Patients undergoing multiple ¹⁸F-FDG PET/CT exams: Assessment of frequency, dose and disease classification

Abstract

Objective:

To analyse the frequency, demographics, primary disease and cumulative effective dose of patients undergoing two or more ¹⁸F-FDG PET/CT examinations in a year.

Methods:

In a retrospective study performed at a tertiary-care hospital, patients who underwent ≥2 ¹⁸F-FDG PET/CT scans in a calendar year were identified for two consecutive years. The CT radiation dose was calculated using dose-length-product and sex-specific conversion factors. The primary malignancy of patients was retrieved from electronic medical records.

Results:

10,714 ¹⁸F-FDG PET/CT exams were performed for 6,831 unique patients in 2 years, yielding an average of 1.6 exams per patient. The maximum number of ¹⁸F-FDG PET/CT examinations any patient underwent in a single year was seven. 20.9% patients had ≥2 ¹⁸F-FDG PET/CT exams in any single year. Thirty nine percent patients in the cohort were below 60 years age. The median dose for ¹⁸F-FDG PET/CT examination was 25.1 mSv and maximum value reaching 1.7 to 2.9 times the median value. Cumulative effective dose (CED) was≥100 mSv in 12–13% of the patients. The cumulative dose for both years combined demonstrated the 25^th percentile, 50^th percentile and 75^th percentile as well as the mean to be over 100 mSv, with the 25^th percentile being 109 mSv. The dominant primary malignancies contributing to serial ¹⁸F-FDG PET/CTs in decreasing frequency were melanoma, non-Hodgkin’s lymphoma (NHL), gastrointestinal cancer, breast cancer and Hodgkin’s lymphoma.

Conclusions:

A sizeable number of patients undergo≥2 ¹⁸F-FDG PET/CT exams with one out of every eight patients receiving cumulative dose≥100 mSv and that includes patients with long-life expectancy.

Advances in knowledge:

The study found that one of five patients had≥2 ¹⁸F-FDG PET/CT exams in a calendar year, one of four patients in two years and one of eight patients received cumulative dose≥100 mSv. Top malignancies associated with serial imaging in decreasing order of frequency included melanoma, non-Hodgkin’s lymphoma (NHL), gastrointestinal cancer, breast cancer and Hodgkin’s lymphoma.

Introduction

Positron-emission tomography (PET) typically combined with computed tomography (CT) is a well-established imaging modality with proven applications in many clinical situations, dominantly in oncology. Nearly 2 million PET/CT examinations are performed annually in USA, ¹ While the benefits to patients from PET/CT exams are well known, the concern over relatively higher radiation doses has been raised. ² Moreover, the frequency of multiple PET/CT exams is not well established. ³ It has been demonstrated by case studies in a group of 33 hospitals in Helsinki region in Finland where tracking of patients’ imaging examinations and radiation doses were implemented that patient-specific justification and optimization becomes possible using the tracking of radiologic procedures and radiation dose of individual patients. ⁴ A number of studies in recent years have highlighted the importance of assessing the frequency of imaging exams as they have provided useful information for strengthening measures for implementing rational practice and achieve patient safety. ^5–7 There could be many reasons for lack of information on the frequency of PET/CT imaging, with prominent factors including lack of permanent medical identification number (ID) for a patient across different healthcare systems (which is more pertinent in USA), ⁸ lack of wider availability of systems that can track all radiology exams for a patient, lack of information on radiation dose associated with PET in the electronic medical records of the patient (somewhat similar to CT, where dose estimations are easily available in the dose management systems and measured dose quantities are incorporated in the output of the machine) ⁹ and finally lack of motivation to optimize the radiation safety profile secondary to common perceptions such as oncology patients get a lot of radiation any way and the additional dose from PET/CT exams should not be of much consequence. ^5,10 On the last point, it is important to realize that oncology patients get higher doses targeted to cancer tissues, not to rest of the body where every effort must be made to optimize radiation dose in manner similar to non-oncology patients. With advances in medicine, many oncology patients now have long survival exceeding decade(s) and a sizable fraction do get cured. ¹¹ In this study on oncology patients receiving serial ¹⁸F-FDG PET/CT examinations, we have assessed the exam frequency, effective dose and the primary disease oncologic diagnosis.

Methods and materials

This retrospective study requiring no interaction or intervention with patients was approved by the institutional review board (IRB) and the requirement to obtain informed consent was waived.

