Metabolic Tumor Volume and TLG in FDG PET

Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) move FDG PET beyond the single hottest voxel: MTV is the volume of tumor above a segmentation threshold, and TLG is that volume multiplied by its mean SUV—so together they describe how much tumor is present and how active it is. Both are only as trustworthy as the segmentation method, SUV calibration, and acquisition protocol behind them, which is why volumetric PET is a quantitative-imaging discipline, not a button press. ^{1, 2}

For two decades, FDG PET reporting has leaned on SUVmax, the brightest voxel in a lesion. It is easy to compute and reproducible for a single point, but it is blind to tumor size and sensitive to noise. A 1-centimeter nodule and a 10-centimeter mass can share the same SUVmax while representing wildly different disease burdens. MTV and TLG were developed to capture that difference. ^{1, 3}

This guide explains what MTV and TLG are, the physics and segmentation choices that determine them, a worked calculation, the evidence that they carry prognostic weight across cancers, and the quality-control foundations a medical physicist puts in place so the numbers mean the same thing today, next quarter, and on the scanner down the hall. DRPS supports facilities with PET/CT and nuclear medicine physics and medical physics consulting across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware.

Introduction

FDG PET measures glucose metabolism. The standardized uptake value (SUV) normalizes the measured activity concentration in a voxel to the injected activity and the patient's body mass, giving a semiquantitative index of uptake. SUVmax, the maximum SUV in a region, is the most common reporting metric because it is operator-independent once a region is drawn and relatively insensitive to partial-volume blurring. ²

But SUVmax reflects one voxel. It says nothing about how large the tumor is, how heterogeneous its uptake is, or how much total glycolytic tissue the patient carries. Volumetric biomarkers were introduced to fill that gap. Total lesion glycolysis was proposed by Larson, Ginsberg, and colleagues in 1999 as a global measure of tumor glycolysis, computed from the metabolic volume of a lesion and its average uptake. ¹ Metabolic tumor volume followed as the volumetric component reported on its own. Today both are widely studied as prognostic and response biomarkers. ^{3, 5, 6}

Topic Explanation

What MTV and TLG actually measure

Metabolic tumor volume (MTV) is the volume of voxels inside a tumor whose SUV exceeds a defined segmentation threshold. It is reported in cubic centimeters (cm³), equivalent to milliliters. It answers the question "how much metabolically active tumor is there?"

Total lesion glycolysis (TLG) is MTV multiplied by the mean SUV (SUVmean) of that same volume. It answers "how much total glycolysis is there?"—weighting the volume by how avid it is. Larson's original definition framed it exactly this way, and also introduced the change in TLG during treatment (ΔTLG, sometimes called the Larson–Ginsberg index) as a response measure. ¹

The three common lesion-level metrics therefore form a hierarchy of information:

SUVmax — a single voxel; intensity only.
MTV — a volume; burden only.
TLG — volume weighted by mean intensity; burden and avidity combined.

When a patient has multiple lesions, the per-lesion values can be summed into whole-body totals, usually written TMTV (total metabolic tumor volume) and total TLG, which represent systemic tumor burden. ^{7, 10}

Why these are "quantitative imaging" biomarkers

SUV, and therefore MTV and TLG, are only meaningful if the entire measurement chain is controlled: scanner calibration against a dose calibrator, accurate injected-activity and residual-activity records, patient weight, a controlled uptake (incubation) time, blood glucose, a consistent reconstruction, and corrections for attenuation, scatter, randoms, and dead time. The EANM procedure guidelines for tumour FDG PET imaging exist precisely to standardize this chain so that semiquantitative values are repeatable within a patient and reproducible across sites. ² For the SUV fundamentals underneath these volumetric metrics, see our primers on SUV quantification and EARL SUV harmonization.

Key Technical Principles

The core relationship

The defining equation for TLG is simple, which is part of its appeal:

TLG = MTV \times \overline{SUV}

where $\overline{SUV}$ is the mean SUV within the segmented metabolic volume. MTV in turn is the integral of the voxel volume over all voxels whose SUV exceeds the segmentation threshold $T$ :

MTV = i : SUV_{i} \geq T \sum v_{i}

with $v_{i}$ the volume of voxel $i$ . The threshold $T$ is where all the subtlety lives.

