Current status of treatment of localized disease and future aspects☆

Pretreatment anemia is an adverse prognostic variable in squamous cell head-and-neck cancer (HNC) patients treated with radiotherapy (RT) alone. Tumor hypoxia is an adverse parameter for treatment with RT alone or with RT and concurrent chemotherapy (CCT). Tumor hypoxia is more prevalent in patients who present with pretreatment hemoglobin (Hgb) concentrations less than 13 g/dL. RT/CCT improves survival over RT alone in advanced HNC, and its use is becoming more widespread. This study was performed to evaluate whether pretreatment Hgb less than 13 g/dL was correlated with treatment outcome in patients with advanced HNC treated with a uniform regimen of RT/CCT.

The study population consisted of patients with AJCC Stage III or IV, M0 HNC who were treated with 70 to 72.5 Gy accelerated hyperfractionated RT (1.25 Gy b.i.d.) and CCT consisting of 2 cycles of CDDP (12–20 mg/m²/d × 5 days) and continuous infusion 5-FU (600 mg/m²/d × 5 days) during Week 1 and Week 6. A planned break in RT occurred during Week 4. These patients were enrolled on the experimental arm of a prospective randomized trial that compared this regimen to hyperfractionated irradiation alone from 1990 to 1996. RT/CCT was delivered as standard therapy from 1996 to 2000. The primary endpoint was failure-free survival (FFS). Secondary endpoints included local-regional control and overall survival.

One hundred and fifty-nine patients were treated from 1990 to 2000. The median (25–75%) pretreatment Hgb was 13.6 (12.2–13.5) g/dL. Hgb was 13 g/dL or higher in 105 patients and less than 13 g/dL in 54 patients. Primary tumor sites included oropharynx (43%), hypopharynx/larynx (36%), oral cavity (9%), and nasopharynx (6%). Seventy-eight percent of the patients with Hgb 13 g/dL or higher and 92% of the patients with Hgb less than 13 g/dL had a primary tumor stage of T3 or T4 (p = 0.01). Node-positive disease was present in 74 of 105 (70%) of patients with Hgb 13 g/dL or higher patients and in 36/54 (67%) of patients with Hgb less than 13 g/dL patients. Median follow-up of surviving patients was 42 months (range, 4–128 months). Five-year FFS was 75% for patients with Hgb 13 g/dL or higher vs. 50% for patients with Hgb less than 13 g/dL had a (p < 0.01). A total of 49 failures occurred in both patient cohorts. The median (25–75%) decrease in Hgb during RT/CCT was 2.2 (1.3–3.1) g/dL, both in patients who failed and in those who remained disease-free.

Pretreatment Hgb less than 13 g/dL is correlated with adverse outcomes in advanced HNC patients treated with RT/CCT. Whether anemia actually causes poor outcomes remains unknown. The therapeutic effect of anemia correction is being evaluated in prospective trials.

$\text{Purpose}$: It is well established that anemia predicts diminished radiocurability in cervix cancer. However, the therapeutic benefit of measures to correct the anemia remains controversial. The objective of this study was to determine the impact of routine transfusion in patients with hemoglobin level (hb-l) ≤11 g/dl.

$\text{Methods and Materials}$: Since 1985, it has been departmental policy to attempt to correct hb-l ≤11 g/dl before and/or during radiotherapy by red blood cell transfusion (RBCT) in patients undergoing radical radiotherapy for primary cervix cancer. To assess the benefit of RBCT, the charts of 204 patients (FIGO: IB–IV) treated until 1997 were reviewed. Parameters analyzed for their impact on disease-specific survival (DSS), pelvic control (PC), and metastases-free survival (MFS) included pretreatment hb-l, treatment hb-l, stage, tumor size, and lymph node status. To determine any differences in outcome according to type of anemia, a separate analysis was performed, grouping patients by cause of anemia (tumor vs. other medical illness related).

$\text{Results}$: Each of the parameters tested was significantly correlated with the end points studied in univariate analysis. Patients whose hb-l were corrected (18.5%) had an outcome that did not differ significantly from that of nontransfused patients, whereas DSS, PC, and MFS (all: p < 0.001) were significantly decreased in nonresponders to RBCT. Subgroup analysis showed no impact of hb-l in patients with other medical illness-related anemia (n = 12). In multivariate analysis treatment, but not pretreatment, hb-l remained predictive for DSS, PC, and MFS. Persistent anemia was associated with a significantly increased risk of death (relative risk: 2.1) and pelvic failure (relative risk: 2.4) compared with nontransfused patients. If only patients with tumor anemia were considered, the respective risks increased (2.7; 3.6). None of the patients with other causes of anemia recurred, whether or not their hb-l was maintained. Assessment of the therapeutic gain in patients who responded to RBCT showed improved PC (p = 0.02) and a trend toward increased DSS (p = 0.06), but no effect on MFS after adjustment for tumor size and lymph node status.

$\text{Conclusion}$: Treatment hb-l, in addition to tumor size and lymph node status, independently predicted outcome. Although our final multivariate analysis showed a therapeutic benefit for patients whose hb-l was corrected, the response to RBCT was disappointing. Results of our subgroup analysis suggest that the cause of anemia in patients with cervical cancer warrants in-depth investigation.

