Milos J. Janicek

Rituximab and CHOP Induction Therapy for Newly Diagnosed Mantle-Cell Lymphoma: Molecular Complete Responses Are Not Predictive of Progression-Free Survival

PURPOSE To evaluate the efficacy of rituximab and cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) induction therapy in patients with newly diagnosed mantle-cell lymphoma (MCL). PATIENTS AND METHODS From March 1997 to May 1999, 40 previously untreated patients with stage II through IV MCL were treated with six cycles of rituximab and CHOP chemotherapy in a phase II trial. Pretreatment and interval peripheral-blood (PB) and bone marrow (BM) specimens were also analyzed by polymerase chain reaction (PCR) for tumor-specific BCL-1/immunoglobulin H (IgH) translocations and clonal IgH rearrangements. Study end points included clinical and molecular response rates and long-term progression-free survival (PFS). RESULTS Forty-eight percent of patients achieved a complete response (CR)/CR unconfirmed (CRu), and 48% of patients obtained a partial response (PR). However, 28 of the 40 patients have already relapsed or developed progressive disease with a median PFS of 16.6 months. Twenty-five patients had PCR-detectable BCL-1/IgH or clonal IgH products in PB or BM at diagnosis. Nine of the 25 informative patients had no evidence of PCR-detectable disease in PB or BM after rituximab and CHOP therapy. However, patients who achieved molecular remissions in PB or BM had PFS similar to patients without molecular remissions (16.5 v 18.8 months, P =.51). CONCLUSION Favorable clinical and molecular response rates associated with rituximab and CHOP chemotherapy do not translate into prolonged PFS in MCL. Nevertheless, rituximab and combination chemotherapy may transiently clear PB or BM of detectable tumor cells, prompting additional consideration of antibody-based in vivo purging in subsequent clinical trials.

Phase I Clinical Trial of Recombinant Human Endostatin Administered as a Short Intravenous Infusion Repeated Daily

PURPOSE To perform a phase I trial of recombinant human endostatin (rhEndostatin; EntreMed, Rockville, MD) given as a daily 20-minute intravenous (IV) injection in adult patients with refractory solid tumors. PATIENTS AND METHODS The daily dose was increased from 15 to 240 mg/m(2) by a factor of 100% in cohorts of three patients. In the absence of dose-limiting toxicity, uninterrupted treatment was continued until the tumor burden increased by more than 50% from baseline. Correlative studies included dynamic contrast-enhanced magnetic resonance imaging of tumor blood flow, urinary vascular endothelial growth factor and basic fibroblast growth factor levels, rhEndostatin serum pharmacokinetics, and monitoring of circulating antibodies to rhEndostatin. RESULTS There were no notable treatment related toxicities among 15 patients receiving a total of 50 monthly cycles of rhEndostatin. One patient with a pancreatic neuroendocrine tumor had a minor response and two patients showed disease stabilization. Linearity in the pharmacokinetics of rhEndostatin was indicated by dose-proportionate increases in the area under the curve for the first dose and the peak serum concentration at steady state. Daily systemic exposure to rhEndostatin in patients receiving 240 mg/m(2)/d was approximately 50% lower than that provided by the therapeutically optimal dose in preclinical studies. CONCLUSION rhEndostatin administered as a 20-minute daily IV injection at doses up to 240 mg/m(2) showed no significant toxicities. Evidence of clinical benefit was observed in three patients. Due to high variability between the peak and trough serum concentrations associated with the repeated short IV infusion schedule, daily serum drug levels only briefly exceeded concentrations necessary for in vitro antiangiogenic effects.

High-dose CHOP as initial therapy for patients with poor-prognosis aggressive non-Hodgkin's lymphoma: a dose-finding pilot study.

