Hypoxia and radiation response in human tumors

H y p o x i a and Radiation R e s p o n s e in H u m a n Tumors Michael H6ckel, Karlheinz Schlenger,MargareteMitze, Uwe Schaffer, andPeterVaupel This study demonstrates by an updated analysis of an ongoing prospective study that tumor oxygenation, as measured with a validated standardized polarographic needle electrode method before treatment, powerfully predicts the prognosis of patients receiving radiotherapy for intermediate and advanced stage cancer of the uterine cervix. First evidence for a host component in tumor oxygenation based on a significant correlation between median pO2 values determined in normal subcutaneous fatty tissue and in cervical cancer is also presented. Further investigations are necessary to clarify

whether tumor hypoxia is just a marker of intrinsic tumor aggressiveness or whether the negative impact of tumor hypoxia on survival is related to radiobiological mechanisms caused by hypoxia per se, which may include (1) the reduced oxygen enhancement effect, (2) increased radioresistance due to expression of genes for cell cycle delay and stress proteins, and/or [3) accelerated tumor progression to more radioresistant and metastatic variants by increased genetic heterogeneity. Copyright 9 1996 by W.B. Saunders Company

olid tumors are multicellular "ecosystems" with chaotic structural and functional characteristics. As a consequence of the unrestricted tumor cell proliferation with increased demand for oxygen and the chaotic microvasculature, which is not able to supply oxygen sufficiently, the existence of hypoxic areas within the tumor can be predicted. In fact, low and even zero oxygen partial pressures (pO2) have been measured in experimental, as well as human solid tumors, j-7 We have shown in intermediate and advanced stage cancer of the uterine cervix that within an individual tumor, oxygen partial pressures exhibit an anisotropic spatial distribution. Yet intraindividual heterogeneity is less pronounced than interindividual heterogeneity allowing a categorization of tumors with respect to their oxygenation status. We have also presented preliminary data from the first analysis of a prospective study that indicate hypoxia, per se, is a powerful prognostic factor negatively influencing overall and recurrencefree smwival. ~,9 From these results, t h e following questions arise: (i) why are some tumors less well oxygenated than others of the same size or stage? Is hypoxia dictated by the tumor cells or do (extrinsic) host factors also play a role? (2) Why do patients with

low pO 2 cervical cancers have a significantly poorer prognosis than those with better oxygenated tumors of the same size or stage? Are hypoxic tumors more aggressive (ie, more progressed in clonal evolution) or is the treatment less effective at lower oxygen tensions? In this report, we investigate the relevance of tumor hypoxia in patients with intermediate and advanced stage cancer of the uterine cervix receiving radiation therapy based on the updated results of our trial.

S

From the Departments of Obstetrics and Gynecology, and Pathology and Institute cffPhysiolo~ and Pathophysiology, University of Mainz, Mainz, Germany. Supported by Grant No. M40/91/Valfrom the Deutsche KrebshilJe. M.H. was supported by.fundsfiom LTS Ix~hmann Therapie Sysleme and Else-Kr6ner-Fresenius Sti~ung. Address reprint requests to Michael Ho?kel, MD, PhD, Department of Obstetrics and Gynecology, Universi~ ~ Mainz Medical School D-55101 Mainz, Germany. Copyright 9 1996by W.B. Saunders Company 1053-4296/96/0601-0002505.00/0

Methods Patients and Treatments

All patients with celwical cancers of at least 3 cm diameter as determined by clinical and computed tomography (CT) or magnetic resonance imaging (MRI) who had been referred to the Department of Ob/Gyn, University of Mainz Medical Center, were eligible for the prospective study. The study design was approved by an ethics committee. Patients were accrued once informed consent had been obtained. Clinical staging was performed according to the International Federation of Gynecology and Obstetrics (FIGO) classification. Patients.treated with curative intent received radiation with or without adjunctive chemotherapy. Definitive radiation was administered as combined teletherapy and brachytherapy. External beam irradiation was delivered using 10-MV photons produced by a linear accelerator in the Department of Radiation Therapy. For brachytherapy, a high-dose rate Ir192 afterloading machine in the Ob/Gyn Department was used. The patients usually received 45 Gy whole pelvis external beam radiation by the standard four-field box technique in 2 Gy/day fractions, five times per week.