A retrospective analysis of the radiation dose resulting from ¹⁸F-FDG PET/CT examinations conducted at a tertiary-care academic medical centre over a two-year period between 1 January 2019 and 31 December 2020 was performed. The search criteria utilized to identify examinations for inclusion were date of the examination (1 January 2019 to 31 December 2020) and institution-specific exam code assigned to ¹⁸F-FDG PET/CT at our hospital (PT.NM.PETCT) for adult patients. This search strategy identified all PET examinations, including those performed with low-dose CT for attenuation correction only as well as those performed with diagnostic CT. The search results were exported from the internally developed institution specific radiology report and image search application into an encrypted, password-protected excel spreadsheet, which was stored on an encrypted, password-protected drive and accessed using computers located within the radiology department.

For each ¹⁸FDG PET/CT examination identified during the search, the results that were extracted from radiology report included examination accession number, date of the examination, medical record number, age and gender of the patient as well as PET/CT report. Five PET/CT machines (two Siemens Biograph, two Siemens MCT Flow, one GE Discovery STE) were utilized to perform the examinations during the study period. Each system had CT dose reduction techniques (CareDose4D, Auto mA) and two MCT flow systems also had more advanced techniques (CarekV, Admire). The typical protocol for ¹⁸FDG PET/CT at our institution for clinical imaging is base of skull to mid-thigh imaging with either diagnostic CT or low-dose attenuation correction CT. The weight-based dosing is practiced as a routine and is in line with practice guideline. ¹² For adults, the protocol has dose bands based for BMI ≤ 30.9; 31–44.9 and ≥45. The dosing chart for Siemens mCT flow involves almost half the administered activity as compared to Discovery and Biograph. The radioactive dose of FDG administered in mCi (later converted into MBq) during each PET examination was recorded and retrieved from the radiology report. Typical imaging starts 60 ± 10 min after injection.

The sorting function in Microsoft Excel was utilised to sort the examinations identified for inclusion in this study based on the patient’s numerical medical record number. Next, the subset of patients who received greater than two ¹⁸F-FDG PET/CTs in any given calendar year during the two-year study period were identified using excel algorithms. Additional information about this subset of patients was extracted from radiology report including the age of the patient (age at the time of the first ¹⁸F-FDG PET/CT), the race/ethnicity of the patient (if available), patient gender, total number of ¹⁸F-FDG-PET/CTs received within the two-year study period, number of ¹⁸F-FDG PET/CT examinations in 2019 (between 1 Jan 2019 and 31 Dec 2019), number of ¹⁸F-FDG PET/CT examinations in 2020 (between Jan 1, 2020 and Dec 31, 2020) and the type of disease or active malignancy undergoing treatment or surveillance. Patients receiving ≥2 ¹⁸F-FDG PET/CT examinations in any calendar year (either 2019 or 2020) were included in the study.

Effective dose (E) resulting from radioactive F-18 fluorodeoxyglucose (FDG) administration during the PET portion of the examination was estimated using dose coefficient of 0.70 mSv/mCi (0.019 mSv/MBq) for F-18 FDG as reported in the International Commission on Radiological Protection (ICRP)’s publication on “Radiation Dose to Patients with Commonly Used Radiopharmaceuticals”. ¹³

CT examination data were extracted from a commercial dose monitoring platform (Radimetrics v. 2.9; Bayer Healthcare, Whippany, NJ). For each acquisition series, effective dose was calculated using dose length product (DLP) and a sex-specific conversion factor. For a scan from the base of skull to mid-thigh, the conversion factor was 0.012 and 0.014 mSv/(mGy∙cm) for male and female patients, respectively. ¹⁴ For other anatomic regions, we used the conversion factors in ICRP Publication 102 (Table A.2). ¹⁵

Summing the E from FDG and CT part resulted in total E for the combined ¹⁸F-FDG PET/CT examination. Summing E for an individual patient from ¹⁸F-FDG PET/CTs done at different times provided cumulative effective dose (CED) for each patient.

The primary disease being serially evaluated through multiple ¹⁸F-FDG PET/CT examinations was obtained from reviewing the electronic medical records. These data were compiled to quantify the frequency with which various categories of malignancies were represented in our patient cohort.