Segmentation methods and why they disagree

There is no single "correct" tumor boundary on a blurred PET image. The common segmentation approaches are:

Fixed absolute threshold — include voxels above an absolute SUV (for example SUV ≥ 2.5). Simple, but sensitive to how avid the tumor is and to background.
Percentage of SUVmax — include voxels above a fraction of the lesion's own peak, commonly 41% or 50% of SUVmax. The EANM guideline describes a 41% SUVmax approach among recommended methods for volume delineation. ²
Gradient / edge-based — find the boundary where the image intensity gradient is steepest rather than at a level set.
Statistical / adaptive and, increasingly, machine-learning methods that adapt the threshold to local contrast and noise.

These methods do not agree. In a study comparing FDG PET segmentation against the measured volume of resected head and neck, lung, and colorectal tumors, the mean segmented volume shifted substantially with method—gradient-based, 30%, 40%, and 50% SUVmax thresholds produced different volumes and different agreement with pathology, with the higher percentage thresholds systematically underestimating small tumors. ⁴ The practical rule follows directly: fix one segmentation method and keep it constant within a study, a protocol, and an individual patient's follow-up.

Metric	What it captures	Units	Governed primarily by	Main weakness
SUVmax	Single hottest voxel	SUV (unitless)	Noise, reconstruction, recovery	Ignores size; noise-sensitive
SUVpeak	Mean of a small fixed volume (~1 cm³) at the hottest region	SUV (unitless)	Reconstruction, recovery	Still intensity-only
MTV	Volume above threshold	cm³ (mL)	Segmentation threshold, contrast	Threshold-dependent
TLG	Volume × mean SUV	SUV·cm³ ("g")	Threshold and SUV calibration	Inherits both error sources

A worked example

Suppose a lesion is segmented with a 41% SUVmax threshold and the software reports a metabolic volume of 40 cm³ with a mean SUV of 6.5 inside that volume. Then:

TLG = 40 cm^{3} \times 6.5 = 260 SUV\cdotcm^{3}

Now imagine two different reporting choices for the same scan. If instead a 50% SUVmax threshold were applied, the volume might fall to roughly 28 cm³ with a higher SUVmean of about 7.6 (the boundary now hugs the avid core), giving:

TLG = 28 cm^{3} \times 7.6 \approx 213 SUV\cdotcm^{3}

The same tumor, the same scan, two defensible thresholds—and MTV changed by 30% and TLG by about 18%. This is not measurement error; it is definitional. It is why a volumetric value without a stated segmentation method is uninterpretable, and why cross-site comparison requires harmonization of both the SUV chain and the segmentation rule. ^{2, 4}

Response assessment and ΔTLG

For treatment response, the change matters more than the absolute. PERCIST 1.0 formalized FDG PET response assessment around a consistent protocol, a liver reference region, a fixed small region of interest in the most active lesion, and a 30% decline in SUL as a threshold for response. ³ Volumetric change extends this idea: the fractional change in TLG from baseline to follow-up,

Δ TLG = \frac{TLG _{pre} - TLG _{post}}{TLG _{pre}} \times 100%

is the modern expression of Larson and Ginsberg's original response index. ¹ It can register response that SUVmax alone misses—for example when a tumor shrinks in active volume while its hottest voxel changes little.

Clinical Impact

Across many solid tumors and lymphomas, higher baseline MTV and TLG have been associated with worse survival—often more strongly than SUVmax. The signal is consistent enough across independent meta-analyses to be clinically credible, even though absolute cutoffs vary by disease and method.