Seminars in Radiation Oncology, Vol 6; No 1 (Janua~), 1996.'pp3-9

3

H6ckel et al

Radiation treatment was completed by three to four endocavitary insertions with a Hentschke applicator delivering a total dose of 24 to 28 Gy to point A at a dose rate of 0.5 to 1 Gy/min in weekly intervals. Lateral tumor extension was treated with additional teletherapy boosts to the parametria of up to 15 Gy. Postoperative adjuvant whole pelvic external beam radiation was given to a treatment field to include the upper half of the vaginacaudally. The cephalad margin was L4/5 and the lateral borders were 1.5 cm beyond the linea terminalis. Systemic chemotherapy regimens were used as follows: cis-platinum; vincristine and bleomycin as induction therapy followed by definitive radiation; cis-platinum or carboplatinum for concomitant chemoradiotherapy. The majority of patients in this study were seen at regular follow-up intervals in this department. In the case of tumor recurrences, the site(s) of relapse were identified by clinical and radiological work-up. Complete follow-up reports were obtained from those patientswhose after-care examinations were not performed at the University of Mainz. No patient was lost to follow-up.

Intratumoral pO2 Histography Tumor oxygenation was measured pretherapeutically with the Eppendorf histograph system adhering to the standard procedure as developed and validated earlier. 79 Immediately after calibration of the ethylene oxide-sterilized pO2 electrode, the conscious patient was placed into a defined lithotomy position, pO2 readings were performed along linear tracks, first in the subcutaneous fat of the mons pubis followed by measurements at the 12 o'clock and 6 o'clock sites of the vital tumor tissue. Within the cervical tissue 25 to 35 pO2 measurements were taken on each tumor track, (50 to 70 readings in total) starting at a tissue depth of 5 mm. The measuring points were placed 0.7 m m apart from each other resulting in an overall measuring track length of approximately 2 cm. Using an online computing system, the pO2 data of each track were displayed (1) as absolute pO2 values related to the location of the measuring point along the track and (2) as relative frequencies within a pO2 histogram ranging from 0 to 100 m m Hg with a class width of 2.5 mmHg. After the pO2 measurements, core biopsies of 2 m m diameter and 2 cm length were taken from those tumor areas, where the pO~ determination had been made using a Biopty device (Radiplast AB,

Uppsala, Sweden). The biopsies were fixed in buffered formalin and processed for histology to assure that the pO2 measurements had been performed in vital tumor tissue and not in gross tumor necrosis or cervical stroma without tumor cells. Concomitantly with the pO2 determinations intravaginal temperature, hemoglobin concentration and hematocrit were monitored. The pO2 measurements were usually performed 1 to 5 days before the oncologic treatment.

Data Analysis The oxygenation status of each tumor was represented by the median pO2, and the low pO2 fraction (corresponding to the relative frequency of pO2 readings from 0 to 2.5 m m Hg or 0 to 5 mmHg) derived from the pooled histograms. Endpoints of the study were recurrence-free survival and overall survival. For comparison, the following statistical methods were applied: determination of correlation coefficients, Fisher's exact test, and Mann-Whitney-Wilcoxon test (U-test). Survival and recurrence-free survival were calculated with the Kaplan-Meier method. Differences between survival curves were analyzed with the log-rank test. Variables influencing survival and recurrence-free survival were evaluated with the univariate and multivariate Cox proportional hazards model.

Results From June 1989 until March 1995, 44 patients with advanced cancers of the uterine cervix, FIGO stages Ib bulky (n = 1), IIa,b (n = 18), IlIa,b (n = 25) treated with primary radiation with curative intent have entered the study. Fifty-two percent of the tumors had median pO2 < 10 mm Hg; in 68% pO2 readings < 5 m m Hg, and in 43% pO2 readings < 2.5 m m Hg were recorded. No correlation could be detected between age, menopausal status, parity of the patients, and the oxygenation pattern of their tumors. Likewise, no influence of clinical tumor stage and tumor size, histologic type, and differentiation on tumor oxygenation has been found. However, a significant correlation between tumor oxygenation and host oxygenation measured in the mons puns fatty tissue became evident in the scatter plot (Fig 1). Twenty-nine patients received radiation therapy alone, 11 patients were treated with radiation subsequent to induction chemotherapy, and four patients underwent concomitant chemoradiotherapy. All patients were analyzed for outcome. Within a median

Hypoxia and Radiation Response

50-

chemotherapy before or concomitant with radiation (Table 1). Cox regression analysis showed tumor oxygenation as the strongest independent prognostic factor in this patient cohort.

