The rate of incidental secondary tumours in 18-fluorodeoxyglucose positron-emission tomography in current literature—a systematic review and meta-analysis
Original Article

The rate of incidental secondary tumours in 18-fluorodeoxyglucose positron-emission tomography in current literature—a systematic review and meta-analysis

Aydin Mohammadi^, Thomas Havas

Otolaryngology Head and Neck Research Group, Prince of Wales Hospital, Randwick, NSW, Australia

Contributions: (I) Conception and design: A Mohammadi; (II) Administrative support: T Havas; (III) Provision of study materials or patients: A Mohammadi; (IV) Collection and assembly of data: Both authors; (V) Data analysis and interpretation: Both authors; (VI) Manuscript writing: Both authors; (VII) Final approval of manuscript: Both authors.

^ORCID: 0000-0003-4272-2733.

Correspondence to: Aydin Mohammadi, BMed, MS. Otolaryngology Head and Neck Research Group, Prince of Wales Hospital, Randwick, NSW, Australia. Email: Aydin.Mohammadi@health.nsw.gov.au.

Background: 18-Fluorodeoxyglucose (18-FDG) Positron emission tomography (PET) is a commonly used nuclear medical imaging modality utilised in the primary staging of cancer, treatment planning, and treatment response. It is often used for these indications in head and neck squamous cell carcinoma where incidental secondary areas of uptake are rarely discovered. The significance of these often remains controversial and sometimes attributed to normal metabolic activity. The authors sought to determine the incidence of true secondary tumours published in the literature.

Methods: The authors performed a systematic review and meta-analysis of the literature looking at incidental findings on 18FDG-PET scanning in 8 studies comprising 10,068 total scans.

Results: Eight studies were identified which looked at the number of incidental findings on 18FDG-PET scans. A total of 10,068 scans were performed. Of these 487 (4.8%) incidental positive results were identified. One hundred and sixty-six patients were not investigated further due to prognosis, patient wishes, unavailability of results, or loss to follow-up. Of the 321 remaining positive findings, 210 (2.1% overall) were found on biopsy to be true positives of either malignant or benign nature. One hundred and eleven (1.1% overall) false positive results were identified. Meta-analysis demonstrated a 2.29% effect rate (95% CI: 1.07–3.96).

Discussion: The above study demonstrates that a number of patients (210 patients, or 2.1% of all patients studied) with one neoplastic process may indeed have a second primary disease elsewhere. This puts the 18FDG-PET at an advantage in identifying potentially treatment altering disease. The study also demonstrates that 18FDG uptake often felt to be metabolic may indeed represent a second primary and these should be investigated fully by appropriate teams.

Keywords: Second primary disease (SPC); PET/CT; incidental


Received: 21 May 2021; Accepted: 27 September 2021; Published: 31 December 2021.

doi: 10.21037/ajo-21-17


Introduction

18-Fluorodeoxyglucose Positron Emission Tomography (18FDG-PET) scanning is a nuclear medical imaging modality which provides a way to assess the biochemical and metabolic activity of tissues. Tissues which are metabolically active, such as cancer cells, have an increased utilisation of the radiolabelled glucose analogue 18-FDG. The metabolic activity is detected by localisation of annihilation photons (1). There is good evidence that this metabolic activity decreases with appropriate treatment, which may not be congruent with size decreases seen on other modalities such as Computed Tomography (CT) or Magnetic Resonance Imaging (MRI) alone (2). PET can detect cancers when they reach a size of 4–10 mm, and studies have demonstrated a strong correlation between PET estimated and histopathological tumour volumes. This is in contrast to CT which showed significant overestimation of tumour size compared with pathology (3).

Furthermore, previous large database studies including Dong (633,964 patients) which looked at the Swedish Family-Cancer Database, and Ueno (24,498 patients) who analysed the Cancer Institute Hospital database demonstrated that a significant proportion of patients with one primary had a second primary diagnosed during their treatment (4,5). Dong and colleagues found a 8.4% incidence of a second primary cancer in males, and 8.7% in females. The rate was lower in the Ueno group where a 5.2% incidence was demonstrated. The differences are likely explained by the fact that Ueno and colleagues were looking primarily from the point of colorectal cancer whereas the Dong study looked at a more extensive database of patients encompassing all cancers. Nevertheless, there is at least a 1:20 rate of second primary if we are to take the lower number and identifying these at the time of the diagnosis of the first primary with FDG-PET may be a possibility.