Head and neck cancer: a meta-analysis of 13 studies and 1,180 patients found high MTV and high TLG each associated with roughly a threefold increase in adverse events and in death (pooled hazard ratios around 3.0–3.5). ⁵
Non-small cell lung cancer: a meta-analysis of 36 studies and 5,807 surgical patients found MTV and TLG independently predicted disease-free and overall survival, with adjusted hazard ratios in the 1.9–3.0 range. ⁶
Breast cancer: a meta-analysis of 975 patients found high primary-tumor MTV and TLG associated with more progression and recurrence. ⁷
Advanced cervical cancer: pretreatment MTV and TLG were independent predictors of overall and recurrence-free survival and outperformed SUVmax. ⁸
Endometrial cancer: MTV and TLG were significant prognostic factors for disease-free survival where SUVmax was not, and MTV helped stratify who needed adjuvant therapy. ⁹
T-cell lymphoblastic lymphoma: baseline total MTV and TLG were strong predictors of progression-free and overall survival and sharpened a risk model. ¹⁰

The through-line is that disease burden—captured by volume and glycolysis—carries prognostic information that a single voxel cannot. For response, volumetric metrics complement PERCIST-style SUL reporting; see PERCIST tumor response quantification and, for lymphoma specifically, FDG PET lymphoma Deauville response.

Practical Optimization Tips

A volumetric PET program lives or dies on protocol discipline. The priorities:

1. Lock the SUV chain

MTV and TLG inherit every SUV error. Confirm scanner-to-dose-calibrator cross-calibration on schedule, record injected and residual activity, verify patient weight at the scanner (not from the chart), and control blood glucose. A 15% SUV bias becomes a 15% TLG bias before segmentation even begins.

2. Fix the uptake time

SUV rises with uptake time for most tumors. Standardize the FDG incubation interval (commonly about 60 minutes) and keep it within a narrow window, because a drifting uptake time silently changes both SUVmean and the threshold-defined volume. See PET uptake time for why this matters.

3. Freeze the reconstruction

Point-spread-function and time-of-flight reconstructions raise recovery and SUVmax and shift segmented volumes. Use one reconstruction for quantitative reporting and keep an EANM-compliant harmonized reconstruction available when values must be comparable across systems. ²

4. Standardize segmentation

Choose one segmentation method (for example a 41% SUVmax threshold) and apply it identically at baseline and follow-up. Document it in the report. Never compare a 41%-threshold MTV to a gradient-based MTV. ^{2, 4}

5. Account for the partial-volume effect

Small lesions (below roughly two to three times the scanner's spatial resolution) suffer partial-volume underestimation of SUVmean, biasing TLG low and making small-volume segmentation unstable. Flag sub-centimeter lesions as quantitatively unreliable. See the PET partial-volume effect.

Common pitfalls

Reporting MTV or TLG without the segmentation method. The number is meaningless alone.
Changing reconstruction or uptake time between scans. This masquerades as real change.
Comparing values across unharmonized scanners. Calibration and recovery differ.
Trusting volumetrics on tiny or low-contrast lesions. Partial-volume and noise dominate.
Treating a published cutoff as universal. Cutoffs are method- and disease-specific.

Regulatory Considerations

MTV and TLG are not separately "regulated," but the quantitative chain that produces them is governed by accreditation and guideline frameworks a facility must satisfy. FDG itself is a positron-emitting byproduct material administered under NRC or Agreement State medical-use rules; the quantitative accuracy of the images is governed by PET/CT accreditation and professional guidelines.

EANM procedure guidelines for tumour FDG PET imaging (version 2.0) define the standardized acquisition, calibration, and SUV-harmonization requirements that make volumetric metrics reproducible, including recommended volume-delineation methods. ²
PERCIST 1.0 provides the response-assessment framework—reference-region normalization, consistent protocol, and defined response thresholds—within which volumetric change is interpreted. ³
ACR–ACNM and comparable PET/CT accreditation programs require documented scanner calibration, phantom performance, and dose-calibrator consistency, which are the physics underpinnings of any SUV-derived value.

Of the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are NRC Agreement States regulating the medical use of FDG under their own radiation-control rules, while Washington, DC and Delaware are regulated directly by the NRC. A facility pursuing quantitative PET should align its protocol with accreditation requirements and with a documented quality-control program. DRPS provides this support through PET/CT and nuclear medicine physics and accreditation support.

