Traci Leong

Prognostic value of serum markers of bone metabolism in untreated multiple myeloma patients.

Bone involvement is a central feature of multiple myeloma (MM). We investigated whether serum markers of osteoblastic and osteoclastic activity correlate with the presence of bone disease and survival in 313 MM patients enrolled in a phase III trial (E9486). Five markers were measured, including osteocalcin (OC), carboxy‐terminal propeptide of type I collagen (PICP), bone alkaline phosphatase (BAP), carboxy‐terminal telopeptide of type I collagen (ICTP) and tartrate‐resistant acid phosphatase (TRAP). We analysed the relationship between serum levels of these markers and the presence of bone manifestations, and survival. Serum levels of ICTP and BAP correlated significantly with bone pain, lesions and fractures. Serum level of ICTP was also higher in stage II–III compared with stage I disease. The serum level of ICTP was significantly associated with shortened survival in the univariate analysis. The median survival times were 4·1 and 3·5 years for low and high ICTP respectively (P = 0·02). There was a strong relationship between ICTP and beta‐2‐micrgolobulin (B2M). ICTP stands out as a significant marker of bone disease. Incorporation of these markers into clinical trials assessing the use of bisphosphonates in MM is needed to determine whether they might serve as indicators of effectiveness of these agents.

Childhood optic chiasm gliomas: Radiographic response following radiotherapy and long-term clinical outcome

PURPOSE In children with chiasmal gliomas, radiation therapy can arrest progressive visual and neurologic impairment. We examined the radiographic response and clinical outcomes after irradiation. METHODS AND MATERIALS Forty-two children (median age at diagnosis, 6.6 years) with chiasmal gliomas were managed as follows: 11 asymptomatic patients with neurofibromatosis-1 (NF-1) were observed only; 2 patients, less than 3 years old, underwent surgery and chemotherapy to delay irradiation; and 29 patients with progressive disease received radiation with or without prior surgery or chemotherapy. Time to radiographic response, long-term tumor control and late sequelae were reviewed for the 29 irradiated patients. RESULTS The probability of at least 50% radiographic response at 24 months after irradiation was 18.1% and increased to 38.2% by 48 months and 45.9% by 60 months. By actuarial analysis, the median time for such radiographic response was 62 months. For the 29 irradiated patients, the 10-year freedom from progression and overall survival rates were 100% and 89%, respectively (median follow-up for surviving patients, 108 months). Stabilization or improvement in vision occurred in 81% of 26 evaluable irradiated patients. CONCLUSIONS Notable radiographic response may be observed years after irradiation. Radiation therapy provides excellent long-term tumor control and vision preservation or improvement in the majority of patients with progressive chiasmal gliomas.

Is flow cytometric DNA content hypodiploidy prognostic in multiple myeloma

Hypodiploid multiple myeloma is uncommon when assessed by DNA content flow cytometry, having been reported in less than 6% of patients with newly diagnosed multiple myeloma. Previous studies have shown these patients to be unresponsive to therapy and to have short survival. To address this further, we studied 349 of 504 patients eligible for Eastern Cooperative Oncology Group (ECOG) treatment trial E9486 and laboratory correlative study E9487 who had marrow mononuclear cells available for ploidy analysis. Marrow samples were studied by dual channel flow cytometry, using propidium iodide to measure the DNA content and kappa and lambda light chain antisera to identify the clonal cells. A DNA index < 0.95 was considered hypodiploid. Five patients (1.4%) were found to have hypodiploid DNA content in their marrow plasma cells. Three of the 5 patients with hypodiploid myeloma had a partial objective response to chemotherapy, which is not different from the overall objective response rate for all patients enrolled on E9486. All five patients with hypodiploid multiple myeloma died within 4 years from diagnosis, but these patients had a similar overall median survival (2.6 years) compared to the patients with diploid DNA content. Our studies confirm the poorer survival of patients with diploid versus hyperdiploid myeloma; we cannot confirm, however, the previously reported very poor outcome associated with hypodiploid myeloma using DNA content flow cytometry. Hypodiploid DNA content of plasma cells by flow cytometry may not be as ominous a factor as previously reported.

Altered T cell repertoire usage in CD4 and CD8 subsets of multiple myeloma patients, a Study of the Eastern Cooperative Oncology Group (E9487).