Background and purpose: With the recent development of hemopoietic growth factors and alternatives to transfusion, there has been a renewed interest in the relationships between anemia, tumor hypoxia and treatment outcome in a number of human malignancies. This review is intended to provoke a reconsideration of these issues and their effect on clinical trials, aimed at improving treatment outcome in patients with cervix cancer.

Materials and methods: Using data from the literature and from our own prospective series of tumor oxygenation in cervix cancer, we modeled the impact of anemia on tumor blood flow and hypoxia in animal models and human tumors, examined the relationship between anemia and hypoxia and treatment outcome in patients, and reviewed the impact of transfusion on tumor hypoxia and treatment outcome in cervix cancer.

Results: Anemia may result in a significant reduction in oxygen delivery to tumors, but compensatory mechanisms reduce the impact on tumor oxygenation. Anemia is associated with inferior treatment outcome in cervix cancer, but hemoglobin levels prior to and during treatment are strongly correlated with tumor size, and this may explain the prognostic impact of anemia in older studies. Transfusion and erythropoietin ameliorate hypoxia in only a proportion of anemic patients. Critical analysis of the published data from the Princess Margaret Hospital randomized trial of transfusion in cervix cancer reveals that, when analyzed by intention-to-treat, transfusion did not result in a benefit to patients.

Conclusions: This review suggests that the relationships among anemia, hypoxia, transfusion and treatment outcome are complex. Further study of anemia as an independent prognostic factor is required and randomized studies of transfusion alternatives, such as erythropoietin, must be of sufficient size to detect small treatment effects.

Purpose: The use of methylxanthine derivatives has been postulated as a means of increasing tumor perfusion and thus ameliorating tumor hypoxia. The aim of this study was to quantify and compare the effects of three methylxanthine derivatives: pentoxifylline (PX), torbafylline (TB), and HWA 138 (HW) on tumor perfusion and oxygenation.

Methods and Materials: Anesthetized Sprague Dawley rats with DS-sarcomas implanted subcutaneously onto the hind foot dorsum were used in this study. Mean arterial blood pressure (MABP) was measured throughout experiments. Regional red blood cell (RBC) flux was monitored using a multichannel laser Doppler device and tumor oxygenation on a more global level was assessed polarographically using an O₂-sensitive catheter electrode. The methylxanthine derivatives were administered as a single dose intraperitoneally (for PX 50 mg/kg; for TB and HW 75 mg/kg).

Results: Following drug administration, initial decreases in MABP down to 75% of baseline values were observed for all three substances. PX, HW, and TB caused initial transient reductions in mean RBC flux followed by gradual increases to values of 137 ± 27 %, 139 ± 14 %, and 122 ± 14 % respectively at t = 60 min. Following a small initial decrease upon drug administration, O₂ partial pressure (pO₂) rose to 160 ± 31 %, 153 ± 34 %, and 121 ± 11 % for PX, HW, and TB, respectively at t = 60 min. At the end of the observation period (t = 90 min), increases in RBC flux and pO₂ were still evident. When individual tumors were considered, a variety of patterns (including opposing effects) for changes in RBC flux were seen, not necessarily reflected in the mean values. Thus, while the methylxanthine derivatives caused an increased average tumor perfusion, there is evidence suggesting that a redistribution of tumor blood flow occurs which may amplify preexisting heterogeneity.

Conclusions: Substantial improvements in tumor oxygenation and perfusion were observed after administration of the methylxanthine derivatives. These substances may therefore be of use during tumor therapies in which the outcome may be detrimentally affected by the presence of hypoxia.

Aim: In order to investigate possible pathophysiological mechanisms underlying the postulated preferential protective effect of hypoxia on normal tissue during radiotherapy, the impact of acute respiratory hypoxia (8.2% O₂ + 91.8% N₂) on tissue oxygenation was assessed.

Methods: Tumor and normal tissue oxygenation was directly determined using O₂-sensitive electrodes in two experimental rat tumors (DS and Yoshida sarcomas) and in the normal subcutis of the hind foot dorsum.

Results: During respiratory hypoxia, arterial blood O₂ tension (pO₂), oxyhemoglobin saturation and mean arterial blood pressure decreased. Changes in the arterial blood gas status were accompanied by a reflex hyperventilation leading to hypocapnia and respiratory alkalosis. In the subcutis, tissue oxygenation worsened during acute hypoxia, with decreases in the mean and median pO₂. Significant increases in the hypoxic fractions were, however, not seen. In tumor tissues, oxygenation also worsened upon hypoxic hypoxia with significant decreases in the mean and median pO₂ and increases in the size of the hypoxic fractions for both sarcomas.

Conclusion: These results suggest that during respiratory hypoxia, radiobiologically relevant reductions in the oxygenation (and a subsequent selective radioprotection) of normal tissue may not be achieved. In addition, in the tumor models studied, a worsening of tumor oxygenation was seen which could result in an increased radioresistance.