Frequently Asked Questions (FAQs)

What is metabolic tumor volume (MTV)?

Metabolic tumor volume is the volume of tumor tissue whose FDG uptake exceeds a chosen segmentation threshold on a PET image. It is reported in cubic centimeters (equivalently milliliters) and represents the metabolically active tumor burden rather than the single hottest voxel. Because MTV depends on the threshold used to draw the tumor boundary, the segmentation method must be stated and kept consistent for the value to be meaningful.

What is total lesion glycolysis (TLG)?

Total lesion glycolysis is the metabolic tumor volume multiplied by the mean standardized uptake value (SUVmean) within that volume. It combines how much tumor is active (volume) with how active it is (mean uptake) into a single number, originally introduced by Larson and Ginsberg in 1999. TLG is reported in units of SUV times cubic centimeters, often written as grams.

How are MTV and TLG different from SUVmax?

SUVmax is the single highest-uptake voxel in a lesion, so it ignores tumor size and is sensitive to image noise. MTV measures the volume of active tumor, and TLG combines volume with mean activity. In many cancers MTV and TLG predict outcomes better than SUVmax because they reflect total disease burden rather than one voxel.

Why does the segmentation method matter so much for MTV?

MTV is defined entirely by where the tumor boundary is drawn. A fixed absolute threshold (for example SUV 2.5), a percentage of SUVmax (commonly 41% or 50%), and gradient-based or statistical methods can give substantially different volumes for the same lesion. Studies comparing PET segmentation to resected-tumor volume show the method changes both the volume and its agreement with pathology, so the method must be fixed within a study or a patient's follow-up.

Can MTV and TLG be compared across different scanners or sites?

Only if the quantitative chain is harmonized. SUV depends on scanner calibration, injected activity and residual, patient weight, uptake time, reconstruction, and recovery. The EANM procedure guidelines and SUV harmonization programs exist precisely so semiquantitative values are comparable across systems. Without harmonization and identical segmentation, cross-scanner MTV and TLG comparisons are unreliable.

What is the clinical value of MTV and TLG?

Across head and neck, lung, breast, cervical, endometrial, and lymphoid malignancies, higher baseline MTV and TLG have been associated with worse progression-free and overall survival in published meta-analyses and cohort studies. They are prognostic and increasingly used in response assessment and radiotherapy target definition, though standardization remains a prerequisite for routine quantitative use.

Does measuring MTV and TLG require a special PET protocol?

It requires a disciplined standard protocol: a calibrated scanner, controlled uptake time, documented injected activity and glucose, a consistent reconstruction, and a fixed segmentation method. These are the same quality-control foundations that support any quantitative SUV reporting, which is why a medical physicist is central to a defensible volumetric PET program.

Key Takeaways

MTV is volume above a threshold; TLG is that volume times its mean SUV. Together they describe tumor burden and avidity, which SUVmax cannot. ¹
Segmentation defines the result. Fixed-absolute, percentage-of-SUVmax (41% or 50%), and gradient methods give different volumes; the method must be stated and held constant. ^{2, 4}
Volumetrics inherit every SUV error. Calibration, injected activity, weight, uptake time, and reconstruction must be controlled before segmentation. ²
They carry prognostic weight. Higher baseline MTV and TLG track with worse survival across head and neck, lung, breast, gynecologic, and lymphoid cancers. ^{5, 6, 7, 8, 9, 10}
Standardization is the price of admission. Without a harmonized chain and a fixed segmentation rule, cross-scanner and cross-time comparisons are not defensible.

Conclusion

Metabolic tumor volume and total lesion glycolysis are the natural answer to SUVmax's blind spot: they measure how much tumor is metabolically active and how active it is, not just the brightness of one voxel. The evidence that they carry prognostic information is broad and consistent. But that value is conditional. MTV and TLG are defined by a segmentation threshold layered on top of an SUV that is itself the product of a long calibration and acquisition chain. Change the threshold, the reconstruction, or the uptake time, and the number changes—without the tumor changing at all.