Previous investigations have demonstrated that an expanding circulating T cell population is able to modulate the malignant clone in multiple myeloma. More recently, an expansion of T cell subsets exhibiting a restricted T cell repertoire has been detected in some MM patients. To further elucidate if a selected T cell expansion occurs in MM, we studied the T cell receptor (TCR) variable (V) region expression from a cohort of previously diagnosed and treated MM patients (N=37). The latter was done by assessing the reactivity of a panel of monoclonal antibodies specific for different V region families (alpha or beta) in combination with anti-CD4 or anti-CD8, for purified blood T cells from MM patients. TCR V region usage in MM patients was compared to blood T cells from age matched (N=13) control individuals. The multivariate analysis of variance did not uncover a difference for distribution of TCR V region usage between the normal controls and the MM cohort. However, there were individual MM patients who had expanded T cells with specific TCR V region expression when compared to the control group. Several MM patients had multiple, expanded CD4 and/or CD8 subsets based on TCR V region expression. The majority of MM patients had expanded T cell subsets that constituted less than 10% of the total blood T cell pool. However, a few MM patients (N=3) had larger percentages (range 34-84%) of these expanded T cell subsets within their blood T (CD3+) cells. The stage of disease and treatment status (currently on or off therapy) did not associate with the pattern of restricted T cell repertoire. Finally, a smaller cohort of newly diagnosed, untreated MM patients (N=13) also demonstrated an expanded T cell repertoire. However, these patients had more CD4 than CD8 cell subsets involved in the altered V region expression in several Vbeta families. Thus, these results add to the evidence that this malignant B cell disorder whether newly diagnosed or of longer duration, may be accompanied by an altered T cell repertoire characterized in part by expanded T cell clones.

Seizures in Children with Supratentorial Astroglial Neoplasms

We reviewed the records of 98 consecutive patients, 18 years of age or younger, with pathologically confirmed supratentorial astroglial neoplasms at the Childrens Hospital, Boston, to evaluate the importance of seizures in their presentation and natural history. Tumors were diagnosed using the WHO criteria as pilocytic astrocytomas, astrocytomas, anaplastic astrocytomas, glioblastomas, giant cell glioblastomas, oligoastrocytomas and gangliogliomas. Our results were as follows. (1) Fifty percent of children had seizures as part of their presentation and 30% had seizures as their only presenting phenomenon. (2) The seizures varied in nature, but complex (55%) or simple (28%) partial seizures were by far the most common types, occurring in 77% of cases. (3) Preoperative electroencephalography (EEG) accurately lateralized to the tumor side in 88% of the cases and localized to the correct lobe in 56%. (4) Tumors involving cerebral cortex significantly correlated with seizures at presentation as compared to noncortical locations; 59% of patients with cortical tumors presented with seizures, and only 15% of patients with noncortical tumors experienced seizures. Lesions in the temporal and frontal lobes had the highest incidence of seizures. (5) Patients with gangliogliomas and oligoastrocytomas had the highest incidence of seizures, 88 and 86%, respectively, whereas patients with anaplastic astrocytoma had the lowest incidence, 21%. (6) Histopathologic calcification was associated with seizures at presentation approaching significance (p = 0.06). (7) Seizures at presentation were significantly associated with good prognosis (p = 0.02).

T-helper phenotypes in the blood of myeloma patients on ECOG phase III trials E9486/E3A93

T‐helper blood populations are frequently altered in multiple myeloma (MM). We measured the numbers of naive and activated cell subsets in the blood of a cohort of both previously untreated and treated MM patients. Two‐colour flow cytometry to detect total CD4+, CD4+, CD45RA+ (naive) and CD4+, CD45RO+ (activated) subsets was then used to quantify the T‐cell subsets in controls and MM patients. Previously treated MM patients either on or off treatment (n = 105) had significantly reduced (P < 0.0001) total CD4 and naive/activated cells than controls. Previously treated MM patients sampled for naive/activated cells while currently off therapy (n = 45) had no difference in the levels of CD4 and naive/activated cells compared to the currently treated patients (n = 60). However, newly diagnosed patients (n = 58) had a significantly reduced total CD4 (P = 0.023) and activated CD4 (P = 0.004), but not naive CD4 subsets, compared to controls. CD19+ cell levels above 125/μl were positively associated with higher T‐helper cell levels. There was a strong positive association for better overall survival for patients with > 395 CD4 cells/μl (P = 0.0001). These data indicate that MM patients at diagnosis have altered T‐helper subsets, with a selective reduction in activated but not naive cells. Subsequent chemotherapy or the disease process contributes to a further reduction in CD4 cells. Importantly, the association of higher CD19+ cell levels with higher T helper cells indicates that certain myeloma patients can be identified with a more quantitatively intact immune system.

