If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address for correspondence: Subba R. Digumarthy, MD, Division of Thoracic Imaging and Intervention, Massachusetts General Hospital, 55 Fruit St, Founders 202, Boston, MA 02114
Young patients with non–small-cell lung cancer (NSCLC) often present with nonspecific symptoms, mimicking other pathologies.
•
Young patients typically have advanced NSCLC at presentation.
•
NSCLC that arise in young patients frequently have actionable molecular alterations.
•
Imaging features and patterns of metastasis may be driven by these alterations.
Abstract
Background
Non–small-cell lung cancer (NSCLC) in young adult patients is rare, with scarce data available in patients aged < 40 years and even less in those aged < 35 years. Our goal was to determine the presenting symptoms, clinicopathologic characteristics, and imaging features of young patients with NSCLC at time of diagnosis and compare them to those of older adults.
Patients and Methods
We retrospectively analyzed the medical records and imaging of young patients (≤ 40 years old) with NSCLC treated at our institution between 1998 and 2018. Patients < 35 years old were compared to those between 35 and 40 years old. Characteristics of patients ≤ 40 years old were compared to older patients (> 40 years) from publicly available data sets.
Results
We identified 166 young patients with NSCLC (median age, 36.6 years; range, 18-40 years). Most presented with nonspecific respiratory symptoms and were diagnosed with pneumonia (84/136, 62%). Compared to patients < 35 years old, patients 35-40 years old were more likely to have malignancy detected incidentally (15% vs. 5%, P = .04). Patients < 35 years old were more likely to have central tumors (55% vs. 33%, P = .02) and to have bone (38% vs. 19%, P = .007) and lung (39% vs. 24%, P = .03) metastases. Compared to older patients (> 40 years), young patients were more likely to be never smokers (65.0% vs. 14.7%, P < .001) and to have advanced disease (88% vs. 66%, P < .001).
Conclusion
Young patients with NSCLC often present with nonspecific symptoms and have advanced disease at diagnosis, often mimicking other pathologies. Awareness of the clinical presentation and imaging features of NSCLC in young patients may help minimize delays in diagnoses.
Despite significant advances in the diagnosis and management of non–small-cell lung cancer (NSCLC) over the past 2 decades, NSCLC remains the leading cause of cancer-related mortality in the United States and worldwide.
NSCLC typically affects older patients, with most presenting in their 60s or 70s. Less than 5% of patients with NSCLC present before the age of 50, and even fewer patients present before the age of 40.
Although there is no agreed-on strict definition as to what age range is considered a young adult, psychosocial theories of human development conventionally estimate young adulthood to be between ages of 18 and 39,
and for statistical purposes, the American Cancer Society defines cancers in young adults as those that occurring in patients between the ages of 20 and 39.
In young adults aged 40 years or less, cancer is the leading cause of disease-related mortality, with the most common malignancies including lymphoma, melanoma, sarcoma, breast, and testicular cancer.
The diagnosis and treatment of cancer during young adulthood may come with unique health and psychosocial challenges and considerations. Fertility issues, for instance, may be of concern, as many of these patients are in their peak reproductive years.
Financial concerns may also arise, as they are the most likely to be uninsured and are at the greatest risk or unemployment among all other age groups.
Young adults with cancer may be faced with emotional and psychological issues, as the diagnosis comes at a time marked by rapid changes in cognitive and emotional development.
Studies have suggested that NSCLC in the young may have unique clinicopathologic characteristics and outcomes compared to their older counterparts. In particular, studies demonstrate that NSCLC occurring in these younger patients more often have actionable molecular alterations, with the most common being ALK rearrangements and EGFR mutations.
few data exist specifically pertaining to patients aged < 40 years, and within this group, there is significantly higher prevalence over the age of 35 years.
The primary goals of our study were to determine the clinical and imaging features of NSCLC in young patients and to perform subgroup analysis to determine whether differences exist between those aged < 35 years and those aged 35-40 years. We hypothesized that young adult patients have clinical and imaging features that are distinct compared to those of older adults, and that there may also be differences in patients younger than 35.