N = 44 r = 0.475 p = 0.011

40-

e

32

E E 30-

o a

o TM

Q

Q_ L_ 2 0 O

i

e o

9

10-

~

#

9 9

0-

|

e

eoe 9

9

9 i

9

**

9

5'o

20

30

9

Q

e

o 9

Discussion

o

~|

6'o

8'0

7'o

NormQItissuepO2 [mmHg] F i g u r e 1. Scatter plot demonstrating a significant (r = 0.475; P = .01) correlation between median pO2 readings of normal subcutaneous fatty tissue at the mons pubis and median intratumoral pO2 readings in patients with intermediate and advanced stage cancer of the uterine cervix (n = 44). observation time of 42 months (range, 3 to 72 months) 16 patients progressed or relapsed. All but two patients failed locoregionally with or without simultaneous distant metastases. Patients with low pO 2 tumors (median pO2 < 10 m m Hg) had a significantly worse disease-free survival probability compared with better oxygenated tumors (pO2-> 10 m m Hg; Fig 2). Both subgroups, which resulted from the stratification according to their oxygenation status, did not differ significantly in the variables: age of the patient, menopausal status, clinical stage and size of the tumor, histologic type, and addition of

Diseose-free survival probobility 1.o- L 0.8-

Median pO2 ~ 10 mmHg (n = 21)9

0.6I Median pO2 < 10 rnmHg (n = 23) o

0.4-

0.2Log-rank: 0.0- (~

llo

p = 0.0484 210

310

4~0

5nO

60

7nO

8~0

Time [months] Figure

2. Disease-free survival probability calculated

with the Kaplan-Meier method of 44 cervical cancer patients treated with primary radiation stratified for tumor oxygenation.

Although various methods have been established to detect hypoxia-related features in human solid tumors, only polarographic electrode probes can measure microregional tissue pO2 directly. This assay of tumor oxygenation is, therefore, presently regarded as the "gold standard technology'U ~ Introduction of the commercially available Eppendorf histograph system has significantly facilitated intratumoral pO 2 measurements in humans and enabled clinical investigations with larger patient cohorts, which could not be performed before due to major technical obstacles with the needle electrode. Within the last years, the Eppendorf polarographic probe has been technically further improved. It has also been regarded as most accurate and reliable in determining tumor oxygenation by a recent International Expert Consensus Workshop held at the 43rd Annual Meeting of the Radiation Research Society, San Jose, California (April 5, 1995). T a b l e 1. Subgroup Characteristics

Age _<50yr >50yr Menopausal status Pre Post FIGO stage Ib IIa,b IIIa,b Tumor size (cm) Mean (range) Histology Squamous Adeno Treatment RT only CT before RT Concomitant RT + CT

Low p02 Tumors (~ = 23)

High p02 Tumors (~ = 21)

6 17

7 14

7 16

8 13

0 9 14

I 9 11

6.0 (3-8) .

5.9 (3-8)

23 0

19 2

16 5

13 6

2

2

NOTE. Patient demographics, tumor characteristics, and treatment specifications of 23 patients with low pO~ cervicalcancers (median pO2 < 10 mm Hg) as compared with 21 patients with high pO2 cervicalcancers (median pO2 >- 10 mm Hg). Abbreviations:RT, radiotherapy;CT, chemotherapy.