PURPOSE To purpose of this study was to develop a more effective approach to the treatment of patients with poor-prognosis aggressive non-Hodgkins lymphoma (NHL). PATIENTS AND METHODS Thirty newly diagnosed patients with bulky (> or = 10 cm) advanced-stage aggressive NHL were enrolled onto a pilot study. The study was designed to determine the maximum-tolerated dosages (MTD) of cyclophosphamide and doxorubicin that could be used in a high-dose cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) regimen with granulocyte colony-stimulating factor (G-CSF) support and to assess preliminarily the efficacy of the regimen. RESULTS In the initial dose-finding portion of the study, cumulative thrombocytopenia was the dose-limiting toxicity. At the MTD, the regimen included four 21-day cycles of cyclophosphamide 4 gm/m2, doxorubicin 70 mg/m2, vincristine 2 mg, and prednisone 100 mg for 5 days with mesna and G-CSF support. At the MTD, 65% of treatment cycles were complicated by febrile neutropenia, 84% of patients received at least one platelet transfusion for platelet counts less than 20,000/microL, and there was one treatment-related death. Nineteen of 22 (86%; 90% confidence interval [CI], 68 to 96) patients treated at the MTD achieved an initial complete response (CR), and 79% (90% CI, 58 to 92) of the complete responders and 69% of all patients remain progression-free with 20 months median follow-up. CONCLUSION The high-dose CHOP regimen may be an effective alternative for patients with poor-prognosis aggressive NHL.

Early restaging gallium scans predict outcome in poor-prognosis patients with aggressive non-Hodgkin's lymphoma treated with high-dose CHOP chemotherapy.

PURPOSE This prospective study assessed the predictive value of early restaging gallium (Ga) and computed tomographic (CT) scans in poor-prognosis patients with aggressive non-Hodgkins lymphoma (NHL) who were treated with high-dose cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy. PATIENTS AND METHODS Thirty newly diagnosed patients with bulky (> or = 10 cm) advanced-stage aggressive NHL were treated with a four-cycle high-dose CHOP regimen (22 patients at maximum-tolerated dose [MTD]: cyclophosphamide 4 g/m2, doxorubicin 70 mg/m2, vincristine 2 mg, and prednisone 100 mg orally for 5 days). All patients had chest/abdominal/pelvic CT scans and 10-mCi Ga scans at baseline and following two and four cycles of therapy. Scans were reviewed in a blinded manner for CT-documented rates of response and sizes of residual masses and Ga avidity of residual masses. The results of early (post-cycle 2) and final (post-cycle 4) restaging were subsequently associated with clinical outcome. RESULTS CT-documented rates of response and residual mass sizes were indistinguishable in complete responders who remained continuously disease-free (CR-Cont), complete responders who subsequently relapsed (CR-Rel), and partial responders who then progressed (PR/Prog). In marked contrast, early restaging (post-cycle 2) Ga scans accurately delineated these three categories of patients: CR-Cont 90% Ga-negative (18 of 20 patients) versus CR-Rel 25% Ga-negative (one of four patients) versus PR/Prog 0% Ga-negative (zero of six patients) (P = .000014). At a median follow-up duration of 31 months (range, 21 to 46), 94% of patients who had negative early restaging Ga scans remain free from progression (FFP), whereas only 18% of patients who had positive early restaging Ga scans remain FFP (P = .000007). Early restaging Ga scans were more predictive for FFP than final restaging Ga scans because patients who required four full cycles of therapy to become Ga-negative were more likely to develop recurrent disease. CONCLUSION Early restaging Ga scans delineate patients who are likely to have prolonged disease-free survival from those who fail to respond to intensive induction therapy. Patients whose tumors remain Ga-positive midway through high-dose CHOP therapy have a poor outcome and may be candidates for alternative treatment.

Dynamic Infrared Imaging of Newly Diagnosed Malignant Lymphoma Compared with Gallium-67 and Fluorine-18 Fluorodeoxyglucose (FDG) Positron Emission Tomography

Staging and therapy monitoring of malignant lymphomas relies heavily on imaging using arbitrary size criteria from computed tomography (CT) and sometimes non-specific radionuclide studies to assess the activity of the disease. Treatment decisions are based on early assessment of the response to therapy and the residual volume of the disease. Our initial experience is reported using a new noninvasive, inexpensive, and reproducible passive imaging modality, Dynamic Infrared Imaging (DIRI), which may add a new dimension to functional imaging. This system relies on its ability to filter the raw infrared signal using biological oscillatory behavior. It detects and analyzes minute oscillations of temperature and heat distribution in tumors.