Patients and Methods
This study was approved by our institutional research review board, with requirement for informed consent waived because of its retrospective nature. The study was performed at a large tertiary-care specialty referral center.
Subject Selection and Data Extraction
We searched multiple databases containing data of patients with resected and advanced lung cancer seen at our institution between January 1998 and December 2018 to identify all lung cancer patients aged ≤ 40 at presentation. The study population was restricted to patients with NSCLC who had available cross-sectional imaging studies at presentation. Those with histology other than NSCLC, such as carcinoid, mucoepidermoid, and small-cell carcinoma, and those with inaccessible baseline imaging studies (irrespective of diagnosis) were excluded (Figure 1). Age, race, sex, smoking history, presenting symptoms, date of diagnosis, and pathology results were extracted from electronic medical records. A subset of cases in the database also had information available regarding molecular testing as part of routine care.
Figure 1CONSORT Diagram Illustrating Patient Selection and Inclusion
At diagnosis, the disease of all patients was staged using computed tomography (CT) of the chest, abdomen, and pelvis. For a subset of patients, staging studies included 18-fluorodeoxyglucose (FDG)-positron emission tomography (PET), CT and/or magnetic resonance imaging of the brain, and nuclear skeletal scintigraphy. The CT scans were performed on various multidetector scanners with automatic exposure control or fixed mA at a tube potential of 100-120 kV. Image reconstruction was performed at a slice thickness of 1 to 2.5 mm for chest and at a maximum thickness of 5 mm for the abdomen and pelvis using standard soft tissue reconstruction kernels. Intravenous iodinated contrast was administered with standard oncology imaging protocol, unless there were contraindications.
Images were reviewed concurrently by a thoracic radiologist with expertise in lung cancer imaging (S.R.D.) and a thoracic imaging fellow (S.G.) on the institutional picture archiving and communication system (Visage 7; ProMedicus, Richmond, Australia). Imaging findings were determined and recorded by consensus. The primary tumor, when identifiable, was analyzed by its bidimensional size, lobar location (right upper lobe, right middle lobe, right lower lobe, left upper lobe, lingula, and left lower lobe), axial location (central vs. peripheral), and density (solid, part solid, ground glass), and was assessed for cavitation, air bronchograms, calcification, and lymphangitic carcinomatosis. Regional metastatic lymph nodes were assessed and classified as N1, N2, or N3 per the International Association for the Study of Lung Cancer (IASLC), 8th edition, of the tumor, node, metastasis classification system (TNM).
The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours.
Distant metastatic lymph nodes were assessed in the neck, thorax (axillary, internal mammary, diaphragmatic/cardiophrenic), abdomen, and pelvis. Lymph nodes measuring 10 mm or greater in short-axis diameter on CT and/or increased radiotracer uptake on FDG-PET imaging were considered pathologic. Metastatic sites outside of lymph nodes were recorded. On the basis of the findings, patients were grouped into those with potentially resectable NSCLC (stage IIIA and below) or advanced/unresectable disease (stage IIIB and above) per IASLC.
Pathology and Molecular Analysis
The diagnosis of NSCLC was made from direct tissue sampling of either the primary tumor or metastasis. Disease was subdivided into adenocarcinoma, squamous-cell carcinoma, adenosquamous carcinoma, undifferentiated carcinoma, or NSCLC not otherwise specified according to a pathologist’s assessment. Assessment for oncogenic driver mutations was determined by fluorescence in situ hybridization, immunohistochemistry, SNaPshot platform (Applied Biosystems; Thermo Fisher Scientific, Waltham, MA), and/or next-generation sequencing. The testing platform used varied depending on the year of presentation.
Comparison to Patients > 40 Years of Age From National Database and Literature
Because the inclusion period covered 2 decades, it was not logically feasible to include all patients > 40 years old from our institute as a control group. For the control group, data from the Surveillance, Epidemiology, and End Results (SEER) Program of 189,727 patients with histologically confirmed NSCLC diagnosed between 2008 and 2012 was used to identify patients aged > 40 years.
