validation and added value of18F-fluorodeoxyglucose positron emission tomography. Chest. 2005;128(4):2490-2496. [CrossRef] [PubMed]
 
Verstegen NE, Lagerwaard FJ, Haasbeek CJ, Slotman BJ, Senan S. Outcomes of stereotactic ablative radiotherapy following a clinical diagnosis of stage I NSCLC: comparison with a contemporaneous cohort with pathologically proven disease. Radiother Oncol. 2011;101(2):250-254. [CrossRef] [PubMed]
 
Gould MK, Ananth L, Barnett PG, Veterans Affairs SCSG; Veterans Affairs SNAP Cooperative Study Group. A clinical model to estimate the pretest probability of lung cancer in patients with solitary pulmonary nodules. Chest. 2007;131(2):383-388. [CrossRef] [PubMed]
 
Harkenrider MM, Bertke MH, Dunlap NE. Stereotactic body radiation therapy for unbiopsied early-stage lung cancer: a multi-institutional analysis. Am J Clin Oncol. 2014;37(4):337-342. [CrossRef] [PubMed]
 
NSCLC Meta-Analyses Collaborative Group. Chemotherapy in addition to supportive care improves survival in advanced non-small-cell lung cancer: a systematic review and meta-analysis of individual patient data from 16 randomized controlled trials. J Clin Oncol. 2008;26(28):4617-4625. [CrossRef] [PubMed]
 
van Klaveren RJ, Oudkerk M, Prokop M, et al. Management of lung nodules detected by volume CT scanning. N Engl J Med. 2009;361(23):2221-2229. [CrossRef] [PubMed]
 
Arias E, Curtin LR, Wei R, Anderson RN. United States life tables, 2008. Natl Vital Stat Rep. 2012;61(3):1-64. [PubMed]
 
Shavelle RM, Paculdo DR, Strauss DJ, Kush SJ. Smoking habit and mortality: a meta-analysis. J Insur Med. 2008;40(3-4):170-178. [PubMed]
 
Wiener RS, Schwartz LM, Woloshin S, Welch HG. Population-based risk for complications after transthoracic needle lung biopsy of a pulmonary nodule: an analysis of discharge records. Ann Intern Med. 2011;155(3):137-144. [CrossRef] [PubMed]
 
Grutters JP, Kessels AG, Pijls-Johannesma M, De Ruysscher D, Joore MA, Lambin P. Comparison of the effectiveness of radiotherapy with photons, protons and carbon-ions for non-small cell lung cancer: a meta-analysis. Radiother Oncol. 2010;95(1):32-40. [CrossRef] [PubMed]
 
Shah A, Hahn SM, Stetson RL, Friedberg JS, Pechet TT, Sher DJ. Cost-effectiveness of stereotactic body radiation therapy versus surgical resection for stage I non-small cell lung cancer. Cancer. 2013;119(17):3123-3132. [CrossRef] [PubMed]
 
Longworth L, Rowen D. Mapping to obtain EQ-5D utility values for use in NICE health technology assessments. Value Health. 2013;16(1):202-210. [CrossRef] [PubMed]
 
Lagerwaard FJ, Aaronson NK, Gundy CM, Haasbeek CJ, Slotman BJ, Senan S. Patient-reported quality of life after stereotactic ablative radiotherapy for early-stage lung cancer. J Thorac Oncol. 2012;7(7):1148-1154. [CrossRef] [PubMed]
 
Jang RW, Isogai PK, Mittmann N, et al. Derivation of utility values from European Organization for Research and Treatment of Cancer Quality of Life-Core 30 questionnaire values in lung cancer. J Thorac Oncol. 2010;5(12):1953-1957. [CrossRef] [PubMed]
 
Sturza J. A review and meta-analysis of utility values for lung cancer. Med Decis Making. 2010;30(6):685-693. [CrossRef] [PubMed]
 
Doyle S, Lloyd A, Walker M. Health state utility scores in advanced non-small cell lung cancer. Lung Cancer. 2008;62(3):374-380. [CrossRef] [PubMed]
 
Gould MK, Sanders GD, Barnett PG, et al. Cost-effectiveness of alternative management strategies for patients with solitary pulmonary nodules. Ann Intern Med. 2003;138(9):724-735. [CrossRef] [PubMed]
 
Fryback DG, Lawrence WF Jr. Dollars may not buy as many QALYs as we think: a problem with defining quality-of-life adjustments. Med Decis Making. 1997;17(3):276-284. [CrossRef] [PubMed]
 
