Imaging Analysis

How can you explain decreasing SUVs despite obvious progression of metastatic breast cancer?

A 55-year-old woman presented to our center for restaging PET/CT. Her history was significant for recurrent breast cancer in the mediastinum and lungs. She initially underwent left mastectomy, adjuvant chemotherapy and radiotherapy. She was started on palliative chemotherapy. Her initial staging PET/CT showed subcentimeter medistinal lymph nodes (maximal SUV 8.4), right lower lobe mass 1.4 × 1.2 cm (SUV 5.3), scattered subcentimeter nodules in the lung. Her restaging scan with a PET/CT after chemotherapy showed progressive disease with right axillary lymph nodes measuring 1 cm (SUV 2.0), medistinal lymph nodes measuring 1 cm (SUV 3.6 to 10.1), right lower lobe mass measuring 2.1 × 1.4 cm (SUV 7.1), relatively stable pulmonary nodules and new left pleural effusion.

After a change in her chemotherapy regimen, her repeat PET/CT showed increased AP window lymph nodes measuring up to 2.5 cm (however, SUV decreased to 6.3 from 10.1), new subcarinal soft tissue measures 1.4 cm (SUV 5.5), increase in size of epicardial lymph nodes to 1.1 cm (SUV 4.4), right lower lobe mass increased in size to 2.3 × 1.7 cm (however, SUV decreased to 4.7 from 7.1), new pulmonary nodule, unchanged right axillary lymph nodes and diffusely intense uptake of FDG activity in the bone marrow. This patient has progressive disease based on the increasing size of the metastatic lesions. How can we explain decreasing SUVs despite obvious progression of disease?

Figure 1: Axial CT and PET images demonstrating progression of prevascular lymphadenopathy on anatomic images.
Figure 1: Axial CT (left column) and PET (right column) images demonstrating progression of prevascular lymphadenopathy on anatomic images (second row is a follow up imaging study). However, SUVs are declining despite visible increase in intensity and extent of hypermetabolism. Note interval increased marrow activity as depicted by increasing intensity of uptake in the ribs and thoracic spine.

Figure 2: Axial CT and PET images demonstrating progression of a lung lesion on anatomic images.
Figure 2: Axial CT (left column) and PET (right column) images demonstrating progression of a lung lesion on anatomic images (second row is a follow up imaging study). However, SUVs are declining despite visible increase in intensity and extent of hypermetabolism.

Source: M. Ghesani

DISCUSSION

FDG-PET is shown to detect metastatic lesions and identify equivocal lesions on conventional imaging. One of the limitations of PET/CT is the sensitivity to detect malignant lesions decreases with low-grade tumors and lesions <1 cm and in breast cancer, axillary metastasis.

A retrospective analysis in use of PET/CT in recurrent breast cancer showed a sensitivity of 90%, specificity of 71% and accuracy of 83% with a change in management being recorded in 51% of patients. In a prospective study done by Souvatzoglou et al in recurrent breast cancer, PET/CT was positive in 67% of patients and negative in 10% of patients. Local recurrence was identified in 12% of patients; metastatic sites were lymph nodes (36%), bone (32%), liver (14%), lungs (10%), adrenal glands (3%) and pleura (2%). PET/CT had changed the management in 33% of the patients.

Interval response to chemotherapy in nonosseous metastases can be in the form of an increase in FDG avidity and increasing SUVs, indicating a treatment failure. However, “steal phenomenon” occurs when an increased uptake of FDG is noted in the bone marrow with relatively lower available FDG activity to the nonosseous metastatic lesion. This steal phenomenon could be due to the effect of chemotherapy or usage of growth factors. This case illustrates steal phenomenon and importance of the need to correlate overall interval change on both anatomic and functional imaging and not solely rely on SUVs as a marker of either a favorable response or progression of disease.

Vamsee Torri, MD, is a Fellow in Hematology/Oncology at St. Luke’s-Roosevelt Hospital Center.

Iwao Tanaka, MD, is a Resident in Radiology at St.Luke’s-Roosevelt Medical Center.

Rami Daya, MD, is a practicing oncologist in Brooklyn, N.Y.

Munir Ghesani, MD, is Associate Clinical Professor of Radiology at Columbia University College of Physicians and Surgeons and Attending Radiologist at St.Luke’s-Roosevelt Medical Center.

For more information:

  • Radiographics. 2007;27:S215-S229.
  • Cancer. 2006;107:2545-2551.
  • J Nucl Med. Meeting Abstracts, 2008. 49:18P
  • Informa Health Care. 2008. ISBN 0849380871, 9780849380877.