201Tl/99mTc-HMPAO SPECT imaging of treated childhood brain tumors

To assess whether thallium-201 thallous chloride (Tl) can detect childhood tumors and whether diagnostic effectiveness improves with combined blood flow imaging, 28 children (1.0-18.6 years) were studied using single photon emission computed tomography (SPECT): Tl (1.3-1.8 mCi intravenously), followed in 13 of the patients by technetium-99m-hexamethylpropyleneamine oxime (99mTc-HMPAO; 8-18 mCi intravenously). Tl-uptake was markedly increased with histologically confirmed recurrent brain tumors (N = 12). Tl-avid tumors comprised several histologic types, including 6 astrocytomas/gliomas as well as nonastrocytic neoplasms, such as medulloblastoma and ependymoma. A questionable false-positive study was observed with a treated medulloblastoma. Tl failed to detect 5 tumors (i.e., 2 medulloblastomas, 1 ependymoma, 1 malignant schwannoma, and initially 1 low-grade astrocytoma). The sensitivity and specificity of 201Tl-SPECT for detection of childhood brain tumors was 76.9% and 93.3%, respectively. The mean tumor-to-normal brain ratio for Tl was 2.5 +/- 0.5 (N = 7). In some of the patients, 201Tl-SPECT allowed a more precise assessment of the functional state of the tumor than was possible with computed tomography and magnetic resonance imaging. HMPAO distribution was variously normal, increased or decreased at the site of tissue abnormality, and abnormal blood flow was demonstrated in the remaining neuraxis, in 3 of the 7 patients. Changes in tissue perfusion did not correlate with Tl findings, but were evaluated in only one false-negative study.

Monitoring therapeutic response in skeletal metastases using dual-energy x-ray absorptiometry: a prospective feasibility study in breast cancer patients.

Response to systemic therapy in breast cancer patients with lytic skeletal metastases manifests as a shift from increased bone resorption to new bone formation. We hypothesized that dual-energy x-ray absorptiometry (DXA) could be used to prospectively quantitate changes in bone mineral density (BMD) in metastatic skeletal lesions in breast cancer patients receiving systemic therapy. Nine metastatic breast cancer patients with one or more assessable lytic skeletal metastases receiving systemic therapy were prospectively evaluated with DXA, skeletal radiographs, computed tomography (CT), and radionuclide bone scans at baseline (t = 0 months, 2 months, and 6 months). The median (range) percentage change in BMD in skeletal lesions among patients responding to systemic therapy was 10.7% (0.1-21.85), 5.0% (-1.3-23.8), and 16.7% (-2.0-50.8) at 0-2, 2-6, and 0-6 months, respectively. Changes in BMD between 0-2, and 0-6 months were significant (Wilcoxin signed rank test; p = 0.013 and p = 0.017, respectively). The percentage change in BMD skeletal lesions between 0-2 and 2-6 months correlated with the changes imaged on skeletal x-rays (Spearman rank order correlation coefficient [Rs] = 0.511, p = 0.011) and CTs (Rs = 0.416, p = 0.046) but less so with bone scans (Rs = 0.293, p = 0.189). It is technically feasible to use DXA to prospectively monitor changes in lytic skeletal metastases in breast cancer patients receiving systemic therapy. The BMD of skeletal metastases increases in patients responding to treatment and was significantly correlated with the changes imaged on skeletal x-rays and CTs. Additional studies of DXA to evaluate response in skeletal metastasis are warranted.

Scintigraphic and radiographic patterns of skeletal metastases in breast cancer: value of sequential imaging in predicting outcome.

ObjectiveTo determine whether temporal changes in scintigraphic and bone radiographic findings have prognostic significance in patients with skeletal metastases from breast cancer.DesignClinical information and films were retrospecitvely reviewed in 101 randomly selected patients with skeletal metastases. Images from sequential bone scans and bone radiographs were correlated with survival after detection of the metastases.ResultsTime to detection of skeletal metastases and the length of time for which patients were classified as radiologically stable after development of skeletal metastases correlated with survival (r=0.843; r=0.821, respectively). Failure to develop a radiographically and scintigraphically stable pattern after treatment was associated with significantly decreased survival compared with the rest of the patients (mean survival 2.1±1.3 years vs 4.3±2.3 years; p<0.001). Scintigraphic regression of metastases was associated with significant survival benefit and longer stabilization of disease compared to all other patterns (mean survival 5.0±2.7 years for regressive disease vs 3.7±1.9 years for stable disease and 2.2±1.3 years for progressive disease; p<0.001).ConclusionsSequential scintigraphic and radiographic imaging is useful in breast cancer patients not only for detection of metastases and monitoring of treatment effect, but also because these studies provide valuable prognostic information.To determine whether temporal changes in scintigraphic and bone radiographic findings have prognostic significance in patients with skeletal metastases from breast cancer. Clinical information and films were retrospecitvely reviewed in 101 randomly selected patients with skeletal metastases. Images from sequential bone scans and bone radiographs were correlated with survival after detection of the metastases. Time to detection of skeletal metastases and the length of time for which patients were classified as radiologically stable after development of skeletal metastases correlated with survival (r=0.843; r=0.821, respectively). Failure to develop a radiographically and scintigraphically stable pattern after treatment was associated with significantly decreased survival compared with the rest of the patients (mean survival 2.1±1.3 years vs 4.3±2.3 years; p<0.001). Scintigraphic regression of metastases was associated with significant survival benefit and longer stabilization of disease compared to all other patterns (mean survival 5.0±2.7 years for regressive disease vs 3.7±1.9 years for stable disease and 2.2±1.3 years for progressive disease; p<0.001). Sequential scintigraphic and radiographic imaging is useful in breast cancer patients not only for detection of metastases and monitoring of treatment effect, but also because these studies provide valuable prognostic information.