Louie AV, Rodrigues G, Hannouf M, et al. Withholding stereotactic radiotherapy in elderly patients with stage I non-small cell lung cancer and co-existing COPD is not justified: outcomes of a Markov model analysis. Radiother Oncol. 2011;99(2):161-165. [CrossRef] [PubMed]
 
Raz DJ, Zell JA, Ou SH, Gandara DR, Anton-Culver H, Jablons DM. Natural history of stage I non-small cell lung cancer: implications for early detection. Chest. 2007;132(1):193-199. [CrossRef] [PubMed]
 
National Comprehensive Cancer Network. NCCN guidelines for non-small cell lung cancer (V.3.2014). National Comprehensive Cancer Network website. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp. Accessed March 21, 2014.
 
Cronin P, Dwamena BA, Kelly AM, Carlos RC. Solitary pulmonary nodules: meta-analytic comparison of cross-sectional imaging modalities for diagnosis of malignancy. Radiology. 2008;246(3):772-782. [CrossRef] [PubMed]
 
Menn P, Leidl R, Holle R. A lifetime Markov model for the economic evaluation of chronic obstructive pulmonary disease. Pharmacoeconomics. 2012;30(9):825-840. [CrossRef] [PubMed]
 
Cerfolio RJ, Bryant AS. Survival of patients with true pathologic stage I non-small cell lung cancer. Ann Thorac Surg. 2009;88(3):917-922. [CrossRef] [PubMed]
 
Ost DE, Gould MK. Decision making in patients with pulmonary nodules. Am J Respir Crit Care Med. 2012;185(4):363-372. [CrossRef] [PubMed]
 
Senan S, Paul MA, Lagerwaard FJ. Treatment of early-stage lung cancer detected by screening: surgery or stereotactic ablative radiotherapy? Lancet Oncol. 2013;14(7):e270-e274. [CrossRef] [PubMed]
 
Field JK, Smith RA, Aberle DR, et al; IASLC CT Screening Workshop 2011 Participants. International Association for the Study of Lung Cancer Computed Tomography Screening Workshop 2011 report. J Thorac Oncol. 2012;7(1):10-19. [CrossRef] [PubMed]
 
Almeida FA, Casal RF, Jimenez CA, et al. Quality gaps and comparative effectiveness in lung cancer staging: the impact of test sequencing on outcomes. Chest. 2013;144(6):1776-1782. [CrossRef] [PubMed]
 
Deppen SA, Phillips S, McPheeters M, et al. Benign disease prevalence after surgical lung resection varies geographically in the US Medicare population, implications for lung cancer screening. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; April 6-10, 2013; Washington, DC. Abstract 3628.
 
Detterbeck FC, Gibson CJ. Turning gray: the natural history of lung cancer over time. J Thorac Oncol. 2008;3(7):781-792. [CrossRef] [PubMed]
 
Mehta HJ, Ravenel JG, Shaftman SR, et al. The utility of nodule volume in the context of malignancy prediction for small pulmonary nodules. Chest. 2014;145(3):464-472. [CrossRef] [PubMed]
 
McWilliams A, Tammemagi MC, Mayo JR, et al. Probability of cancer in pulmonary nodules detected on first screening CT. N Engl J Med. 2013;369(10):910-919. [CrossRef] [PubMed]
 
Balekian AA, Silvestri GA, Simkovich SM, et al. Accuracy of clinicians and models for estimating the probability that a pulmonary nodule is malignant. Ann Am Thorac Soc. 2013;10(6):629-635. [CrossRef] [PubMed]
 
Crott R, Versteegh M, Uyl-de-Groot C. An assessment of the external validity of mapping QLQ-C30 to EQ-5D preferences. Qual Life Res. 2013;22(5):1045-1054. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1 –  Decision tree depicting treatment options for a solitary pulmonary nodule (SPN) on CT scan. Sensitivities and specificities of diagnostic tests inform initial probabilities of health states at the terminal branches, where the Markov model begins. Bx = biopsy; M = Markov model; SABR = stereotactic ablative radiotherapy.Grahic Jump Location
Figure Jump LinkFigure 2 –  State transition diagram of health states. Note that DEATH includes death from other causes, biopsy-related death, and cancer-related death. DM = distant metastasis; FN = false negative; FP = false positive; LR = local recurrence; NED = no evidence of disease (true positive); RR = regional recurrence; TN = true negative; TP = true positive.Grahic Jump Location
Figure Jump LinkFigure 3 –  Suggested algorithm for a patient with an SPN, suspicious for lung cancer on CT scan. Horizontal arrows indicate factors that may increase or decrease the lung cancer prior probability threshold between treatment strategies. See Figure 1 and 2 legends for expansion of abbreviations.Grahic Jump Location

Tables

Table Graphic Jump Location
TABLE 1 ]  Biopsy Rates in a Sample of Studies Using SABR in Stage I NSCLC

There appears to be a trend toward increased rates of biopsy in North America compared with other parts of the world. SABR = stereotactic ablative radiotherapy.