National Institutes of Health; National Cancer Institute; Surveillance, Epidemiology, and End Results (SEER) SEER cancer statistics review, 1975-2012. Confirmed cases, 2008-2012. n.d.
This data set included histologic subtypes, patient demographics (age, sex, ethnicity), and disease stage (localized or advanced). Information regarding smoking status and prevalence of oncogenic mutations was not available in the SEER report. The data pertaining to smoking status of those with diagnosed NSCLC (current smokers, past smokers, never smokers) was obtained from a study cohort of National Comprehensive Cancer Network member institutions participating in the NSCLC Database Project between January 2006 and December 2011.
Advanced adenocarcinoma of the lung: comparison of CT characteristics of patients with anaplastic lymphoma kinase gene rearrangement and those with epidermal growth factor receptor mutation.
Two-tailed Student t tests with equal variance were used to compare continuous variables between subsets of patients aged 35-40 years and < 35 years. Two-tailed chi-square tests were used to compare the categorical values. In patients with targetable genetic mutations, the chi-square test was also used to further test the distribution frequency of the mutations. P < .05 was deemed statistically significant. Similar analyses were performed to compare the characteristics of young NSCLC patients to literature reports of patients aged > 40 years. The reference studies used are indicated.
Results
Characteristics of Young Patients With NSCLC
There were 166 patients aged ≤ 40 years in the study, including 102 in 35-40-year and 64 in < 35-year age groups, respectively. Table 1 summarizes the clinicopathologic characteristics of the two groups. Most patients in both groups were white never smokers. The sex distribution in both groups was comparable. The mean age at diagnosis was 30.3 and 38 years for < 35- and 35-40-year age group, respectively. Among current or past smokers, the mean number of pack-years was 9.2 and 16 for the < 35- and 35-40-year age groups (P = .014), respectively. In both groups, adenocarcinoma was the most common histology, and most patients had a targetable oncogenic driver alterations. Twenty-four patients did not undergo molecular assessment for oncogenic driver mutations; they are included in the “absent” group with regard to presence or absence of driver mutations (Table 1). Most patients in both groups (93% and 84%, respectively) had advanced or surgically unresectable disease at the time of initial presentation.
Table 1Clinicopathologic Characteristics of Young Patients With NSCLC and Comparison Between Patients < 35 Years Old and Those Aged 35 to 40 Years
Details of presenting symptoms and duration were available for 136 patients (Table 2). The details for the remainder, most of whom were patients referred for a second opinion, were not clearly documented. Cough (64, 47%) and shortness of breath (36, 27%) were the most common presenting symptoms. A majority of patients (84, 62%) were initially treated for pneumonia and respiratory tract infection (Figure 2). Symptoms related to metastatic disease such as headache, altered mental status, abdominal pain, and back pain were the second most common presentation (29, 21.3%). Cancer was detected incidentally on imaging performed for unrelated causes such as trauma (18, 13.2%), which was performed more commonly in those aged < 35 years (15% vs. 5%, P = .04). The median number of days from initial presentation to tissue diagnosis was 61 days, ranging from 1 day to 1076 days (2.9 years). The longest interval from presentation to tissue diagnosis was for an incidentally detected nodule on CT that enlarged over time.