Results

A total of 6831 patients underwent 10,714 ¹⁸F-FDG PET/CT examinations between 1 January 2019 and 31 December 2020, yielding an average 1.6 examinations per patient. 5,333 ¹⁸F-FDG PET/CT examinations were performed in 2019 and 5381 ¹⁸FDG PET/CT examinations were performed in 2020, despite the COVID-19 pandemic. 1428 patients received ≥2 ¹⁸F-FDG PET/CT examinations during the two-year study period, including 834 patients in 2019, 825 patients in 2020, and 231 patients in both years.

The demographics of the patients who received 2 or more ¹⁸F-FDG PET/CT examinations during the study period (either in 2019 or 2020) are detailed in Table 1. Of the patients receiving 2 or more ¹⁸F-FDG PET/CT examinations, about half of the patients were between the ages of 61 and 80 and nearly a quarter of the patients were between the ages of 41 and 60. The study cohort included 98 more male patients compared to female patients, with male patients contributing to 53.4% and female patients contributing to 46.6% of the cohort.

Table 1.

Age and sex distribution of patients receiving two or more PET/CTs in a year

Age (in years)	Number	%
≤40	166	11.6%
41 to ≤60	382	26.8%
61 to ≤80	743	52.0%
≥81	137	9.6%
Total	1428	100%
Gender
Female	665	46.6%
Male	763	53.4%

The frequency with which ¹⁸F-FDG PET/CT examinations were performed is summarised in Table 2. The maximum number of ¹⁸F-FDG PET/CT examinations any patient underwent in a year was in 6 in 2019 and 7 in 2020. One patient underwent 12 ¹⁸F-FDG PET/CT examinations over the two-year study period, the maximum frequency observed in our cohort. A total of five patients underwent 10 or more ¹⁸F-FDG PET/CT examinations cumulatively over the two-year study period. Since the study cohort only included those patients who underwent 2 or more ¹⁸F-FDG PET/CT in a single year (either 2019 or 2020), the least number of ¹⁸F-FDG PET/CT exams in our cohort was 4, and thus no data are provided in last column of Table 2 against first two rows.

Table 2.

Number of patients with ≥2 PET/CT examinations in 2019, 2020 and 2019 as well as 2020 combined

Number of PET/CT exams	2019	2020	2019 and 2020 combined
2	530 (63.5%)	533 (64.6%)	*
3	197 (23.6%)	179 (21.7%)	*
4	75 (9.0%)	70 (8.5%)	79 (34.2%)
5	24 (2.9%)	28 (3.4%)	48 (20.8%)
6	8 (1.0%)	13 (1.6%)	46 (19.9%)
7	0	2 (0.24%)	25 (10.8%)
8–9	0	0	28 (12.1%)
10–12	0	0	5 (2.2%)
Total	834	825	231

For all ¹⁸F-FDG PET/CT examinations (including the patients receiving only one exam), the average administered activity of FDG was 15.5 ± 3.0 mCi (575 ± 111 MBq).

The information on dose per exam for PET, CT and combined ¹⁸F-FDG PET/CT part is presented in Table 3, split into different quartiles (median, mean ± SD, 75^th percentile with minimum and maximum dose). The median dose was 10.5 mSv for PET, 14.5 mSv for CT and 25.1 mSv for ¹⁸F-FDG PET/CT, with the maximum value reaching up to 1.7–2.9 times the median value, as can be expected from dose variation related to patient body habitus. Please note that the low dose values presented under the column labelled “Minimum” in Table 3 represent the doses associated with attenuation correction CT.

Table 3.

Effective dose (mSv) per PET/CT examination for 2019, 2020 and both years combined split by PET, CT and combined examination dose

Year	No. of patients	Exam	Minimum	25th	Median	Mean ± SD	75th	Maximum
2019	834	PET	2.3	10.46	10.52	11.3 ± 1.6	12.2	17.7
		CT	0.2	10.0	13.9	14.9 ± 7.2	18.4	61.1
		PET/CT	3.9	20.8	24.7	26.3 ± 8.2	30.1	75.0
2020	825	PET	4.2	9.2	10.5	10.3 ± 1.7	10.7	18.1
		CT	0.4	9.3	13.7	14.7 ± 7.4	19.1	46.2
		PET/CT	6.7	19.4	23.8	25.0 ± 8.2	29.6	58.1
2019 plus 2020	231	PET	7.0	10.0	10.5	11.0 ± 1.5	11.6	17.8
		CT	0.3	10.6	14.5	16.0 ± 7.5	20.3	41.8
		PET/CT	7.3	20.8	25.1	27.0 ± 8.3	32.3	54.4

Table 4 presents CED for the PET portion, the CT portion and combined ¹⁸F-FDG PET/CT for 2019, 2020 as well as 2019 and 2020 combined. The last column of Table 4 presents the number and percentage of patients receiving CED ≥ 100 mSv.