Table Graphic Jump Location
TABLE 2 ]  Model Parameters

DM = distant metastasis; FN = false negative; GOLD = Global Initiative for Chronic Obstructive Lung Disease; HR = hazard ratio; LR = local recurrence; NED = no evidence of disease; RR = regional recurrence; SA = sensitivity analysis. See Table 1 legend for expansion of other abbreviation.

a 

Range studied in deterministic sensitivity analysis.

b 

Lung cancer prior probability threshold range based on two-way SA, varying prior probability with each parameter in the model.

c 

As sensitivity and specificity are related, sensitivity analysis was performed on these using a constant diagnostic OR, calculated from reference values.

d 

Transitions are listed as annual rates to allow for easier interpretation. In the model, these are converted into monthly probabilities to coincide with the 1-mo cycle length.

Table Graphic Jump Location
TABLE 3 ]  Model Validation

The model was validated internally for survival outcomes in patients treated with and without pathology. External validation of observation strategy performed with data from the California Cancer Registry. See Table 1 legend for expansion of abbreviation.

References

Cykert S, Dilworth-Anderson P, Monroe MH, et al. Factors associated with decisions to undergo surgery among patients with newly diagnosed early-stage lung cancer. JAMA. 2010;303(23):2368-2376. [CrossRef] [PubMed]
 
Howington JA, Blum MG, Chang AC, Balekian AA, Murthy SC. Treatment of stage I and II non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143(5_suppl):e278S-e313S. [CrossRef] [PubMed]
 
Donington J, Ferguson M, Mazzone P, et al; Thoracic Oncology Network of American College of Chest Physicians; Workforce on Evidence-Based Surgery of Society of Thoracic Surgeons. American College of Chest Physicians and Society of Thoracic Surgeons consensus statement for evaluation and management for high-risk patients with stage I non-small cell lung cancer. Chest. 2012;142(6):1620-1635. [CrossRef] [PubMed]
 
Simone CB II, Wildt B, Haas AR, Pope G, Rengan R, Hahn SM. Stereotactic body radiation therapy for lung cancer. Chest. 2013;143(6):1784-1790. [CrossRef] [PubMed]
 
Timmerman R, Paulus R, Galvin J, et al. Stereotactic body radiation therapy for inoperable early stage lung cancer. JAMA. 2010;303(11):1070-1076. [CrossRef] [PubMed]
 
Haasbeek CJ, Palma D, Visser O, Lagerwaard FJ, Slotman B, Senan S. Early-stage lung cancer in elderly patients: a population-based study of changes in treatment patterns and survival in the Netherlands. Ann Oncol. 2012;23(10):2743-2747. [CrossRef] [PubMed]
 
Gould MK, Donington J, Lynch WR, et al. Evaluation of individuals with pulmonary nodules: when is it lung cancer? Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143(5_suppl):e93S-e120S. [CrossRef] [PubMed]
 
Hiraki T, Mimura H, Gobara H, et al. CT fluoroscopy-guided biopsy of 1,000 pulmonary lesions performed with 20-gauge coaxial cutting needles: diagnostic yield and risk factors for diagnostic failure. Chest. 2009;136(6):1612-1617. [CrossRef] [PubMed]
 
Guckenberger M, Allgäuer M, Appold S, et al. Safety and efficacy of stereotactic body radiotherapy for stage 1 non-small-cell lung cancer in routine clinical practice: a patterns-of-care and outcome analysis. J Thorac Oncol. 2013;8(8):1050-1058. [CrossRef] [PubMed]
 
Grills IS, Hope AJ, Guckenberger M, et al. A collaborative analysis of stereotactic lung radiotherapy outcomes for early-stage non-small-cell lung cancer using daily online cone-beam computed tomography image-guided radiotherapy. J Thorac Oncol. 2012;7(9):1382-1393. [CrossRef] [PubMed]
 
Onishi H, Yoshiyuki S, Yasuo M, et al. Japanese multi-institutional study of stereotactic body radiation therapy for more than 2000 patients with stage I non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2013;87(suppl 2):S9-S10. [CrossRef]
 