6

H&kel et al

We have been using the Eppendorf probe for investigating the clinical relevance of tumor oxygenation in intermediate and advanced stage cancer of the uterine cervix since 1989 in an open prospective trial. Intermediate and advanced stage cancer of the uterine cervix appears to be a promising tumor entity to study the importance of tumor oxygenation because (1) defined invasive access to the tumor under clinical control without the necessity of imaging procedures or anesthesia is possible, (2) the tumors are treated either by primary radiotherapy or radical surgery for local control with the endpoints disease-fi'ee smwival and overall survival, and (3) median disease-free intervals are relatively short (less than 1 year) and the rate of tumor recurrence or progression is high (more than 50%). Clinical tumor oxygenation studies using the Eppendorf probe are currently also being performed in other tumor entities such as primary and metastatic head and neck cancers, IH3 breast cancer, 9'14'L5lung cancer, ~4 rectal cancer, 16 soft tissue sarcoma] 7 and glioblastoma, ta The potential risks of the invasive measurements with the polarographic probe are local infection, hemorrhage, and dislocation of tumor cell clusters along the measuring tracks.~9Although minor discomfort was reported by some patients without anesthesia, neither infection nor serious hemorrhage have been observed by the participants of a recent International User Group Meeting in more than 1,000 intratumoral measurements in patients. 2~ The risk of tumor spread by displacement of tumor cell clusters must be regarded as extremely low based on experience gained with fine needle biopsies, 91,22yet we recommend that the region of the measuring track be included in the locoregional treatment field of radiation or removed by surgery. Informed consent must be obtained from all patients undergoing the investigation. We have developed and validated a standard procedure for measuring tumor oxygenation in intermediate and advanced stage cervical cancer with the Eppendorfprobe. Although intratumoral heterogeneity can be pronounced, the methodological evaluation of the standard measuring procedure showed that the intertumoral variation is significantly higher than intraobserver, interobserver, and intratumoral variations (H6ckel et al, unpublished results). Similar data have been reported recently by others for cervical cancer, as well as for soft tissue sarcomas and lymph node metastases. 23'24 Using the median pO 2 value of the pooled histogram of the two intratumoral pO 2 measurements, a

numeric parameter of the oxygenation of an individual cexwix tumor can be obtained. With respect to the cut-off level of l 0 mm Hg, which we have defined to discriminate between low pO2 tumors and high pO2 tumors, the possibility of changing the tumor category by observer and intratumoral variation is 10% and 11%, respectively (H6ckel et al, unpublished results). In cervical carcinomas of more than 3 cm in largest diameter described as intermediate and advanced stage tum(~rs, oxygenation did not show any relation to patient demographics, clinical tumor size, and stage, as well as histologic type and degree of differentiation. At present, it is not clear whether the independence from tumor size also holds true for small-volume cervical cancer. Interestingly, we found a significant correlation between the oxygenation status of normal subcutaneous tissue measured in the mons pubis and the tumor oxygenation. If supported in larger patient series, this result may point to an influence of the host on tumor oxygenation, which might be of clinical importance. So far, we have been unable to detect any influence of putative factors on overall tissue oxygenation, such as smoking habits and hemoglobin content or hematocrit values. However, based on our data we cannot definitely rule out such effects, as the information on smoking habits used in the analysis was incomplete and the majority of patients in this study had normal hemoglobin concentrations within a relatively narrow range. The most important result of our ongoing investigation is the demonstration of the prognostic relevance of tumor oxygenation in intermediate and advanced stage cancer of the uterine cervix. The updated analysis of our prospective trial again confirmed our previous observation~,9 that tumor oxygenation as measured pretherapeutically with the standard procedure is the most powerful predictor of overall and disease-fi'ee survival, followed by FIGO stage. The difference in the Kaplan-Meier curves stratified tbr tumor oxygenation (median pO2 < 10 mm Hg versus > 10 mm Hg) cannot be related to the local tumor load because FIGO stages and tumor sizes were not different in the two subgroups. Likewise, the addition of chemotherapy either preceding or concomitant to radiation and the median follow-up times were similar. Moreover, multivariate Cox regression analysis identified tumor oxygenation as the most important independent risk factor for poor outcome both as a continuous and as a categorized variable. Several earlier studies have demonstrated a

