Improving specificity of PET/CT for detection of recurrent lymphoma in immunocompromised patient

The patient is a 47-year-old man with fairly well-controlled HIV/AIDS, hepatitis B without cirrhosis and former polysubstance user, with pulmonary dysfunction secondary to idiopathic fibrosis. He was diagnosed with stage IV Hodgkin’s disease involving the liver and bone marrow. His baseline diffusion capacity for carbon monoxide was decreased, precluding therapy with bleomycin. As such, he received CEPP regimen that consisted of cyclophosphamide 600 mg/m² IV on day 1 and 8; etoposide 70 mg/m² IV on days 1 to 3; procarbazine (Matulane, Sigma Tau) 60 mg/m² orally on days 1 to 10; and prednisone 60 mg/m² orally on days 1 to 10.

The treatment was interrupted because of patient noncompliance and infectious complications. He eventually received five cycles, but unfortunately, he refused further treatment and never achieved a complete remission.

He relocated to New York and established care in our institution shortly after. His PET/CT at that time showed diffuse lymphadenopathy with significant hypermetabolic uptake in the appendix, without clinically evident appendicitis, bone, liver and splenules. Of note, the patient had a history of remote splenectomy caused by trauma. The bone marrow biopsy confirmed classic Hodgkin’s disease with lymphohistiocytic infiltrate CD30, PAX5-positive and CD20, CD45-negative.

He received two cycles of ICE salvage therapy that consisted of carboplatin (area under the curve of 6) on day 1; ifosfamide 1,500 mg/m² on days 1 to 3; mesna 300 mg/m² on days 1 to 3; and etoposide 100 mg/m² with growth factor support. The restaging CT of chest, abdomen and pelvis demonstrated excellent response, so he received an additional two cycles.

After four cycles of ICE therapy, his PET/CT showed a complete remission. Prominent bone marrow activity was felt to be benign and attributed to etoposide administration. He was referred to the transplant center for the evaluation of autologous stem cell transplant. Unfortunately, he refused to undergo the stem cell collection. His HIV was not controlled adequately, although he was taking an antiretroviral regimen.

Although the patient is post-splenectomy, there is a rounded soft tissue mass (red circle) in the left upper quadrant that likely regenerated splenic tissue. At the time of this scan in December, there is normal mild FDG activity within this tissue as expected in a normal spleen. Images courtesy of L Makarian; N Gupta; M Ghesani

This subsequent scan in September after multiple rounds of chemotherapy demonstrates a new subtle small hypodense lesion in the splenule (red circle). The corresponding functional PET image demonstrates a focus of abnormal FDG uptake with a standard uptake value of 3, which is above the background splenic activity. This is highly suspicious for recurrent disease.

About 4 months later, he started to complain of abdominal pain, and the abdomen and pelvis CT showed an interval development of a 2.5-cm right paravertebral soft tissue at T9-T10 level that was suspicious for recurrent lymphoma. He also had minimally increased small and borderline lymph nodes in the abdomen and pelvis.

His axillary lymph nodes were stable, and small bilateral inguinal lymph nodes became slightly larger. PET/CT was recommended for suspected recurrence, although bilateral axillary and inguinal lymph nodes were felt to be indeterminate, as they often are with HIV-associated lymphadenopathy. PET/CT not only showed increasing metabolic activity in the nodes but also a new hypodense lesion in the accessory spleen with associated subtle but suspicious hypermetabolic activity. The latter finding strongly supported the possibility of recurrent lymphoma.

This initial PET/CT in December demonstrates foci of mildly increased FDG activity within the bone marrow of the pelvic bones suspicious for disease involvement.

This PET/CT in May demonstrates intense uptake throughout the marrow containing spaces of the pelvis, markedly increased from the initial scan. Notice on the whole body MIP image, there has been interval development of intense uptake throughout the ribs, spine, and pelvis compared to the prior exam (Figure 3). This is consistent with the effect of etoposide, although evaluation for focal lesions cannot be made on this background.

He also had new foci of hypermetabolic activity associated with the cervical and portacaval lymph nodes. There was 3.1-cm × 1.1-cm right paravertebral soft tissue mass at the level of T9 and T10, with suggestion of metabolic activity associated with this soft tissue. He had an interval progression of skeletal involvement. Interestingly, there was a significant decrease in prominent diffuse bone marrow activity, as etoposide was no longer being given. The biopsy of the paravertebral mass is currently planned.

Finally, this PET/CT in September demonstrates decrease in the diffuse marrow uptake throughout the spine, ribs, and pelvis that was seen on the May exam; however there has been interval development of multiple focal regions of abnormal intense uptake (red circles along the left sacrum and left iliac wing). The decrease in diffuse marrow activity is consistent with resolving etoposide effect, while the development of focal abnormal uptake is highly suspicious for recurrent disease.

Discussion

PET/CT is recommended and is widely used for staging and after completion of the therapy to assess the response in the Hodgkin’s lymphomas. Additionally, the importance of interim PET/CT in the midst of therapy to predict response was reported by Dann and colleagues. He studied it after two cycles of BEACOPP therapy in 47 patients with standard and high-risk disease. Almost 30% of patients relapsed if the interim PET/CT was positive, and only 2.3% patients relapsed if interim PET/CT was negative. Several current trials in Hodgkin’s disease include interim PET/CT in the algorithm, and if positive, they escalate the therapy after two cycles of doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD).

The role of PET/CT in post-therapy surveillance, however, remains controversial. In fact, the National Comprehensive Cancer Network guidelines do not recommend it for routine use. Moreover, in patients with HIV, the finding could be harder to interpret. The fluorine-18 fluorodeoxyglucose detects active lymphoid tissue associated with any condition, including HIV infection.

Goshen retrospectively studied the role of PET/CT and correlated it with concurrent clinical, immunologic and viral load data. PET/CT accurately correlated with the extent of lymphoma in 12 of 16 patients. In the cases in which PET and CT findings were discrepant, increased viral loads and CD4 levels implied benign HIV-related lymphadenopathy.

In our patient, presence of a new hypodense hypermetabolic lesion in the splenule makes it very suspicious for recurrence. This is further supported by evidence of worsening of skeletal involvement.

Liana Makarian, MD, is an oncology Fellow at St Luke’s-Roosevelt Hospital Center.

Neil Gupta, MD, is a resident in radiology at St Luke’s-Roosevelt Hospital Center.

Munir Ghesani, MD, is an attending radiologist at St. Luke’s-Roosevelt Hospital Center and associate clinical professor of radiology at Columbia University College of Physicians and Surgeons.

For more information:

• Dann EJ. Blood. 2007;109:905-909.
• Goshen E. Clin Nucl Med. 2008;33:610-614.