Table 2Most Common Presenting Symptoms and Time to Diagnosis in 136 Young Patients With NSCLC in Whom Presenting Symptoms Were Reported
Figure 2Biomarkers Analyzed in Patients With Non–Small-Cell Lung Cancer. (A) ROS1-Positive Adenocarcinoma in 31-Year-old Male Never Smoker Initially Treated for Presumed Pneumonia. Computed Tomography (CT) Demonstrates Consolidation in Left Lower Lobe (Arrows) With Multifocal Bilateral Ground-Glass Opacities. (B) ALK-Positive Adenocarcinoma in a 25-Year-old Female Never Smoker Who Presented With Cough and Shortness of Breath. CT Shows a Central Mass in Hilum Extending Across Left Lower Lobe and Lingula (Arrowheads) Associated With Significant Subcarinal and Left Hilar Lymphadenopathy (White Arrows) and Sclerotic Osseous Metastases (Black Arrows). (C) EGFR-Positive Adenocarcinoma in a 28-Year-old Male Never Smoker Presenting With Shortness or Breath and Back Pain. CT Shows a Dominant Mass in Left Lower Lobe (Arrowheads) Associated With Diffuse, Bilateral Miliary-type Lung Metastases. CT and Corresponding Positron Emission Tomography (PET) Demonstrates Multiple Lytic Osseous Metastases (Arrows)
Imaging Features and Metastatic Patterns of NSCLC in the Young
Imaging features of the primary tumor in the < 35- and 35-40-year age groups are summarized in Table 3. Central tumors were more common in the < 35-year than in the 35-40-year age group (55% vs. 33%, P = .02). Otherwise, there was no significant difference in size, lobar or axial distribution, density, and presence of cavity or air bronchogram within the primary tumor. Metastases to the lung (39% vs. 24%, P = .03) and bone (38% vs. 19%, P = .007) were more common in the < 35-year age group (Table 3). While the overall incidence of metastatic disease at presentation was similar between groups aged < 35 and 35-40 years, there were significantly higher rates of lung and bones metastases in the < 35-year age group. The bone metastases were of both lytic and sclerotic type, with latter found in over a third of patients. Although it did not reach statistical significance, the < 35-year age group had a higher incidence of ALK rearrangements, which has an increased association with sclerotic bone metastases.
Table 3Primary Tumor Imaging Features and Patterns of Metastases in Young Patients With NSCLC and Comparison Between Patients < 35 Years Old and Those Aged 35 to 40 Years
NSCLC in the Young Versus Older Patients Aged > 40 Years From National Databases
Comparisons between young patients aged 18-40 years and older patients aged > 40 years are summarized in Table 4. The median age at diagnosis in the young patients was 36.6 years, versus 70 years in the comparison cohort. Numerically, lung cancer in the young subset was more common in female subjects, whereas male subjects predominated in the older cohort, but this difference was not statistically significant. Both groups were primarily made up of white patients. There were significantly more Asians (16.9%) and fewer blacks (4.8%) in the young patient group compared to those aged > 40 years (5.9% and 11%, respectively) (P ≤ .0001). There were more never smokers (65%) in the young group compared to the older group (14.7%) (P ≤ .0001). Current and past smokers accounted for 22.7% and 12% of patients in the young group compared to 50% and 35% of patients in the comparator population. Adenocarcinoma was the overwhelmingly dominant histologic type in the young (86% vs. 51.9% P = .0001), and there were significantly fewer squamous-cell (5% vs. 27.1%) and other histologic (9% vs. 21%) subtypes (P ≤ .0001).
Table 4Clinicopathologic Characteristics of Young and Older Patients With Non–Small-Cell Lung Cancer
Computed tomography imaging characteristics of non–small-cell lung cancer with anaplastic lymphoma kinase rearrangements: a systematic review and meta-analysis.
A majority of young patients had an identifiable oncogenic driver alteration (69.9% vs. 41.5%, P < .0001). ALK rearrangement was the most common alteration (35%), followed by EGFR mutation (23.5%) and ROS1 rearrangement (7.2%). Although it did not reach statistical significance, the < 35-year age group had a higher incidence of ALK rearrangements, while the 35-40-year age group had a higher incidence of EGFR mutations (Tables 1 and 4). In those aged > 40 years, KRAS mutation was prevalent (22.5% vs. 4.2% P ≤ .0001). At presentation, surgically unresectable and metastatic disease was significantly more common in the young cohort (88% vs. 65.7% P ≤ .0001).
Discussion
This is to our knowledge the largest descriptive study of clinicopathologic characteristics and imaging features of NSCLC in young patients aged ≤ 40 years. This is first to study to compare the age groups < 35 and > 35 years, as within the young group there is a significant increase in the incidence of NSCLC in those aged > 35 years. We found that NSCLC in young patients most commonly occurs in never smokers and is characterized by a higher proportion of adenocarcinoma histology and advanced disease at presentation; in addition, it more often harbors targetable genetic alterations. We also found that most patients present with nonspecific respiratory symptoms such as cough and shortness of breath and are often initially diagnosed with and treated for pneumonia. Compared to the < 35-year age group, patients aged 35-40 years diagnosed with NSCLC are more likely to have the primary lung lesions detected incidentally for imaging performed for unrelated causes, such as trauma. Finally, at imaging, NSCLC in those aged < 35 years are more likely to have a central primary tumor and have higher frequencies of bone and lung metastases.