Table 4.

Cumulative effective dose (CED) in mSv per patient (who received two or greater PET/CT examinations) in 2019, 2020 and over the two-year study period

Year	No. of patients	Exam	Minimum	25th	Median	Mean ± SD	75th	Maximum	No. of ≥100 mSv
2019	834	PET	6	21	27.2	28.7 ± 9.9	31.6	70.6	0
		CT	0.4	23.5	33.4	37.9 ± 22.1	46.9	166	18 (2.2%)
		PET/CT	10.9	45.9	59.1	66.6 ± 29.8	79.7	228	102 (12.2%)
2020	825	PET	8.3	20.7	21.5	26.5 ± 10.7	31.2	83.7	0
		CT	0.8	21.1	32.8	38.1 ± 25.2	47.1	191	23 (2.8%)
		PET/CT	13.4	42.4	55.8	64.6 ± 33.4	75.7	248	103 (12.5%)
2019 plus 2020	231	PET	28.1	48.6	58.5	61.2 ± 19	70.5	130	10 (0.7% of 1428)
		CT	1.2	55.5	79.2	89 ± 48.7	107	279	72 (5.0% of 1428)
		PET/CT	29.3	109	135	150 ± 62.2	178	409	184 (12.9% of 1428)

None of the patients in our cohort received CED ≥ 100 mSv from PET alone in any single year. 0.7% of the patients received ≥100 mSv combined over the two-year study period. In comparison, 2–3% of the patients received ≥100 mSv from CT in a single year (2.2% in 2019 and 2.8% in 2020) and 5% of the patients received ≥100 mSv from CT combined over the two-year study period. 12.2%, 12.5% and 12.9% of the patients received ≥100 mSv from combined ¹⁸F-FDG PET/CT in 2019, 2020 as well as over the two-year study period respectively. It is interesting to note that all quartile values (namely 25^th, 50^th, 75^th) as well as the mean value are over 100 mSv for both years combined, with 25^th percentile itself being 109 mSv and 75^th percentile being 178 mSv. This quantitative statistic is a good indication of the calibre of the doses involved.

Figure 1A is a graphical representation of the dose data presented in Table 3 and Figure 1B is a graphical representation of the cumulative effective dose data presented in Table 4.

Figure 1.

(a) Box-and-whisker plot showing effective dose (in logarithmic scale) for patients receiving greater than 2 PET/CT in a calendar year split by PET, CT and PET/CT for 2019, 2020 and combined two-year study period. Whiskers denote 10th and 95th percentiles; box extremes indicate 25th and 75^th percentiles, and centre bar indicates 50th percentile, or median value. (b) Box-and-whisker plot showing CED (in logarithmic scale) for patients receiving greater than 2 PET/CT in a calendar-year split by PET, CT and PET/CT for 2019, 2020 and combined two-year study period. Whiskers denote 10th and 95th percentiles; box extremes indicate 25th and 75th percentiles, and centre bar indicates 50th percentile, or median value.

Figure 2 illustrates the frequency with which various primary malignancies were represented in our cohort and the frequency with which they received 2 or more ¹⁸F-FDG PET/CT exams over the study period resulting in CED of equal to or greater than 100 mSv. The primary diseases in decreasing order of occurrence are melanoma (including head and neck melanoma), non-Hodgkin’s lymphoma (including diffuse large B-cell lymphoma and T-cell lymphoma), gastrointestinal cancer, breast cancer and Hodgkin’s lymphoma.

Figure 2.

Frequency of various primary malignancies in patients receiving CED over 100 mSv. NHL, Non-Hodgkin’s lymphoma; HL, Hodgkin’s lymphoma.

