Use of PET/CT for staging and radiation therapy planning in patients with non-small cell lung cancer.

Positron emission tomography (PET) and more recently PET/computed tomography (CT) scanning represent major advances in the imaging of lung cancer and have an especially high impact on the management of patients who are candidates for potentially curative or "radical" radiotherapy (RT). This article reviews the current status of PET and PET/CT for staging patients before RT and considers the use of PET and PET/CT images for target volume definition. The relevant literature on the use of PET for staging lung cancer is reviewed and placed in the context of patients who are candidates for RT. Research that specifically considers the use of PET for RT planning is considered critically and some promising areas for future research are discussed. The available literature is almost exclusively devoted to non small cell lung cancer (NSCLC) with few relevant studies of small cell lung cancer (SCLC). The primary PET radiopharmaceutical shown to have value for staging and RT planning is 18F-fluorodeoxyglucose (FDG). In prospective studies where PET imaging was used to stage radical RT candidates, 25-30% of patients were excluded from radical therapy because of PET detected advanced disease. In all studies where "PET-assisted" and conventional target or treatment volumes were compared, there were major differences between PET and conventional volumes. Because PET-assisted staging is proven to be significantly more accurate than conventional staging and because all studies show major differences between PET-assisted and conventional treatment volumes in NSCLC, routine use of PET/CT for RT planning is recommended.

Clinical influence of 18F-fluorodeoxyglucose positron emission tomography on the management of primary tumours of the thymus.

The objective of the current study was to evaluate the effect of PET on the management of primary tumours of the thymus. Patients with a primary tumour of the thymus who underwent PET were identified from a prospective database. Forty-three PET scans were carried out on 26 patients with primary thymic tumours. Sensitivity, specificity and accuracy were 79, 100 and 85%, respectively. Conventional imaging and PET findings were discordant in 10 cases (23%). PET appropriately changed patient management in three (7.0%) cases based on accurate results that differed from pre-PET imaging. Most PET scans carried out (88%) did not influence clinical management. Patient comorbidities, limited treatment options and already planned surgery are factors that may hinder the effect of PET in the setting of thymic tumours. The potential for false-negative results, probably because of a combination of low-fluorodeoxyglucose avidity and small volume residual disease, needs to be considered in management planning. However, the positive predictive value of PET is high and enables appropriate modification of management in a small, but potentially important subset of patients.

Dose escalation of radical radiation therapy in non-small-cell lung cancer using positron emission tomography/computed tomography-defined target volumes: are class solutions obsolete?

This study investigated the maximum theoretical radiation dose that could safely be delivered to 20 patients diagnosed with non-small-cell lung cancer. Two three-dimensional conformal radiation therapy (RT) class-solution techniques (A and B) and an individualized three-dimensional conformal RT technique (C) were compared at the standard dose of 60 Gy (part I). Dose escalation was then attempted for each technique successfully at 60 Gy, constrained by predetermined limits for lung and spinal canal (part II). Part I and part II data were reanalysed to include oesophageal dose constraints (part III). In part I, 60 Gy was successfully planned using techniques A, B and C in 19 (95%), 18 (90%) and 20 (100%) patients, respectively. The mean escalated dose attainable for part II using techniques A, B and C were 76.4, 74 and 97.8 Gy, respectively (P < 0.0005). One (5%) patient was successfully planned for 120 Gy using techniques A and B, whereas four (20%) were successfully planned using technique C. Following the inclusion of additional constraints applied to the oesophagus in part III, the amount of escalated dose remained the same for all patients who were successfully planned at 60 Gy apart from two patients when technique C was applied. In conclusion, individualized three-dimensional conformal RT facilitated greater dose conformation and higher escalation of dose in most patients. With modern planning tools, simple class solutions are obsolete for conventional dose radical RT in non-small-cell lung cancer. Highly individualized conformal planning is essential for dose escalation.

Positron emission tomography of stage IV mucosa-associated lymphoid tissue lymphoma confined to the four major salivary glands.

In a patient with stage IVA marginal zone lymphoma, (18)F-fluorodeoxyglucose-positron emission tomography indicated that the disease was confined to the four major salivary glands. The positron emission tomography findings encouraged the use of radiotherapy with curative intent in this case. After 30 Gy of external beam radiotherapy to the parotid and submandibular glands, the patient entered a complete remission and remains free from progression more than 4 years later.