Prognostic significance of the S-phase fraction of light-chain-restricted cytoplasmic immunoglobulin (cIg) positive plasma cells in patients with newly diagnosed multiple myeloma enrolled on eastern cooperative oncology group treatment trial E9486†

The bone marrow plasma cell labeling index (PCLI) as measured by bromodeoxyuridine uptake is a well‐established independent prognostic factor for patients with newly diagnosed multiple myeloma, but the test is not easily done in most laboratories. The purpose of this study was to determine if the proliferative activity (% S‐phase) as determined by two‐color flow cytometry for cytoplasmic immunoglobulin (cIg) light chain and DNA content also had prognostic significance. As part of Eastern Cooperative Oncology Group clinical trial E9486, 500 patients had successful performance of the bone marrow PCLI. Of 349 patients who had flow cIg and DNA content cytometry, 210 had adequate data to reliably calculate S‐phase %. Patients with low % S‐phase fraction (<2%) had a significant overall survival advantage over patients high % S‐phase fraction (≥2%), median survivals 4.1 vs. 2.9 years (P = 0.032). Measurement of the S‐phase % by flow cytometry gives significant prognostic information in patients with newly diagnosed myeloma. However, in multivariate analysis, S‐phase % did not add prognostic information when PCLI was in the model. S‐phase % added prognostic information only when all cases with flow measurement of S‐phase % were included, and when PCLI was excluded from the model. Discriminating a population of only cIg positive cells proved difficult in patients with a low percentage of bone marrow plasma cells. Methodology to measure S‐phase % in patients with a low percent plasma cells is needed before this technique can be used for diagnosis and prognosis in myeloma. Am. J. Hematol. 61:232–237, 1999.

Incomplete covariates in the Cox model with applications to biological marker data

A common occurrence in clinical trials with a survival end point is missing covariate data. With ignorably missing covariate data, Lipsitz and Ibrahim proposed a set of estimating equations to estimate the parameters of Coxs proportional hazards model. They proposed to obtain parameter estimates via a Monte Carlo EM algorithm. We extend those results to non-ignorably missing covariate data. We present a clinical trials example with three partially observed laboratory markers which are used as covariates to predict survival.

Treatment of adult patients with acute lymphocytic leukemia in relapse

Thirty-eight patients with a diagnosis of relapsed acute lymphocytic leukemia were accrued to a treatment program of reinduction therapy by the Eastern Cooperative Oncology Group (ECOG). A combination of mitoxantrone, etoposide (VP-16), and high-dose cytarabine (ARA-C) were administered over a five day period. Thirty-four patients were eligible for follow-up subsequent to treatment. Twenty-seven patients were in first relapse and seven were in second relapse. Fifteen of the thirty-four patients treated were given two cycles of induction chemotherapy. The complete remission (CR) rate for the entire group treated was 17%. The median duration of the CR was 2.4 months and the estimated median survival for first relapse patients was 4.5 months and 5.0 months for second relapse patient group. There were five deaths attributable to toxicity associated with the chemotherapy. The study emphasizes the difficulty in achieving durable remissions in adult patients with relapsed ALL.

Sequential Phenotyping of Myeloma Patients on Chemotherapy: Persistence of Activated T-cells and Natural Killer Cells

To better detail the status of functional T cell subsets and natural killer cells in multiple myeloma, we undertook a detailed immunophenotypic study of circulating mononuclear cells in myeloma. We studied myeloma patients entered on a large prospective, randomized ECOG chemotherapy trial EST 9486 for patients with newly diagnosed multiple myeloma. All patients were studied prior to entry and then two months after initiation of therapy (e.g. post two cycles of Vincristine, BCNU, melphalan, cyclophosphamide and prednisone (VBMCP)). The chemotherapy protocol was a three-arm protocol utilizing either VBMCP, VBMCP alternating with interferon, or VBMCP with intermittent high dose cyclophosphamide. The major findings in this analysis include significant reductions in the white blood cell count, total lymphocytes, T cell (CD3+), T helper (CD4+), and T suppressor (CD8+) cells, after 2 cycles of VBMCP. However, there was a relative sparing of Natural killer (CD16+) and activated T cell (CD2+, HLADR+) reduction in these same patients. In summary, only two cycles of combination chemotherapy resulted in significant reductions in white blood cell and lymphocyte counts in multiple myeloma patients. All cell types appear to have been reduced by chemotherapy except for activated T cells and natural killer cells. The impact of selective modulation of functional T cells subsets during therapy for patients with multiple myeloma is an important parameter which needs to be addressed in the overall approach to these patients.