Hypoxia and Radiation Response

relation between tumor hypoxia and poor treatment results of radiation. Kolstad 25 measured oxygen tensions at the (vaginal) surface of cervical cancers with a bare polarographic electrode. Tumors with the lowest pO2 values between 0 and 9 m m H g had more local recurrences than those with lowest pO2 values > 10 m m Hg. Gatenby et aF 6 determined the tumor oxygenation status in lymph node metastases of squamous cell carcinomas of the head and neck with a CT-guided polarographic needle electrode. He noted a significantly higher mean pO9 in the subgroup of patients responding to radiotherapy as compared with the nonresponders. Recently, the predictive value of oxygen electrode measurements for radiation response could be simulated with an experimental mouse tumor m o d e l s Based on our measurements in cervical cancer, the mean and median pO2 levels strongly correlated with the lowest two pO 2 classes in the oxygenation histogram obtained with the Eppendorf probe, ie, the 0 to 2.5 m m Hg fraction or the 0 to 5 m m Hg fraction. If these low pO2 fractions are used as variables instead of the median pO 2 for the statistical analysis of disease-free or overall survival probability in our group of patients, the results are the same. We prefer the median pO9 to characterize the oxygenation status of an individual tumor rather than the frequency of the lowest pO9 classes because the methodologically inherent variation increases when measuring very low oxygen partial pressures with the polarographic probe. ~~Although our predictive tumor oxygenation data in patients treated with radiation for intermediate and advanced stage cancer of the uterine cervix are in accordance with the assumption that photon irradiation is less effective in hypoxic tumors because of a reduced oxygen enhancement effect due to the lack of molecular oxygen, this may not be the only or even the most probable explanation. Tumor cells, like proliferating normal cells, when stressed by hypoxia can respond by the expression of growth arrest genes (such as p53) to induce a GI/S check point. 28 Similarly, genes coding for stress proteins can also be activated by hypoxia. 29 Both cellular phenomena at the transcriptional level may contribute to the increase of radioresistance under hypoxia. 3~ Microregional tumor hypoxia appears to be a major stimulus for tumor angiogenesis. Vascular endothelial growth factor (VEGF), a strong mediator substance for neovascularization, has been immunohistochemically highlighted at areas of hypoxia in histologic tumor sections. ~-35 As a consequence of

the formation of new microvessels, reoxygenation of former hypoxic microregions may occur. Cells that were in the S phase during the time of hypoxia react on reoxygenation by DNA overreplication resulting in the amplification of further resistancy genes. 36 DNA overreplication-recombination events may also lead to chromosomal aberrations and rearrangements increasing cellular heterogeneity and malignant progression. 37 Hypoxia-reoxygenation induced enhancement of the metastatic potential of tumor cells has been demonstrated in vitro. ~a In addition, we propose another hypothesis: hypoxia-reoxygenation events in a solid tumor might be regarded as microregional counterparts of the ischemia-reperfusion syndrome with the production of oxygen-derived free radicals (eg, by the xanthine oxidase pathway) as the key event. 39,4~ The generation of hydroxyl radicals may increase genetic instability through mutations and facilitate clonal evolution more aggressive cellular phenotypes. 4I We have recently demonstrated that tumor microvascularity as determined with the closest-individual method in defined core biopsies of cervical cancer is independent of tumor oxygenation and also has a significant impact on survival.42 Contrary to earlier views, 95,43,44 our data identified high vascularity as a marker of tumor aggressiveness and poor outcome. Tumors that are hypoxic despite high vascularity have an extremely poor outcome independent of their treatment modalities. In these tumors hypoxiareoxygenation events promoting tumor progression may be particularly frequent (H6ckel et al, unpublished results). The hypothesis that tumor hypoxia may accelerate tumor progression into metastatic, radio- and chemoresistant variants is strongly supported by our observation that hypoxic cervical cancers treated by surgery (1) also relapsed significantly more frequently than well oxygenated tumors of the same size and stage, and (2) the majority of these relapses occurred in the pelvis and responded poorly to salvage radiation or chemotherapy (H6ckel et al, unpublished results). What might be the clinical consequences of the finding that intratumoral low pO2 indicates a much poorer prognosis in patients with intermediate and advanced stage cancer of the uterine cervix treated with standard therapy? Patients with stage IB low pO2 tumors identified by polarographic electrode measurements should be offered participation in controlled randomized prospective trials aimed at investigating new methods to circumvent the hypoxia effect, such as thermoradiotherapy or radio-