Improving radiation therapy for non-small cell lung cancer: molecular imaging and a team-based approach.

The successful integration of molecular imaging and radiation therapy has been shown to significantly impact the management of patients with non-small cell lung cancer (NSCLC). The collaboration of multidisciplinary team members, including radiation oncologists, radiation therapists, nuclear medicine physicians and physicists, has enabled PET/CT to be utilised for routine use throughout the radiotherapy treatment trajectory. Applications include disease diagnosis and staging, target volume definition for radiation therapy and monitoring tumour response to treatment. Not only has the adoption of this technology demonstrated benefits for our current patients, it is also opening doors for significant research in the future.

Failure of T stage to predict survival in patients with non-small-cell lung cancer treated by radiotherapy with or without concomitant chemotherapy.

PURPOSE: Because T stage does not consistently reflect tumor size in non-small-cell lung cancer (NSCLC), we hypothesized that T stage may be of limited prognostic value in patients with locoregional NSCLC treated by nonsurgical means. METHODS AND MATERIALS: The study population consisted of 243 patients with histologically or cytologically proven NSCLC treated in three consecutive prospective trials between 1989 and 1998. The eligibility criteria for this analysis included planned for and began treatment at 60 Gy; Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1; weight loss < or = 10%; no prior treatment; and no supraclavicular nodes, pleural effusion, or distant metastases. In the first study, 204 patients were randomized to receive conventional or accelerated radiotherapy (RT) with or without concomitant carboplatin. In the second, 15 patients were treated with concomitant cisplatin, etoposide, and RT in a single-arm study. In the third, 24 patients were treated with concomitant carboplatin, 5-fluorouracil, and RT in a dose-escalation study. RESULTS: A total of 231 patients for whom the T and N stage were known met the eligibility criteria. The patient characteristics were 77% male, 64% squamous histologic features, 33% ECOG status of 0, and 69% no weight loss. The nodal status was 36% N0, 7% N1, 52% N2, and 5% N3. The estimated median survival for all patients was 1.4 years (95% confidence interval 1.2-1.6), with an estimated 10% surviving 5 years (95% confidence interval 7-15). No significant difference was found in survival among the three trials (p = 0.16). The estimated median survival time and 5-year survival rate according to T stage were as follows: T1 (n = 29), 1.6 years and 16%; T2 (n = 88), 1.3 years and 9%; T3 (n = 59), 1.4 years and 9%; and T4 (n = 55), 1.4 years and 9%. No significant trend was found in overall survival according to T stage (p = 0.85, log-rank). To test whether a significant effect of T stage on overall survival existed after adjusting for N stage, trial, ECOG status, and weight loss, a multifactor analysis using Cox proportional hazards regression analysis was carried out. There was still no significant effect of T stage on survival (p = 0.66) when all factors were taken into account. CONCLUSION: Although there is some evidence that T stage is an independent prognostic factor in patients with NSCLC treated surgically, it did not appear to be of value in this series of patients treated with RT with and without concomitant chemotherapy.

FDG-PET imaging in the management of non-small-cell lung cancer.

Lung cancer is currently the leading cause of cancer-related death in both men and women in most Western countries. Tumour stage is the strongest prognostic factor and the most important parameter guiding treatment decision making. Metabolic positron emission tomography imaging with fluorodeoxyglucose (FDG-PET) has consistently proved superior to conventional imaging for staging of non-small-cell lung cancer and provides information of greater prognostic significance than can be obtained using conventional approaches. FDG-PET has been approved in the USA, Germany and the UK as a basic and invaluable tool in the management of lung cancer.

Exacerbation, then clearance, of mutation-proven Darier's disease of the skin after radiotherapy for bronchial carcinoma: a case of radiation-induced epidermal differentiation?