We present the following article in accordance with the PRISMA reporting checklist (available at https://dx.doi.org/10.21037/ajo-21-17).

Objectives

The authors were interested in what the published rate of unknown synchronous cancers found on PET scanning was in adults who underwent imaging for cancer staging. The goals of this systematic review were to assess: the rate of unknown synchronous disease found on PET, and whether various PET protocols changed the rates of these findings.


Methods

A systematic review and meta-analysis were performed in multiple databases (CENTRAL, MEDLINE, EMBASE, PsychINFO, and CINAHL) between January and March of 2021. The literature search was performed by the two authors independently and blinded to one another. Where there was discrepancy in the inclusion or exclusion of an article, the full article was accessed to allow a final decision to be made. The primary author extracted the data and this was reviewed by the second author for veracity.

The dataset extracted from the selected papers included fasting time, FDG dose, post-dose waiting time, bed positions, acquisition time per bed position, resolution, field of view (FOV), pixel size, and scanner type used. Where data was lacking or not specifically provided, the authors were contacted for clarification.

The authors adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist. A grey literature search was performed in Google. Non-English papers were included and authors were contacted for clarification and further information where required. The details of the search criteria are given in the flowchart below (Figure 1). Studies were excluded when they included single case reports, or small case series less than 10 patients, as well as when the study did not show relevance to the search terms or the study. Duplicates were removed during the process of screening abstracts. Of 58 identified relevant papers, 8 studies were included in the study which looked at the criteria set out by the authors.

Figure 1 Literature review flowchart and search strategy.

Statistical analysis

All descriptive statistics were performed using Microsoft’s Excel (Microsoft Co., USA). The data is provided as Incidence Rate, False Positive Rate, and True Positive Rate. Meta-analysis of the dataset was performed using a commercially available version of MedCalc (MedCalc Software Ltd., Belgium).

A meta-analysis for proportions was performed using the MedCalc statistical software package. The results of this are given in Tables 1 and 2, and graphically demonstrated in the Forest plot in Figure 2. The results of a Q-test indicate the presence of heterogeneity between the studies. The I2 index supports this proposition with a finding of significant heterogeneity between the studies which required a random effects model to be used for the meta-analysis.

Table 1

Meta-analysis of proportions

Study Sample size Proportion (%) 95% CI Weight (%)
Fixed Random
Ozkol et al. (6) 2,370 2.911 2.272 to 3.670 23.53 13.07
Williams et al. (7) 609 0.657 0.179 to 1.673 6.05 12.26
Fuertes et al. (8) 2,290 0.83 0.500 to 1.293 22.74 13.06
Oozeer et al. (9) 299 3.01 1.385 to 5.637 2.98 11.29
Garcheva et al. (10) 1,408 0.426 0.157 to 0.925 13.98 12.87
Malik et al. (11) 591 4.23 2.756 to 6.181 5.88 12.23
Strobel et al. (12) 589 9.508 7.262 to 12.169 5.86 12.23
Ishimori et al. (13) 1,912 1.151 0.722 to 1.737 18.99 13
Total (fixed effects) 10,068 1.754 1.507 to 2.030 100 100
Total (random effects) 10,068 2.298 1.077 to 3.961 100 100

Table 2

Tests for heterogeneity

Test for heterogeneity Static value
Q 157.4717
DF 7
Significance level P<0.0001
I2 (inconsistency) 95.55%
95% CI for I2 93.16 to 97.11
Figure 2 Forest plot of meta-analysis of proportions.

Subgroup analysis was performed on the proportions of patients who were lost to follow-up. Analysis of the scan acquisition protocols did not demonstrate a change to the rate of incidental findings.


Results

A total of 8 studies were selected for inclusion in the study which equated to a total of 10,068 PET scans performed in eight separate institutions. Of these, 487 incidental findings were reported. Two hundred and ten true positive pathologies were demonstrated on histopathology on further investigation of these incidental findings. This equates to a true positive rate of 2.1% and a false positive rate of only 1.1%.