H6ckel et al

therapy combined with drugs cytotoxic to hypoxic cells. Likewise, a variation of the combined operative and radiotherapeutic treatment (CORT), which we have developed for patients with postirradiation pelvic side wall disease 45,46 m i g h t be c o n s i d e r e d . T h i s t h e r a p y would i n c l u d e initial pelvic i r r a d i a t i o n followed by pelvic e v i s c e r a t i o n a n d r e s e c t i o n o f t h e pelvic wall muscles, t r a n s p o s i t i o n o f a u t o l o g o u s u n i r r a d i a t e d tissue i n t o t h e pelvis by m u s c u l o c u t a n e o u s a n d o m e n t u m m a j u s flaps, a n d delivery of a s e c o n d r a d i a t i o n to p a r t s o f t h e pelvis by b r a c h y - or teletherapy. H y p o x i c s t a g e I b a n d II t u m o r s m i g h t b e t r e a t e d w i t h h i g h e r r a d i a t i o n doses, w h i c h are o t h e r w i s e r e s e r v e d for s t a g e III cervical c a n c e r s or c o n c o m i t a n t c h e m o r a d i o t h e r a p y . A l t e r n a t i v e l y , e x t e n d e d radical s u r g e r y s u c h as P i v e r type I V h y s t e r e c t o m y supplem e n t e d w i t h c o m p l e t e pelvic a n d p e r i a o r t i c l y m p h n o d e dissection m i g h t b e p e r f o r m e d . T h e s e s t u d i e s could a n s w e r t h e q u e s t i o n as to w h e t h e r local c o n t r o l in hy-poxic cervical c a n c e r c a n b e i m p r o v e d a n d w h e t h e r this w o u l d also r e s u l t in h i g h e r l o n g - t e r m survival as s u g g e s t e d g e n e r a l l y by Suit a n d W e s t gate. 47 A p e s s i m i s t i c view, however, m i g h t e x p e c t n o significant i m p r o v e m e n t i n l o n g - t e r m survival, feari n g t h a t d i s t a n t m e t a s t a s e s as t h e final s t a g e in t u m o r p r o g r e s s i o n w o u l d t h e n b e c o m e m o r e preval e n t in p a t i e n t s w h o a r e t u m o r - f r e e in t h e i r pelves.

References 1. Vaupel P, Kallinowski F, OkunieffP: Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: A review. Cancer Res 49:6449-6465, 1989 2. Vaupel P, Schlenger K, Knoop C, H6ckel M: Oxygenation of human tumors: Evaluation of tissue oxygen distribution in breast cancers by computerized 02 tension measurements. Cancer Res 51:3316-3322, 1991 3. Vaupel P, Frinak S, Bicher HI: Heterogeneous oxygen partial pressure and pH distribution in C3H mouse mammary adenocarcinoma. Cancer Res 41:2008-2013, 1981 4. Vaupel P: Hypoxia in neoplastic tissue. Microvasc Res 13:399408, 1977 5. Vaupel P, Fortmeyer HP, Runkel S, et al: Blood flow, oxygen consumption and tissue oxygenation of human breast cancer xenografts in nude rats. Cancer Res 47:3496-3503, 1987 6. Kallinowski F, Schlenger K, Runkel S, et al: Blood flow, metabolism, cellular microenvironment, and growth rate of human tumor xenografts. Cancer Res 49:3759-3765, 1989 7. H6ckel M, Schlenger K, I~mop C, et al: Oxygenation of carcinomas of the uterine cervix: Evaluation by computerized 02 tension measurements. Cancer Res 51:6098-6102, 1991 8. H6ckel M, Knoop C, Schlenger K, et al: Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Onco126:45-50, 1993 9. H6ckel M, Vorndran B, Schlenger K, et al: Tumor oxygen-