We investigated a radiotherapy-induced flare and subsequent clearance of skin lesions of a patient with the rare, dominantly inherited genodermatosis, Darier's disease (DD). The DD gene, ATP2A2, was recently isolated and shown to be a cation pump responsible for regulating intracellular calcium homeostasis. A severe exacerbation of Darier's skin lesions developed within the radiation field when 40 Gy of palliative thoracic external-beam radiation therapy and concurrent chemotherapy (cisplatin and hydroxyurea) were delivered for non-small cell lung cancer. The DD lesions subsequently completely cleared from irradiated skin, as they did when a subsequent course of radiation alone was given for a loco-regional tumor recurrence. The two radiation therapy-treated areas of skin remained free from lesions of the skin disorder until the patient's death from progressive lung cancer 9 months later. The nucleotide sequence of the patient's ATP2A2 gene was determined by PCR-based cycle sequencing. We identified four nucleotide sequence variants in the ATP2A2 gene in this patient. Three were probable polymorphisms and the other appeared to be a novel disease-causing mutation (R751Q), situated in the transmembrane portion of the ATP2A2 protein. This finding confirmed the clinical diagnosis. Since epidermis turns over every 3-4 weeks, total and persistent clearance of the DD lesions by chemoradiotherapy suggests that this treatment induced sustained differentiation of the DD-affected skin by an unknown mechanism. Oncologists treating malignant disease in patients with DD should anticipate temporary deterioration in DD-involved irradiated skin. Radiation therapy has therapeutic potential in severe DD.

Impact of FDG-labelled positron emission tomography imaging on the management of non-small-cell lung cancer.

Lung cancer is currently the leading cause of cancer-related death in both men and women in most Western countries. Although 5-year survival rates have doubled from 1960s, they are low compared with survival rates for other cancer types. Tumour stage is the strongest prognostic factor and the most important parameter that guides treatment decision making. Metabolic imaging with fluorodeoxyglucose-labelled positron emission tomography (FDG-PET) has proved superior to conventional imaging for staging of non-small-cell lung cancer. This new imaging modality permits more accurate planning of treatment with surgery and radiotherapy and provides information of greater prognostic significance than what can be obtained with conventional approaches. FDG-PET has been approved in the USA and the UK as a basic and invaluable tool in the management of lung cancer. This paper reviews current trends in clinical practice on the applications of FDG-PET in the management of non-small-cell lung cancer.

F-18 fluorodeoxyglucose positron emission tomography staging in radical radiotherapy candidates with nonsmall cell lung carcinoma: powerful correlation with survival and high impact on treatment.

BACKGROUND: Successful treatment of nonsmall cell lung carcinoma (NSCLC) with radical radiotherapy (RT) requires accurate delineation of tumor extent. Conventional computed tomography-based noninvasive staging often estimates intrathoracic thoracic tumor extent incorrectly and fails to detect distant metastasis. High sensitivity and specificity are reported for F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) staging in potentially resectable NSCLC. The authors investigated FDG-PET staging in radical RT candidates with unresectable NSCLC. METHODS: The authors prospectively studied 153 consecutive patients with unresectable NSCLC who were candidates for radical RT after conventional staging and had PET scans. Patients were allocated both "before PET" and "after PET" stages. Subsequent management was recorded. Survival analysis was used to compare validity of pre-PET and post-PET staging. RESULTS: After PET, 107 patients (70%) actually received radical therapies (radical RT with or without concurrent chemotherapy, n = 102; radical surgery, n = 5); 46 patients (30%) received palliative treatment because of PET-detected distant metastasis (n = 28; 18%) or extensive locoregional disease (n = 18; 12%). Palliative therapies were RT (n = 33), chemotherapy (n = 12), or supportive care (n = 1). All five surgically treated patients underwent potentially curative resections after downstaging by PET. For radically treated patients, post-PET stage (P = 0.0041) but not pre-PET stage (P = 0.19) was strongly associated with survival. Radically treated patients survived longer than those treated palliatively (P = 0.02; 1-year survival, 69% and 44%, respectively; 2-year survival, 44% radical; no palliative patients had 2-yr follow-up). CONCLUSIONS: Positron emission tomography-assisted staging detected unsuspected metastasis in 20%, strongly influenced choice of treatment strategy, frequently impacted RT planning, and was a powerful predictor of survival. Potential impact of FDG-PET is even greater in radical RT candidates with NSCLC than in surgical candidates.

Combined modality treatment using concurrent radiotherapy and pharmacologically-guided carboplatin for inoperable and incompletely resected non-small cell lung cancer.