The high prevalence (69.9%) of oncogenic driver alterations that we found among our study population aligns with previously published reports.
ALK rearrangements and EGFR mutations were the most common oncogenic driver alterations in young patients, with more than half driven by ALK or EGFR mutations. This finding must be interpreted in the setting of a high volume of consultations for cancers with these alterations historically performed at our center. The high prevalence of these specific genetic alterations can be explained by the higher proportion of white and Asian never smokers and adenocarcinoma histology associated with these driver mutations.
Although it did not reach statistical significance, the < 35-year age group had a higher incidence of ALK rearrangements, while the 35-40-year age group had a higher incidence of EGFR mutations. This is in contrast to the > 40-year age group, where KRAS mutation is the dominant driver mutation uncovered.
The imaging findings we noted in NSCLC with driver mutations among young patients are consistent with other reported studies of NSCLC with driver mutations.
Computed tomography imaging characteristics of non–small-cell lung cancer with anaplastic lymphoma kinase rearrangements: a systematic review and meta-analysis.
A recently published meta-analysis, for example, described the increased frequency of extensive lymphadenopathy, lymphangitic carcinomatosis, and pleural effusions in patients with ALK-positive NSCLC, as well as the increased likelihood of pleural and pericardial metastases in those patients.
Computed tomography imaging characteristics of non–small-cell lung cancer with anaplastic lymphoma kinase rearrangements: a systematic review and meta-analysis.
In young patients in particular, the imaging findings of ALK-positive NSCLC may be initially misinterpreted as lymphoma due to prominent lymphadenopathy and sclerotic metastases. In contrast, EGFR-mutated NSCLC has been associated with increased propensity for subsolid density of and air bronchograms involving the primary lung tumor and association with miliary-type lung metastases.
The imaging findings of EGFR-mutant NSCLC may be misinterpreted as pneumonia or other inflammatory etiologies.
In fact, most of the patients in our cohort were initially diagnosed with and treated for pneumonia (Figure 2). This is likely because of several factors. First, the symptoms of young patients with NSCLC are nonspecific, typically presenting as cough and shortness of breath. It has been reported that young patients are more likely to be symptomatic at presentation, and it was suggested that this is likely due to the advanced nature of the disease at time of presentation.
Nevertheless, approximately 1 in 8 patients in our cohort had their malignancies incidentally detected by imaging. The exponential rise in the utilization of imaging for various indications may increase incidental detection of lung cancers, further highlighting the need for recognizing the imaging features of lung cancer in young patients. In addition to nonspecific symptoms, the imaging features of NSCLC may also mimic those of infections. This is particularly challenging when air bronchograms are present or when the findings are multifocal (Figure 2). Finally, given the rarity of NSCLC in young patients, the index of suspicion for malignancy by physicians is likely to be low. These factors can contribute to delays in diagnosis and treatment.
Notably, central tumors were more common in < 35-year age group (55%), and peripheral tumors were more common in 35-40-year age group (63%). Otherwise, there was no significant difference in the imaging features of the primary tumor between the two age groups. Peripheral location of the primary tumor in NSCLC has been previously associated with those with underlying alterations in ALK, ROS1, and RET.
However, this does not adequately account for such differences in location found among the groups aged < 35 years and 35-40 years, as there was no significant difference for the incidence of alterations between these two groups. Although the mechanism and significance of this observation is unclear, primary tumor location may influence presenting symptoms and approach to biopsy, surgery, or radiation.
Although the overall incidence of metastatic disease at presentation was similar between groups aged < 35 years and 35-40 years, there were significantly higher rates of lung and bone metastases in the < 35-year age group. The bone metastases were both of lytic and sclerotic type, with the latter found in over a third of patients. It was recently reported sclerotic metastases are associated with ALK and ROS1 rearrangements, which are found in 70% of all NSCLC in this group.