Table 3 shows the breakdown of studies analysed including false positive and negative rates. There was significant variability in the rates of true and false positives as well as significant differences in the number of patients lost to follow-up. The explanation for this is not entirely clear but the studies do hint at disease factors in particular, such as untreatable disease, or multiple comorbidities as causes for non-investigation. The significance of a large number of patients lost to follow-up is difficult to analyse. The literature supports a cut-off of <5% loss to follow-up as leading to little bias, whereas a loss to follow-up of >20% leads to significant bias (14). All bar two of the included studies had a greater than 5% loss to follow-up with up to 63% loss to follow-up in one study. This is a significant source of bias which would affect the overall rates of true incidental positives. If all patients lost to follow-up are assumed to have true positive incidental findings, as in a worst-case-scenario, the rate would rise to 3.7% (15).

Table 3

Breakdown of total scans, incidental positive findings, true and false negatives, and patients not investigated or lost to follow-up

Author Number of scans Incidental positives True positives False positives Not investigated
Ozkol et al. (6) 2,370 121 69 (2.9%) 5 47
Williams et al. (7) 609 76 4 (0.7%) 24 48
Fuertes et al. (8) 2,290 27 19 (0.8%) 8 0
Oozeer et al. (9) 299 40 9 (3.0%) 20 11
Garcheva et al. (10) 1,408 11 6 (0.4%) 0 5
Malik et al. (11) 591 64 25 (4.2%) 31 8
Strobel et al. (12) 589 69 56 (9.5%) 13 0
Ishimori et al. (13) 1,912 79 22 (1.2%) 10 47
Totals 10,068 487 210 111 166

Table 4 demonstrates the areas of incidental uptake for the studies with most lesions being colorectal in origin (N=66), lung (N=42), and head and neck (N=24) in descending order. The different methods of data acquisition are compared in Table 5 which includes the bed/cradle position time, acquisition time, and resolution where this data was provided. Where the data was not provided, all attempts were made to contact the author listed in the publication. The known scan acquisition protocols were similar between studies that reported them, with the main differences being scan acquisition time and FDG dose. These differences do not appear to influence the rate of incidental findings.

Table 4

Breakdown of sites of individual true positive lesions

Site of second primary Ozkol Williams Ishimori Fuertes Oozeer Malik Garcheva Strobel Totals
Colorectal 16 4 23 17 1 5 66
Thyroid 8 1 6 2 17
Lung 7 2 7 26 42
Kidney 6 1 7
Prostate gland 6 6
Ovarian 6 1 7
Stomach 4 1 1 6
Uterus and cervix 4 4
Head and neck 2 1 2 4 15 24
Salivary glands 2 1 3
Oesophagus 2 2 5 9
Pancreas 2 1 3
Breast 2 2 1 5
Spleen 2 2
Muscle 1 1
Thymus 1 1 2
Adrenal gland 1 1
Pituitary gland 1 1
Lymphatic nodule 1 1 1 1 4
Brain 1 1
Site not reported 11 11

Table 5

Data acquisition protocols

Study Fasting time FDG dose Post-dose waiting time Bed positions Acquisition time per bed position Resolution FOV Pixel size Scanner
Ozkol et al. (6) 6 hours 5 MBq/kg 45–90 min NR 3 minutes NR NR NR Siemens Biograph 6 LSO
Williams et al. (7) NR NR NR NR NR NR NR NR NR
Fuertes et al. (8) 6 hours 2.4–2.7 MBq/kg 60 min NR NR NR NR NR GE Discovery ST & Siemens Exact HR+
Oozeer et al. (9) NR NR NR NR NR NR NR NR NR
Garcheva et al. (10) NR NR NR NR NR NR NR NR NR
Malik et al. (11) 6 hours 350–400 MBq 55–92 min NR 2.5 minutes 6.5 mm NR 3.91 mm GE Discovery ST & GE Advance
Strobel et al. (12) 4 hours 350 MBq NR 8 to 9 3 minutes NR NR NR GE Discovery LS
Ishimori et al. (13) 4 hours 8.14 MBq/kg 60 min 6 or more 5 minutes 4.5 mm 50 cm 3.91 mm GE Discovery LS

Due to the significant heterogeneity demonstrated in the subgroups, a random effects model was used for the meta-analysis which supports a weighted finding of an effect size of 2.29% (95% CI: 1.08–3.96).