Abbreviations: GI: Gastrointestinal; GU: Genito-urinary; MSK: Musculo-skeletal; Cutaneous SCC: cutaneous squamous cell carcinoma

Discussion

Our retrospective analysis including 6831 patients undergoing serial ¹⁸F-FDG PET/CT imaging at a large tertiary-care academic hospital in USA identified that 20.7 to 22.5% of the patients received 2 or more ¹⁸F-FDG PET/CT exams in a single calendar year. 25.7% of the patients in our cohort received 2 or more ¹⁸F-FDG PET/CT examinations during the study period. Given that an average of 1.6 examinations were performed per patient, any patient receiving one ¹⁸F-FDG PET/CT was more likely than not to receive a second follow-up ¹⁸F-FDG PET/CT. About a quarter of the patients in this cohort were between the ages of 41 and 60 and nearly 12% were under 40 years of age. For the patients presenting with diseases with longer life expectancy, the consideration of radiation risk requires strengthening the optimization actions. The maximum number of ¹⁸F-FDG PET/CT examinations any patient underwent in a single year was 7. However, over the two-year study period, there were 33 patients who underwent 8–12 ¹⁸F-FDG PET/CT exams. This observation may be further studied in future multicentric research, to assess if similar findings are observed at other institutions.

The median dose was 10.5 mSv for PET portion, 14.5 mSv for CT portion and 25.1 mSv for ¹⁸F-FDG PET/CT, with maximum value ranging from 1.7 to 2.9 times the median value, something that has not been reported earlier in terms of magnitude of variation. Patients with higher body mass index (BMI) typically will have high doses and may fall in upper part of the above range. The Health Survey for England 2019 estimates that 28% of adults in England are obese and prevalence of obesity in the United States in 2017–2018 was reported to be 42.4%. ^16,17 Thus, high doses may be expected in obese patients.

It is worthwhile to note that despite many advances in CT technology ^18,19 the dose contribution from the CT portion of the examination significantly outweighed the dose contribution from the PET portion of the examination. With continued development in technology both in PET and CT imaging, it will be interesting to see how these figures change in coming years as the technology provides the greatest hope. ^19,20

Not that 100 mSv provides a yardstick for radiation effects, but as specified in recent publications, at the level of 100 mSv of effective dose, many organs may get over 100 mGy dose, ^5,21–24 a level at which radiation risks are more likely particularly if received in a single exposure. The recent study assessing organ doses in cohort of patients who received CED of 100 mSv through serial CT imaging showed organ doses over 100 mGy for most organs and for some organs ≥ 200 mGy with unignorable associated lifetime attributable cancer mortality rates. ²⁴

The percent of patients receiving ≥100 mSv from ¹⁸F-FDG PET/CT was 12 to 13%, that is not ignorable at least for patients with longer life expectancy. It may be noted that all percentile values, namely 25^th, 50^th, 75^th and of course mean are over 100 mSv for both years combined with 25^th percentile itself being 109 mSv and 75^th percentile 178 mSv. This finding necessitates a study of longer duration to have a better assessment of magnitude.

The dominant primary diseases contributing to multiple ¹⁸F-FDG PET/CTs in decreasing order of occurrence are melanoma (including head and neck melanoma), non-Hodgkin’s lymphoma (including diffuse large B-cell lymphoma and T-cell lymphoma), gastrointestinal cancer, breast cancer and Hodgkin’s lymphoma. It is interesting to note that while lymphoma and melanoma have a lower incidence in the population than breast and gastrointestinal malignancies, ^11,25 their representation was disproportionately more frequent in our cohort than would be expected based on their incidence in the general population. Various contributing factors to this overrepresentation possibly include the curability of these malignancies, utility of ¹⁸F-FDG PET/CT in establishing disease extent for malignancies with diffuse anatomic presentations, clinical referral pattern at our institution (including the clinically perceived role of ¹⁸F-FDG PET/CT in the management) as well as the preference to use other diagnostic or therapeutic modalities for the more prevalent malignancies (for example, colonoscopy for gastrointestinal malignancies, mammograms or breast MRI for breast cancer).

On reviewing the literature for other studies on lymphoma patients, ^26,27 it is noted that a mean CED ranging from 70 to 140 mSv, accrued in 3–4 years and patients with a CED >100 mSv were about 20%. Of these, Crowley et al. [Ref 31 from MB] studied 486 consecutive patients of lymphoma of mean age of 59 years for an observation period of 3.6 years. The median CED was 59 mSv with 14% of patients with a CED >150 mSv. The younger patients (under 40 years) had a higher median CED of 89 mSv. CT contributed to 89% of radiation dose and PET contributed to 8% of the dose.