A 55-year-old woman presented to our center for restaging PET/CT. Her history was significant for recurrent breast cancer in the mediastinum and lungs. She initially underwent left mastectomy, adjuvant chemotherapy and radiotherapy. She was started on palliative chemotherapy. Her initial staging PET/CT showed subcentimeter medistinal lymph nodes (maximal SUV 8.4), right lower lobe mass 1.4 × 1.2 cm (SUV 5.3), scattered subcentimeter nodules in the lung. Her restaging scan with a PET/CT after chemotherapy showed progressive disease with right axillary lymph nodes measuring 1 cm (SUV 2.0), medistinal lymph nodes measuring 1 cm (SUV 3.6 to 10.1), right lower lobe mass measuring 2.1 × 1.4 cm (SUV 7.1), relatively stable pulmonary nodules and new left pleural effusion.

After a change in her chemotherapy regimen, her repeat PET/CT showed increased AP window lymph nodes measuring up to 2.5 cm (however, SUV decreased to 6.3 from 10.1), new subcarinal soft tissue measures 1.4 cm (SUV 5.5), increase in size of epicardial lymph nodes to 1.1 cm (SUV 4.4), right lower lobe mass increased in size to 2.3 × 1.7 cm (however, SUV decreased to 4.7 from 7.1), new pulmonary nodule, unchanged right axillary lymph nodes and diffusely intense uptake of FDG activity in the bone marrow. This patient has progressive disease based on the increasing size of the metastatic lesions. How can we explain decreasing SUVs despite obvious progression of disease?

Figure 1: Axial CT and PET images demonstrating progression of prevascular lymphadenopathy on anatomic images.
Figure 1: Axial CT (left column) and PET (right column) images demonstrating progression of prevascular lymphadenopathy on anatomic images (second row is a follow up imaging study). However, SUVs are declining despite visible increase in intensity and extent of hypermetabolism. Note interval increased marrow activity as depicted by increasing intensity of uptake in the ribs and thoracic spine.

Figure 2: Axial CT and PET images demonstrating progression of a lung lesion on anatomic images.
Figure 2: Axial CT (left column) and PET (right column) images demonstrating progression of a lung lesion on anatomic images (second row is a follow up imaging study). However, SUVs are declining despite visible increase in intensity and extent of hypermetabolism.

Source: M. Ghesani

DISCUSSION

FDG-PET is shown to detect metastatic lesions and identify equivocal lesions on conventional imaging. One of the limitations of PET/CT is the sensitivity to detect malignant lesions decreases with low-grade tumors and lesions <1 cm and in breast cancer, axillary metastasis.

A retrospective analysis in use of PET/CT in recurrent breast cancer showed a sensitivity of 90%, specificity of 71% and accuracy of 83% with a change in management being recorded in 51% of patients. In a prospective study done by Souvatzoglou et al in recurrent breast cancer, PET/CT was positive in 67% of patients and negative in 10% of patients. Local recurrence was identified in 12% of patients; metastatic sites were lymph nodes (36%), bone (32%), liver (14%), lungs (10%), adrenal glands (3%) and pleura (2%). PET/CT had changed the management in 33% of the patients.

Interval response to chemotherapy in nonosseous metastases can be in the form of an increase in FDG avidity and increasing SUVs, indicating a treatment failure. However, “steal phenomenon” occurs when an increased uptake of FDG is noted in the bone marrow with relatively lower available FDG activity to the nonosseous metastatic lesion. This steal phenomenon could be due to the effect of chemotherapy or usage of growth factors. This case illustrates steal phenomenon and importance of the need to correlate overall interval change on both anatomic and functional imaging and not solely rely on SUVs as a marker of either a favorable response or progression of disease.

Vamsee Torri, MD, is a Fellow in Hematology/Oncology at St. Luke’s-Roosevelt Hospital Center.

Iwao Tanaka, MD, is a Resident in Radiology at St.Luke’s-Roosevelt Medical Center.

Rami Daya, MD, is a practicing oncologist in Brooklyn, N.Y.

Munir Ghesani, MD, is Associate Clinical Professor of Radiology at Columbia University College of Physicians and Surgeons and Attending Radiologist at St.Luke’s-Roosevelt Medical Center.

For more information:

  • Radiographics. 2007;27:S215-S229.
  • Cancer. 2006;107:2545-2551.
  • J Nucl Med. Meeting Abstracts, 2008. 49:18P
  • Informa Health Care. 2008. ISBN 0849380871, 9780849380877.