BACKGROUND: In our previous randomised trial, radiotherapy (RT) was given concurrently with carboplatin 350 mg/m(2). We wanted to show that the safety and efficacy of the drug could be improved by pharmacologically-guided dosing based on renal function. PATIENTS AND METHODS: Patients were eligible if they had unresectable or incompletely resected NSCLC, good performance status (ECOG 0-2), weight loss < 10%, no distant metastases and adequate haematology and biochemistry. Radiotherapy was given to the primary site and regional lymph nodes to a total dose of 60 Gy in 30 fractions over 6 weeks. Two cycles of carboplatin were given in divided doses of 1-h infusions daily for 5 days before RT weeks 1 and 6. A total plasma AUC of 7 mg/ml per minute per cycle was targeted. The total dose was calculated by using Calvert or Chatelut formulae. RESULTS: Forty-nine patients were treated. Patient characteristics included: 78% male; mean age 66 (range: 38--78); 80% stage 3A or 3B; incomplete resection in six patients. The median dose of carboplatin administered per cycle was 850 mg (range 435--1650); 89% of patients received a higher carboplatin dose compared with BSA-calculated dose (mean increase 41%). Forty-two patients (86%) completed treatment as planned. Myelosuppression > or = grade 3 occurred in 14 patients (29%) (one patient died of pneumonia while neutropenic); two patients developed > or = grade 3 acute oesophagitis and two patients had > or = grade 3 acute pulmonary toxicity. Late pulmonary toxicity > or = grade 3 occurred in two patients. The mean potential follow-up time was 2.7 years. The estimated proportion of patients alive and free of local or distant progression at 1 year was 42% and the median survival duration was 16 months (95% CI: 11--21 months). CONCLUSIONS: Radical chest irradiation can be combined with two cycles of pharmacologically-guided full-dose carboplatin, however because our study demonstrated significant haematologic toxicity, we recommend carboplatin dosing according to renal function at less than full dose (i.e. AUC 6 mg/ml per minute per cycle).

Characteristics of 49 patients who survived for 5 years following radical radiation therapy for non-small cell lung cancer: the potential for cure.

PURPOSE: To investigate the long-term curative potential of radical radiation therapy (RT) for non-small cell lung cancer (NSCLC) by studying characteristics of patients from a large prospective database who survived >5 years after RT, and by analyzing survival beyond 5 years. METHODS AND MATERIALS: Five-year survivors were identified from a database containing information on 488 patients given radical RT following presentation to the Peter MacCallum Cancer Institute with NSCLC between 1984 and 1990. Additional data were obtained from case notes of survivors. RT was computed tomography (CT)-planned, conventionally-fractionated, and given without chemotherapy. RESULTS: Actuarial survival for 49 5-year survivors was 65% at 10 years. Five 5-year survivors had documented disease progression within the first 5 years and subsequently died. Of 44 patients free-from-progression (FFP) at 5 years, an estimated 81% remained FFP in the second 5 years. Age and histology were not significant prognostic factors, and only 22 patients (4.5%) had weight loss >10%. For 277 patients who had not undergone thoracotomy, median RT dose was 60 Gy and survival at 5 and 10 years was 7% and 3%, respectively. For 207 patients who received radical RT post-thoracotomy, median dose was 60 Gy and survival at 5 and 10 years was 24% and 18%, respectively. Five-year survivors of post-thoracotomy RT had been treated for gross residual disease (n = 10), positive-margin (n = 6), or probable microscopic residual disease (n = 17). Failure to regain ECOG performance status = 0 post-thoracotomy was associated with reduced survival (p<0.0012). FFP in the second 5 years was superior for patients who had postoperative radiotherapy (90%) compared to patients without thoracotomy (62%, p = 0.008). CONCLUSION: Most patients FFP >5 years after radical RT for NSCLC remained FFP in the following 5 years and were apparently cured. RT alone can cure small but significant numbers of patients. Long-term results of combined chemotherapy/RT protocols, which are associated with increased median survival, are awaited for comparison.

What is the prognosis for patients who relapse after primary radiation therapy for early-stage low-grade follicular lymphoma?