A number of studies are available on children and adolescents ^28,29 with malignant lymphoma. Pierobon et al ²⁹ reported the median CED from X-ray and nuclear imaging studies of 518 and 309 mSv in Hodgkin and non-Hodgkin lymphoma, respectively with median follow-up of 5.3 and 7.5 years. Ninety percent of the patients received a CED > 100 mSv. The contribution of nuclear medicine imaging procedures amounted to 46% with the additional observation that 80% of the nuclear medicine dose was attributable to ⁶⁷Ga scans. There are a number of other studies also on children with lymphoma. ^28,30,31 Our study dominantly had adults.

Higher percentage of patients with CED >100 mSv of 43.6% have been reported in literature among those undergoing resection for lung cancer, but these patients have lower life expectancy. ³² Our study had much lower fraction (<5%) of patients with lung cancer.

Using such studies to achieve optimization in imaging remains the goal and has been emphasised in the past but it still needs underscoring. ³³ The diagnostic CT constitutes majority of the radiation dose resulting from a ¹⁸F-FDG PET/CT. The attenuation correction CT performed with PET at our institution is a non-contrast study of non-diagnostic quality. Moreover, our institute participates in ACR Dose Index Registry and our dose values for most CT examinations are consistently lower than national benchmark. ^21,22 Since the diagnostic CT contributes to majority of the radiation dose from a ¹⁸F-FDG PET/CT examination, optimising clinical protocols at various centres to minimise the number of diagnostic CT examinations these patients undergo during serial PET evaluations will lead to a more favourable radiation dose profile for this cohort subset.

Some studies suggested a modification of the surveillance strategy with an elimination of routine imaging after chemotherapy to lower the cumulative radiation dose in these predominantly younger patients and to help in reducing the health care expenses. ^30,31,34 In fact, the current guidelines of the European Society of Medical Oncology ³⁵ do not recommend routine ¹⁸F-FDG PET/CT in the surveillance and follow-up in Hodgkin’s lymphoma. They do recommend the use of ¹⁸F-FDG PET/CT for staging and risk stratification in pre-treatment evaluation,

This study has limitations. Our study was focussed only on dose assessment from ¹⁸F-FDG PET/CT exams and not on total dose. Oncology patients are usually examined by various imaging modalities such as SPECT or SPECT/CT or nuclear medicine studies for heart. In some cases, oncology patients may need a surgery facilitated with intraoperative imaging, which could involve a usage of radioisotopes or X-ray source. These patients typically involve many other diagnostic CT scans as we included only those done along with PET (same accession number). Also, some patients may have undergone fluoroscopic guided interventions that result in substantial radiation doses. ³⁶ Further, our study represents a single institute experience, and the study period of 2 years is relatively short. A study covering 10 years or more may give a better picture. It may be mentioned that while on one hand our institution has clinical decision support system in place and thus streamlined system for imaging appropriateness and lower doses by-and-large for CT part when compared to national benchmark through ACR Dose Index registry, ²² the same may not be the case at other institutions and that may lead to different results. The dose assessment methodologies are also known to lead to differences in dose estimates, which need to be kept in mind.

Many suggestions for future work have emerged from our study as indicated in many points discussed above. The consideration of the radiation dose helps in giving directions for future research. For example, if imaging appropriateness criteria are available for all the diseases with multiple imaging reported in this study and when available are they for initial work-up and diagnosis or for serial ¹⁸F-FDG PET/CT imaging. The conclusion from a recent paper analysing this aspect for CT imaging indicated that criteria are invariably not available for serial imaging ²¹ and thus necessitating development of criteria.

The cost-benefit ratio of serial ¹⁸F-FDG PET/CT imaging must ultimately be investigated to understand whether the additional subsequent ¹⁸F-FDG PET/CT examinations contributed to tangible changes to the patient’s management course.

Conclusions

This cohort study of 2 -year ¹⁸F-FDG PET/CT exams objectively demonstrated intensive use of ¹⁸F-FDG PET/CT for the patients with cancer, on average 1.6 exams per patient. Among those with ≥2 exams, age was ≤40 years (12%), 41–60 years (27%), and ≥61 years (61%). The top primary malignancies contributing to serial PET/CTs in decreasing frequency were melanoma, non-Hodgkin’s lymphoma (NHL), gastrointestinal cancer, breast cancer and Hodgkin’s lymphoma. Overall, one of 8 patients received cumulative effective dose ≥100 mSv, a level warrant continued attention. The comprehensive information of PET/CT usage, patient age, cancer site and radiation dose may be interesting and helpful for all those involved in framing guidelines for medical imaging in cancer care.