One hundred and sixty-six patients were lost to follow-up or did not have complete investigation of the incidental findings. This number represents 34% of total incidental findings and the significance is that the final number of true positives could be underestimated. When the number of patients lost to follow-up are analysed as a subgroup, and the individual false positive rates are applied, the predicted true positive rate increases substantially. A further subgroup meta-analysis of proportions demonstrates a new effect size of 3.22% (95% CI: 1.68–5.24). This would be equivalent to 1 in 30 patients having a missed diagnosis of a second primary disease, were it not for PET/CT staging.


Discussion

The results demonstrate that a significant number of patients with a single neoplastic process may have a second incidental finding which may not be clinically apparent. The importance of positron emission tomography in these patients is to highlight secondary disease which may otherwise go unchecked, potentially leading to treatment failure from unsuspected secondaries, second primary disease progression due to non-treatment, or unnecessary treatment of a primary disease in patients who may have an untreatable second disease.

In Australia, the fee for a whole body FDG PET study is AU$953.00 (16). This is arguably a negligible number of health Dollars for what could significantly alter the course of the patient journey. This is a journey which can often include multiple admissions to hospital, numerous investigations, operative interventions, as well as chemotherapy and/or radiotherapy treatments, the cost of which can and often does run into the tens of thousands of dollars. A dedicated cost-benefit analysis would be required to give an objective breakdown of any cost savings; however, it is at most a modest increase in the overall cost of treatment.

The average radiation dose from an FDG PET scan is estimated to be 12.2 millisieverts (mSv) (17). A whole body CT scan is estimated to impose a radiation dose between 4-24 mSv, compared to 0.04 mSv for a chest X-ray. There is an estimated increase in the possibility of cancer of 1 in 2,000 for each 10mSv radiation exposure (18). This is a considerably smaller dose of radiation than would be expected to be given during the course of cancer treatment. However, treatment doses are delivered in an exact manner and imaging should be performed judiciously to prevent excess radiation exposure.

The age-standardised incidence of colorectal malignancies in Australia as measured in 2013 is 58 cases per 100,000 persons (19). The equivalent incidence for this study would be in the order of over 650 cases per 100,000 persons. The significantly large proportion of colorectal incidental findings is worrying. Given that this is not a typical area of investigation for head and neck cancer secondaries, an incidental primary would not necessarily be picked up in this population. This is yet another reason to consider performing FDG PET as a first line staging investigation. Furthermore, with a false positive rate of 1.1%, FDG PET provides a modality which does not pose a large risk of forcing patients to undergo unnecessary examination or investigation of positive results.

The authors recognise the limitations of attempting to perform a large scale, multicentre investigation of all incidental FDG PET findings such as the logistics of performing the appropriate histopathological investigations, performing intention-to-treat analyses on patients lost to follow-up, as well as the appropriate referral pathways for these second primaries. Despite this, such a study would allow the accurate delineation of the true positive incidental rates and what the longer-term follow-up data on these patients is. This would in turn allow for improved investigation and treatment of all cancer patients. The authors support the pursuit of a large-scale, multicentre prospective study to investigate this further.


Acknowledgments

The views expressed within this research paper are those of the authors and do not reflect the official position of the institution.

Funding: None.


Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://dx.doi.org/10.21037/ajo-21-17

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://dx.doi.org/10.21037/ajo-21-17). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.