PURPOSE: To investigate the potential for long-term survival for patients who relapsed after primary radiation therapy (RT) for early-stage low-grade follicular lymphoma and to assess the relative importance of prognostic factors. METHODS AND MATERIALS: Records were reviewed for 79 patients with stage I (n = 32) and II (n = 47) follicular small cleaved cell (fsc, n = 48) and follicular mixed small cleaved cell and large-cell (fmx, n = 31) lymphoma who relapsed after radical RT at Stanford University. Most patients had received doses of 35 to 45 Gy to involved (n = 30) or extended fields (n = 39) or total/subtotal lymphoid irradiation (n = 9). RESULTS: Median time to relapse was 2 years. Most relapses were detected on history (30%) or physical examination (66%). Positive relapse investigations included lymphangiogram (n = 19), chest radiograph (n = 5), and bone marrow biopsy (n = 6). Known extent of relapsed disease was: stage I, n = 30; stage II, n = 26; stage III, n = 10; and stage IV, n = 8. Patients were managed with "watchful waiting" (37%), further RT (39%), chemotherapy [CT, (17%)], or RT + CT (5%). Actuarial survival rates after relapse at 5, 10, 15, and 20 years were 56%, 35%, 17%, and 17% respectively. Median survival was 5.3 years after relapse. Median survival for relapse stage I, II, III, and IV was 10.2, 5.5, 3.0, and 1.1 years respectively. Progression-free survival rates at 5, 10, 15, and 20 years after relapse were 44%, 22%, 22%, and 22% respectively. Factors associated with reduced survival were increasing age, increasing relapse stage, symptoms, histologic transformation and > or = 3 relapse sites. Survival was the same for initial management with "watchful waiting" or RT. CONCLUSION: Approximately 20% of patients experienced prolonged survival after relapse. Younger, asymptomatic patients with stage I-II relapsed disease had the best outcome but results were inferior to those for newly diagnosed stage I-II disease.

Overview of treatment of localized low-grade lymphomas.

The results of treatment of localized low-grade lymphoma have been reviewed with particular emphasis on the results of long-term follow-up of patients treated with radiation therapy at Stanford University. These data and results from numerous other centers suggest that 40% to 50% of patients with stage I and II follicular low-grade lymphomas can expect to achieve clinical cure of their disease with radiation therapy. Randomized trials published to date do not support the use of adjuvant chemotherapy. Although a policy of initial observation with deferral of treatment until the occurrence of disease progression is a well established approach to patients with advanced disease, no randomized studies exist that support this as a safe alternative to radiation therapy for early stage disease. New systemic therapies are required for the treatment of occult disease to prevent the occurrence of relapse outside of irradiated volumes.

Radiotherapy-associated neutropenia and thrombocytopenia: analysis of risk factors and development of a predictive model.

Risk factors for unscheduled interruptions in radiotherapy courses completed between June 1989 and August 1995, lasting > or = 2 days, and associated with World Health Organization grade III-IV neutropenia or thrombocytopenia were studied retrospectively. A group of controls was randomly selected. Potential risk factors for myelosuppression were analyzed using univariate and multivariate analyses. The most important risk factors for treatment interruption with thrombocytopenia were concurrent chemotherapy (odds ratio [OR], 45.5; P < .001), increasing percentage of marrow irradiated (OR, 4.1 for each 20%; P < .001), and brain metastases (OR, 7.3; P = .01). Other significant (P < .05) factors were leukemia/lymphoma, bone or bone marrow metastases, and prior chemotherapy. The most important risk factors for treatment interruptions with neutropenia were concurrent chemotherapy (OR, 42.1; P < .001) and increasing percentage of marrow irradiated (OR, 3.3 for each 20%; P < .001). Similarly, the most important risk factors for treatment interruptions with both thrombocytopenia and neutropenia were concurrent chemotherapy (OR, 48.6; P < .001) and increasing percentage of marrow irradiated (OR, 3.9 for each 20%; P < .001). Other significant (P < .05) factors in these groups were bone marrow or brain metastases and previous chemotherapy. These data were used to create a model, assigning patients to groups at high, intermediate, or low risk for treatment interruption with thrombocytopenia. High-risk patients may be candidates for clinical trials of a platelet growth factor.

Is radiotherapy curative for stage I and II low-grade follicular lymphoma? Results of a long-term follow-up study of patients treated at Stanford University.