PROGRESSIVE SCLEROSIS OF ISOLATED FOOT METASTASIS OF PROSTATE CANCER

An 82-year-old man presented with foot pain. Three years previously gross hematuria had developed, and serum prostate specific antigen (PSA) level at that time was 11.3 ng./ml. Initial prostate needle biopsies demonstrated a Gleason score of 4 4 8 disease, and the patient underwent external beam radiation therapy. Posttreatment PSA reached a nadir of 0.2 ng./ml. and remained at that level until 1 year prior to the current presentation, at which time pain developed in the right foot. Initial bone scan showed focal uptake in the right cuboid bone (fig. 1, A), and plain films revealed osteopenia of the cuboid bone, but no fracture or sclerosis (fig. 2, A). Magnetic resonance imaging (MRI) demonstrated abnormalities consistent with osteomyelitis or edema, but metastases were thought unlikely. The patient underwent repeat bone scans 6 months and 1 month prior to the current presentation, neither of which showed new lesions. However, followup plain films of the right foot demonstrated progressive sclerosis of the cuboid bone (fig. 2, B and C). A subsequent MRI revealed replacement of the bone marrow by bright signal on T2-weighted images, consistent with tumor (fig. 1, B). A repeat PSA test revealed an increase to 0.5 ng./ml. Needle biopsy of the lesion demonstrated adenocarcinoma cells, which stained focally positive for PSA. The patient underwent radiotherapy to the foot at that time, but experienced swelling and a subsequent ulcer in the region of the biopsy. When the patient presented to us the foot pain and swelling had worsened. PSA level at presentation was 0.6 ng./ml. and prostatic acid phosphatase was 0.9 ng./ml. Hormone ablation therapy was instituted, and within 6 weeks the foot pain improved to the extent that the patient discontinued narcotics.

Anatomic differences after robotic-assisted radical prostatectomy and open prostatectomy: implications for radiation field design

PURPOSE To investigate the anatomy of the pelvis following robotic-assisted radical prostatectomy (RARP) compared to the anatomy of the pelvis following open prostatectomy (OP), and to determine if postoperative radiation field design should take surgical approach into consideration. METHODS AND MATERIALS This report is a retrospective review of the postoperative pelvic magnetic resonance imaging (MRI) scans for all OP patients (10) and all RARP patients (15) who presented consecutively to the radiation oncology clinic and subsequently underwent MRI scanning between January 2007 and December 2008. All patients who presented are included in the study. We measured 13 distinct anatomic distances, and we used t tests to examine mean differences in each of the parameters between RARP and OP and analysis of variance to examine mean differences controlling for length of follow-up MRI postsurgery (in days) and body mass index as covariates. RESULTS Of the measurements, we found that the superior levator separation is statistically significantly greater in the post-RARP group than in the post-OP group (P < .01). Similarly, the post-RARP group had a greater mean resection defect measurement (P = .01) as measured by a larger width of the bladder infundibulum. This suggests that the size of trigonal musculature defect is more pronounced after RARP. The total urethral length was statistically significantly longer in the RARP group (P = .03). The vesicorectal distance was variable depending on the location along the rectal wall but trended toward larger separation in the post-RARP group (P = .05). CONCLUSIONS The pelvic anatomy after RARP is considerably different from that after OP. The current standard field design for post-prostatectomy radiation is defined by the post-OP pelvis. Our data support that the clinical target volume borders be expanded posteriorly and laterally in men who have undergone RARP. As RARP continues to become a more widespread surgical option for the management of localized prostate cancer, radiation field design may need to be adjusted.