Computed tomography imaging characteristics of non–small-cell lung cancer with anaplastic lymphoma kinase rearrangements: a systematic review and meta-analysis.
For radiologists, the presence sclerotic bone lesions can be a misleading feature, as these are not typical for NSCLC and are more commonly associated with prostate and breast cancer, lymphoma, and carcinoid tumor,
and may also contribute to delay in diagnosis. The increased prevalence of lymphangitic carcinomatosis and lung metastases are also more common with ALK and ROS1 rearrangements
Computed tomography imaging characteristics of non–small-cell lung cancer with anaplastic lymphoma kinase rearrangements: a systematic review and meta-analysis.
and could explain the increased prevalence of these findings in our cohort of young patients with NSCLC.
Our study has several limitations. Because of its retrospective, single-institutional design, selection biases may limit the generalization of the findings. Most notable is the high volume of patients with a targetable mutation seen at our hospital over time. Given the rarity of NSCLC in patients aged ≤ 40 years and the strict criteria for inclusion in the study, the sample size is small. To increase our sample size, we included patients aged > 20 years, but this also resulted in variable molecular testing strategies among the group and precluded evaluation for association with comutations with certain drivers. Nevertheless, this is the largest study to investigate the clinicopathologic and imaging features of NSCLC in patients ≤ 40 years old, and our cohort is 2 to 3 times larger than prior studies.
Consensus, nonindependent review, and recording of the imaging features are also inherent limitations. Finally, the utilization of historical controls for characteristics of NSCLC in older patients (> 40 years) from a compendium of databases and other published research may have introduced additional biases; however, the large cohort sizes of historical controls we utilized minimize this effect.
This is the largest study to assess the clinicopathologic characteristics, imaging features, and metastatic patterns in young patients under 40 with primary NSCLC. Our findings suggest that NSCLC in young patients represents a genetically unique subgroup of the disease, with specific diagnostic and prognostic implications. The imaging features and metastatic patterns are similar to described features of underlying genetic alterations. Young patients with NSCLC often present with nonspecific symptoms and have advanced disease at diagnosis, which can mimic other pathologies. The awareness of the features unique to NSCLC in young patients can reduce delays in diagnosis and treatment.
Clinical Practice Points
•
Non–small-cell lung cancer (NSCLC) in young patients is rare, but disease is typically advanced at the time of diagnosis.
•
The clinical and imaging presentation of NSCLC in young patients can be nonspecific and can be mistaken for other pathologies.
•
Imaging features may be driven by certain targetable oncogenic alterations.
•
Familiarity with these clinical and imaging features and a high index of suspicion for disease in a younger population are necessary to avoid delays in diagnosis.
Disclosure
I.D.J. has received honoraria from Foundation Medicine, CEC, and American Lung Association; consulting fees from Boehringer Ingelheim, AstraZeneca, and Catalyst; and research support from Array, Genentech, Novartis, Pfizer, and Guardant Health. L.V.S. has consulting relationships with AZ, Janssen, Genentech, and Flagship; and has received institutional research support from Novartis, AZ, Boehringer Ingelheim, Genentech, LOXO, and Blueprint Medicines. S.R.D. provides independent image analysis for hospital-contracted clinical research trials programs for Merck, Pfizer, Bristol Myers Squibb, Novartis, Roche, Polaris, Cascadian, AbbVie, Gradalis, Clinical Bay, and Zai Laboratories; has received an honorarium from Siemens Healthineers (not related to work); and has received research funding from Lunit Inc. The other authors have stated that they have no conflict of interest.
The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours.
Advanced adenocarcinoma of the lung: comparison of CT characteristics of patients with anaplastic lymphoma kinase gene rearrangement and those with epidermal growth factor receptor mutation.
Computed tomography imaging characteristics of non–small-cell lung cancer with anaplastic lymphoma kinase rearrangements: a systematic review and meta-analysis.
30314-4&id=gr1.jpg){kind=link}
30314-4&id=gr2.jpg){kind=link}