Baumann P, Nyman J, Hoyer M, et al. Outcome in a prospective phase II trial of medically inoperable stage I non-small-cell lung cancer patients treated with stereotactic body radiotherapy. J Clin Oncol. 2009;27(20):3290-3296. [CrossRef] [PubMed]
 
Vansteenkiste J, De Ruysscher D, Eberhardt WE, et al; ESMO Guidelines Working Group. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24(suppl 6):vi89-vi98. [CrossRef] [PubMed]
 
Herder GJ, van Tinteren H, Golding RP, et al. Clinical prediction model to characterize pulmonary nodules: validation and added value of18F-fluorodeoxyglucose positron emission tomography. Chest. 2005;128(4):2490-2496. [CrossRef] [PubMed]
 
Verstegen NE, Lagerwaard FJ, Haasbeek CJ, Slotman BJ, Senan S. Outcomes of stereotactic ablative radiotherapy following a clinical diagnosis of stage I NSCLC: comparison with a contemporaneous cohort with pathologically proven disease. Radiother Oncol. 2011;101(2):250-254. [CrossRef] [PubMed]
 
Gould MK, Ananth L, Barnett PG, Veterans Affairs SCSG; Veterans Affairs SNAP Cooperative Study Group. A clinical model to estimate the pretest probability of lung cancer in patients with solitary pulmonary nodules. Chest. 2007;131(2):383-388. [CrossRef] [PubMed]
 
Harkenrider MM, Bertke MH, Dunlap NE. Stereotactic body radiation therapy for unbiopsied early-stage lung cancer: a multi-institutional analysis. Am J Clin Oncol. 2014;37(4):337-342. [CrossRef] [PubMed]
 
NSCLC Meta-Analyses Collaborative Group. Chemotherapy in addition to supportive care improves survival in advanced non-small-cell lung cancer: a systematic review and meta-analysis of individual patient data from 16 randomized controlled trials. J Clin Oncol. 2008;26(28):4617-4625. [CrossRef] [PubMed]
 
van Klaveren RJ, Oudkerk M, Prokop M, et al. Management of lung nodules detected by volume CT scanning. N Engl J Med. 2009;361(23):2221-2229. [CrossRef] [PubMed]
 
Arias E, Curtin LR, Wei R, Anderson RN. United States life tables, 2008. Natl Vital Stat Rep. 2012;61(3):1-64. [PubMed]
 
Shavelle RM, Paculdo DR, Strauss DJ, Kush SJ. Smoking habit and mortality: a meta-analysis. J Insur Med. 2008;40(3-4):170-178. [PubMed]
 
Wiener RS, Schwartz LM, Woloshin S, Welch HG. Population-based risk for complications after transthoracic needle lung biopsy of a pulmonary nodule: an analysis of discharge records. Ann Intern Med. 2011;155(3):137-144. [CrossRef] [PubMed]
 
Grutters JP, Kessels AG, Pijls-Johannesma M, De Ruysscher D, Joore MA, Lambin P. Comparison of the effectiveness of radiotherapy with photons, protons and carbon-ions for non-small cell lung cancer: a meta-analysis. Radiother Oncol. 2010;95(1):32-40. [CrossRef] [PubMed]
 
Shah A, Hahn SM, Stetson RL, Friedberg JS, Pechet TT, Sher DJ. Cost-effectiveness of stereotactic body radiation therapy versus surgical resection for stage I non-small cell lung cancer. Cancer. 2013;119(17):3123-3132. [CrossRef] [PubMed]
 
Longworth L, Rowen D. Mapping to obtain EQ-5D utility values for use in NICE health technology assessments. Value Health. 2013;16(1):202-210. [CrossRef] [PubMed]
 
Lagerwaard FJ, Aaronson NK, Gundy CM, Haasbeek CJ, Slotman BJ, Senan S. Patient-reported quality of life after stereotactic ablative radiotherapy for early-stage lung cancer. J Thorac Oncol. 2012;7(7):1148-1154. [CrossRef] [PubMed]
 
Jang RW, Isogai PK, Mittmann N, et al. Derivation of utility values from European Organization for Research and Treatment of Cancer Quality of Life-Core 30 questionnaire values in lung cancer. J Thorac Oncol. 2010;5(12):1953-1957. [CrossRef] [PubMed]
 
Sturza J. A review and meta-analysis of utility values for lung cancer. Med Decis Making. 2010;30(6):685-693. [CrossRef] [PubMed]
 