References

  1. Wassef HR, Hanna N, Colletti P. PET/CT in Head-neck Malignancies: The Implications for Personalized Clinical Practice. PET Clin 2016;11:219-32. [Crossref] [PubMed]
  2. Wahl RL, Jacene H, Kasamon Y, et al. From RECIST to PERCIST: Evolving Considerations for PET response criteria in solid tumors. J Nucl Med 2009;50:122S-50S. [Crossref] [PubMed]
  3. Cheebsumon P, Boellaard R, de Ruysscher D, et al. Assessment of tumour size in PET/CT lung cancer studies: PET- and CT-based methods compared to pathology. EJNMMI Res 2012;2:56. [Crossref] [PubMed]
  4. Dong C, Hemminki K. Second primary neoplasms in 633,964 cancer patients in Sweden, 1958-1996. Int J Cancer 2001;93:155-61. [Crossref] [PubMed]
  5. Ueno M, Muto T, Oya M, et al. Multiple primary cancer: an experience at the Cancer Institute Hospital with special reference to colorectal cancer. Int J Clin Oncol 2003;8:162-7. [Crossref] [PubMed]
  6. Ozkol V, Alper E, Aydin N, et al. The clinical value of incidental 18F-fluorodeoxyglucose-avid foci detected on positron emission tomography/computed tomography. Nucl Med Commun 2010;31:128-36. [Crossref] [PubMed]
  7. Williams SP, Kinshuck AJ, Williams C, et al. Incidental head and neck findings on 18F-fluoro-deoxy-glucose positron emission tomography computed tomography. J Laryngol Otol 2015;129:898-902. [Crossref] [PubMed]
  8. Fuertes J, Montagut C, Bullich S, et al. Incidental focal uptake in colorectal location on oncologic 18FDG PET and PET/CT studies: histopathological findings and clinical significances. Rev Esp Med Nucl Imagen Mol 2015;34:95-101. [Crossref] [PubMed]
  9. Oozeer NB, Mackenzie K. Outcome of FDG-Avid Areas on PET-CT Scans Organized for Investigation of Head and Neck Malignancy. Otolaryngol-Head Neck Surg. 2014;151:174-5. [Crossref]
  10. Garcheva M, Zlatareva D, Gocheva L. Positron emission tomography combined with computed tomography for diagnosis of synchronous tumors. Klin Onkol 2014;27:283-6. [Crossref] [PubMed]
  11. Malik V, Johnston C, Donohoe C, et al. (18)F-FDG PET-detected synchronous primary neoplasms in the staging of esophageal cancer: incidence, cost, and impact on management. Clin Nucl Med 2012;37:1152-58. [Crossref] [PubMed]
  12. Strobel K, Haerle SK, Stoeckli SJ, et al. Head and neck squamous cell carcinoma (HNSCC)--detection of synchronous primaries with (18)F-FDG-PET/CT. Eur J Nucl Med Mol Imaging 2009;36:919-27. [Crossref] [PubMed]
  13. Ishimori T, Patel PV, Wahl RL. Detection of unexpected additional primary malignancies with PET/CT. J Nucl Med 2005;46:752-7. [PubMed]
  14. Straus SE. Evidence-Based Medicine: How to Practice and Teach It (Straus, Evidence-Based Medicine). 4th edition. Churchill Livingstone, 2010.
  15. Dettori JR. Loss to follow-up. Evid Based Spine Care J 2011;2:7-10. [PubMed]
  16. Australian Government Department of Health. Medicare Benefits Schedule Online. Published January 14, 2002. Accessed July 12, 2017. Available online: http://www9.health.gov.au/mbs/fullDisplay.cfm?type=item&q=61610
  17. Kwon HW, Kim JP, Lee HJ, et al. Radiation Dose from Whole-Body F-18 Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography: Nationwide Survey in Korea. J Korean Med Sci 2016;31:S69-74. [Crossref] [PubMed]
  18. New South Wales Environmental Protection Agency. Information on Whole Body Scanning. Published June 5, 2013. Accessed July 12, 2017. Available online: http://www.epa.nsw.gov.au/resources/radiation/ctbodyscan.pdf
  19. Australia Institute of Health and Welfare. Colorectal cancer (bowel cancer). Published February 1, 2017. Accessed July 12, 2017. Available online: http://www.aihw.gov.au/cancer/colorectal/
doi: 10.21037/ajo-21-17
Cite this article as: Mohammadi A, Havas T. The rate of incidental secondary tumours in 18-fluorodeoxyglucose positron-emission tomography in current literature—a systematic review and meta-analysis. Aust J Otolaryngol 2021;4:40.

Download Citation