That is the physicist's role. Volumetric PET is not a feature you turn on; it is a quantitative program you maintain. A facility that locks its SUV chain, fixes its segmentation method, harmonizes its reconstructions, and documents all of it can report MTV and TLG that mean the same thing across time and across scanners. A facility that does not is reporting numbers that look quantitative but are not.

How DRPS Can Help

Diagnostic Radiation Physics Services helps PET/CT facilities build and defend the quantitative foundation that volumetric biomarkers require: scanner-to-dose-calibrator cross-calibration, phantom-based performance testing, EANM-style harmonized reconstructions, uptake-time and injection-workflow review, and documentation that survives accreditation. This work is delivered through PET/CT and nuclear medicine physics, accreditation support, and medical physics consulting.

DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware.

A quantitative PET value is only as good as the program behind it—and the program is where the physics lives.

Related Resources

References

Larson SM, Erdi Y, Akhurst T, et al. Tumor Treatment Response Based on Visual and Quantitative Changes in Global Tumor Glycolysis Using PET-FDG Imaging: The Visual Response Score and the Change in Total Lesion Glycolysis. Clin Positron Imaging. 1999;2(3):159-171. doi:10.1016/s1095-0397(99)00016-3. PubMed
Boellaard R, Delgado-Bolton R, Oyen WJG, et al. FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur J Nucl Med Mol Imaging. 2015;42(2):328-354. doi:10.1007/s00259-014-2961-x. PubMed
Wahl RL, Jacene H, Kasamon Y, Lodge MA. From RECIST to PERCIST: Evolving Considerations for PET Response Criteria in Solid Tumors. J Nucl Med. 2009;50(Suppl 1):122S-150S. doi:10.2967/jnumed.108.057307. PubMed
Sridhar P, Mercier G, Tan J, Truong MT, Daly B, Subramaniam RM. FDG PET metabolic tumor volume segmentation and pathologic volume of primary human solid tumors. AJR Am J Roentgenol. 2014;202(5):1114-1119. doi:10.2214/AJR.13.11456. PubMed
Pak K, Cheon GJ, Nam HY, et al. Prognostic value of metabolic tumor volume and total lesion glycolysis in head and neck cancer: a systematic review and meta-analysis. J Nucl Med. 2014;55(6):884-890. doi:10.2967/jnumed.113.133801. PubMed
Liu J, Dong M, Sun X, Li W, Xing L, Yu J. Prognostic Value of 18F-FDG PET/CT in Surgical Non-Small Cell Lung Cancer: A Meta-Analysis. PLoS One. 2016;11(1):e0146195. doi:10.1371/journal.pone.0146195. PubMed
Pak K, Seok JW, Kim HY, et al. Prognostic value of metabolic tumor volume and total lesion glycolysis in breast cancer: a meta-analysis. Nucl Med Commun. 2020;41(8):824-829. doi:10.1097/MNM.0000000000001227. PubMed
Calles-Sastre L, Mucientes-Rasilla J, San-Frutos Llorente LM, et al. Prognostic significance of metabolic tumor volume and total lesion glycolysis in patients with advanced cervical carcinoma. Rev Esp Med Nucl Imagen Mol. 2019;38(1):17-21. doi:10.1016/j.remn.2018.08.001. PubMed
Erdogan M, Erdemoglu E, Evrimler Ş, Hanedan C, Şengül SS. Prognostic value of metabolic tumor volume and total lesion glycolysis assessed by 18F-FDG PET/CT in endometrial cancer. Nucl Med Commun. 2019;40(11):1099-1104. doi:10.1097/MNM.0000000000001091. PubMed
Feng X, Wen X, Li L, et al. Baseline Total Metabolic Tumor Volume and Total Lesion Glycolysis Measured on 18F-FDG PET-CT Predict Outcomes in T-Cell Lymphoblastic Lymphoma. Cancer Res Treat. 2021;53(3):837-846. doi:10.4143/crt.2020.123. PubMed