Doyle S, Lloyd A, Walker M. Health state utility scores in advanced non-small cell lung cancer. Lung Cancer. 2008;62(3):374-380. [CrossRef] [PubMed]
 
Gould MK, Sanders GD, Barnett PG, et al. Cost-effectiveness of alternative management strategies for patients with solitary pulmonary nodules. Ann Intern Med. 2003;138(9):724-735. [CrossRef] [PubMed]
 
Fryback DG, Lawrence WF Jr. Dollars may not buy as many QALYs as we think: a problem with defining quality-of-life adjustments. Med Decis Making. 1997;17(3):276-284. [CrossRef] [PubMed]
 
Louie AV, Rodrigues G, Hannouf M, et al. Withholding stereotactic radiotherapy in elderly patients with stage I non-small cell lung cancer and co-existing COPD is not justified: outcomes of a Markov model analysis. Radiother Oncol. 2011;99(2):161-165. [CrossRef] [PubMed]
 
Raz DJ, Zell JA, Ou SH, Gandara DR, Anton-Culver H, Jablons DM. Natural history of stage I non-small cell lung cancer: implications for early detection. Chest. 2007;132(1):193-199. [CrossRef] [PubMed]
 
National Comprehensive Cancer Network. NCCN guidelines for non-small cell lung cancer (V.3.2014). National Comprehensive Cancer Network website. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp. Accessed March 21, 2014.
 
Cronin P, Dwamena BA, Kelly AM, Carlos RC. Solitary pulmonary nodules: meta-analytic comparison of cross-sectional imaging modalities for diagnosis of malignancy. Radiology. 2008;246(3):772-782. [CrossRef] [PubMed]
 
Menn P, Leidl R, Holle R. A lifetime Markov model for the economic evaluation of chronic obstructive pulmonary disease. Pharmacoeconomics. 2012;30(9):825-840. [CrossRef] [PubMed]
 
Cerfolio RJ, Bryant AS. Survival of patients with true pathologic stage I non-small cell lung cancer. Ann Thorac Surg. 2009;88(3):917-922. [CrossRef] [PubMed]
 
Ost DE, Gould MK. Decision making in patients with pulmonary nodules. Am J Respir Crit Care Med. 2012;185(4):363-372. [CrossRef] [PubMed]
 
Senan S, Paul MA, Lagerwaard FJ. Treatment of early-stage lung cancer detected by screening: surgery or stereotactic ablative radiotherapy? Lancet Oncol. 2013;14(7):e270-e274. [CrossRef] [PubMed]
 
Field JK, Smith RA, Aberle DR, et al; IASLC CT Screening Workshop 2011 Participants. International Association for the Study of Lung Cancer Computed Tomography Screening Workshop 2011 report. J Thorac Oncol. 2012;7(1):10-19. [CrossRef] [PubMed]
 
Almeida FA, Casal RF, Jimenez CA, et al. Quality gaps and comparative effectiveness in lung cancer staging: the impact of test sequencing on outcomes. Chest. 2013;144(6):1776-1782. [CrossRef] [PubMed]
 
Deppen SA, Phillips S, McPheeters M, et al. Benign disease prevalence after surgical lung resection varies geographically in the US Medicare population, implications for lung cancer screening. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; April 6-10, 2013; Washington, DC. Abstract 3628.
 
Detterbeck FC, Gibson CJ. Turning gray: the natural history of lung cancer over time. J Thorac Oncol. 2008;3(7):781-792. [CrossRef] [PubMed]
 
Mehta HJ, Ravenel JG, Shaftman SR, et al. The utility of nodule volume in the context of malignancy prediction for small pulmonary nodules. Chest. 2014;145(3):464-472. [CrossRef] [PubMed]
 
McWilliams A, Tammemagi MC, Mayo JR, et al. Probability of cancer in pulmonary nodules detected on first screening CT. N Engl J Med. 2013;369(10):910-919. [CrossRef] [PubMed]
 
Balekian AA, Silvestri GA, Simkovich SM, et al. Accuracy of clinicians and models for estimating the probability that a pulmonary nodule is malignant. Ann Am Thorac Soc. 2013;10(6):629-635. [CrossRef] [PubMed]
 
Crott R, Versteegh M, Uyl-de-Groot C. An assessment of the external validity of mapping QLQ-C30 to EQ-5D preferences. Qual Life Res. 2013;22(5):1045-1054. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Supporting Data

Online Supplement

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
Invasive Staging of Non-small Cell Lung Cancer*: A Review of the Current Evidence
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543