PURPOSE: To evaluate retrospectively the results of radiotherapy for 177 patients with stage I (n = 73 [41%]) and II (n = 104 [59%]) follicular small cleaved-cell and follicular mixed small cleaved-cell and large-cell non-Hodgkin's lymphoma (NHL) treated in the Department of Radiation Oncology, Stanford University between 1961 and 1994. PATIENTS AND METHODS: Histology was follicular small cleaved-cell in 101 (57%) cases and follicular mixed small cleaved-cell and large-cell in 76 (43%). Forty-five patients (25%) had staging laparotomy; 34 (19%) had extranodal involvement. All patients had received radiotherapy, either to one side of the diaphragm (involved or extended field) or to both sides (total lymphoid irradiation [TLI] or subtotal lymphoid irradiation [STLI]. Radiotherapy doses ranged from 35 to 50 Gy. RESULTS: The median follow-up duration was 7.7 years. The longest follow-up duration was 31 years. Actuarial survival rates at 5, 10, 15, and 20 years were 82%, 64%, 44%, and 35%, respectively. The median survival time was 13.8 years. At 5, 10, 15, and 20 years, 55%, 44%, 40%, and 37% of patients, respectively, were relapse-free. Only five of 47 patients who reached 10 years without relapse subsequently developed recurrence. Survival and freedom from relapse (FFR) were significantly worse for older patients. Relapse rates were lower following treatment on both sides of the diaphragm or staging laparotomy. Univariate analysis showed that youth and staging laparotomy were associated with significantly better survival and that FFR was better following treatment on both sides of the diaphragm or laparotomy. CONCLUSION: Radiotherapy remains the treatment of choice for early-stage low-grade follicular lymphomas. Patients who have remained free of disease for 10 years are unlikely to relapse.

Value of granulocyte colony stimulating factor in radiotherapy induced neutropenia: clinical and laboratory studies.

We report the effect of granulocyte colony stimulating factor (G-CSF) on neutropenia occurring during extended field radiotherapy in two groups of patients. The first group comprised 8 patients receiving craniospinal irradiation for a variety of central nervous system (CNS) neoplasms. None of these patients received cytotoxic chemotherapy. G-CSF was administered when the absolute neutrophil count (ANC) approached 1.5 x 10(9)/l. Neutropenia was promptly corrected in all cases, thereby avoiding unscheduled interruptions in radiotherapy. Following each G-CSF administration, ANC reached a peak on the following day and then declined steadily. Mean ANC rose from 1.33 x 10(9)/l on the day of G-CSF treatment to 7.07 x 10(9)/l the next day. Patients received 2-6 G-CSF injections during radiotherapy. Experiments were carried out in vitro to assess the risk of G-CSF causing increased CNS tumour cell proliferation. 11 human CNS tumour cultures (2 medulloblastomas, 2 primitive neuroectodermal tumours and 7 astrocytic tumours) were cultured in the presence of G-CSF at a range of concentrations up to 100 ng/ml. Their proliferation was compared with that of a G-CSF dependent murine leukemia cell line (NFS-60). None of the human tumour cultures demonstrated a significant increase in proliferation in response to G-CSF. 4 patients undergoing "mantle" type radiotherapy for Hodgkin's Disease or Non-Hodgkin's Lymphoma also received G-CSF treatment for neutropenia. All 4 had previously received cytotoxic chemotherapy. The number of G-CSF injections given per patient during radiotherapy ranged from 3-6. Mean ANC rose from 1.76 x 10(9)/l to 10.8 x 10(9)/l the next day. These results suggest that G-CSF is a reliable treatment for radiotherapy induced neutropenia and that an intermittent dosage schedule is effective.

Use of recombinant granulocyte-colony stimulating factor to treat neutropenia occurring during craniospinal irradiation.

PURPOSE: To investigate the effectiveness of recombinant human Granulocyte-Colony Stimulating Factor, a hematopoietic growth factor which stimulates neutrophil production, in the treatment of neutropenia caused by Craniospinal Irradiation. METHODS AND MATERIALS: Four consecutive patients who developed neutropenia (neutrophils less than 1.5 X 10(9)/l in peripheral blood) during craniospinal irradiation for primary intracranial tumors received intermittent subcutaneous injections of Granulocyte-Colony Stimulating Factor. Two of the patients had medulloblastoma, one had a primitive neuroectodermal tumor and the other a pinealocytoma. No patient received prior or concurrent chemotherapy. RESULTS: In all cases peripheral blood neutrophil counts returned rapidly to normal levels following Granulocyte-Colony Stimulating Factor injections and treatment delays were therefore avoided. Platelet counts were unaffected by Granulocyte Colony Stimulating Factor treatment. In one case, slight elevation of peripheral blood monocyte and lymphocyte counts occurred after each Granulocyte-Colony Stimulating Factor injection. No toxicity was encountered. CONCLUSION: Our results suggest that Granulocyte-Colony Stimulating Factor is a safe and effective treatment for neutropenia caused by extended field radiotherapy.