ation: A new predictive parameter in locally advanced cancer of the uterine cervix. Gynecol Oncol 51:141-149, 1993 10. Stone HB, Brown JM, Phillips TL, et al: Oxygen in human tumors: Correlations between methods of measurement and response to therapy. Radiat Res 136:422-434, 1993 11. Fleckenstein W,JungblutJR, SuckffillM, et al: Sauerstoffdruckverteilungen in Zentrum und Peripherie maligner Kopf-HalsTumoren. Dtsch. Z. Mund Kiefer Gesichts Chir 12:205-211, 1993 12. Martin L, Lartigau E, Weeger P, et al: Changes in the oxygenation of head and neck tumours during carbogen breathing. Radiother Onco127:123-130, 1993 13. Lartigau E, Le Ridant AM, Lambin P, et al: Oxygenation of head and neck tumors. Cancer 71:2319-2325, 1993 14. Falk SJ, Ward R, Bleehen NM: The influence of carbogen breathing on tumour tissue oxygenation in man evaluated by computerized pO2 histography. BrJ Cancer 66:919-924, 1992 15. Runkel S, Wischnik A, Teubner J, et al: Oxygenation of mammary tumors as evaluated by ultrasound-guided computerized-pO2-histography. Adv Exp Med Bio1345:451-458, 1994 16. Kallinowski F, Buhr HJ: Tissue oxygenation of primary, metastatic and xenografted rectal cancers, in Vaupel PW, Kelleher DK, G/inderoth M (eds): Funktionsanalyse biologischer Systeme. Tumor Oxygenation: Stuttgart, Jena, New York, Gustav Fischer Verlag, 1995, pp 205-209 17. Brizel DM, Rosner G~ Harrelson J, Prosnitz LR, Dewhirst MW: Pretreatment oxygenation profiles of human soft tissue sarcomas. IntJ Radiat Oncol Biol Phys 30:635-642, 1994 18. Rampling RG, Cruickshank G, Lewis D, et al: Direct measurements ofpO2 distribution and bioreductive enzymes in human malignant brain tumours. Int J Radiat Oncol Biol Phys 29:427-432, 1994 19. Lartigau E, Lespinasse F, Vitu L, et al: Does the direct measurement of oxygen tension in tumors have any adverse effects? IntJ Radiat Oncol Biol Phys 22:949-951, 1992 20. H6ckel M: Clinical relevance of tumor oxygenation, in Vaupel PW, Kelleher D, Gfinderoth M (eds): Funktionsanalyse Biologischer Systeme. Tumor Oxygenation: Stuttgart, Jena, New York, Gustav Fischer Verlag, 1995, pp 337-338 21. Glasgow BJ, Brown HH, Zargoza AM, et al: Quantitation of tumor seeding from fine needle aspiration of ocular melanomas. AmJ Ophthalmo1105:538-546, 1988 22. Ryd W, Hagmar B, Eriksson O: Local tumour cell seeding by fine-needle aspiration biopsy. A semiquantitative study. Acta Pathol Microbiol Immunol Scand 91:17-21, 1983 23. Nordsmark M, Bentzen SM, Overgaard J: Measurement of human tumour oxygenation status by a polarographic needle electrode. Acta Onco133:383-389, 1994 24. Brizel DM, Rosner GL, Prosnitz LR, et al: Patterns and variability of tumor oxygenation in human soft tissue sarcomas, cervical carcinomas, and lymph node metastases. Int J Radiat Oncol Biol Phys 32:1121- 1125, 1995 25. Kolstad P: Intercapillary distance, oxygen tension, and local recurrence in celvix cancer. Scand J Clin Lab Invest Suppl 106:145-157, 1968 26. Gatenby RA, Kessler HB, Rosenblum JS, et al: Oxygen distribution in squamous cell carcinoma metastases and its relationship to outcome of radiation therapy. Int J Radiat Oncol Biol Phys 14:831-838, 1988 27. Nordsmark M, Horsman MR, Overgaard J: The predictive value of oxygen electrode measurements to radiotherapy: Simulation in an experimental tumor model, 43rd Annual

Hypoxia and Radiation Response

Meeting of the Radiation Research Society San Jose, CA, April I-6, 1995, p 21-311 28. Giaccia AJ, Graeber TG: Regulation of cell proliferation by hypoxia, 43rd Annual Meeting of the Radiation Research Society San Jose, CA, April 1-6, 1995, p 12-397 29. Hall E: Radiosensitivity and cell age in the mitotic cycle, Radiobiology for the Radiologist: Philadelphia, PA, Lippincott, 1994, pp 91 - 105 30. O'Dwyer PJ, Filali M, Hamilton TC, Yao K-S: Mechanisms of altered gene expression under hypoxic conditions, 43rd Annual Meeting of the Radiation Research Society San Jose, CA, April 1-6, 1995, $24-2 31. Zhou S, Hill HZ, Hill GJ: Further characterization and purification of an multitherapy resistance factor (MTRF), 43rd Annual Meeting of the Radiation Research Society San Jose, CA, April 1-6, 1995, P04-54 32. Brown LF, Berse B,Jackman RW, et al: Increased expression of vascular permeability factor (VEGF) and its receptor in kidney and bladder carcinomas. A m J Pathol 143:1255-1262, 1993 33. Plate KH, Breier G, Weich HA, et al: Vascular endothelial growth factor is a potential tumor angiogenesis factor in human gliomas in vivo. Nature 359:845-848, 1992 34. Shweiki D, Itin A, Softer D, et al: Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature 359:843-845, 1992 35. Hlatky L, Tsionou C, Hahnfeldt P, et al: Mammary fibroblasts may influence breast tumor angiogenesis via hypoxia-induced vascular endothelial growth factor up-regulation and protein expression. Cancer Res 54:6083-6086, 1994 36. Rice GC, Hoy C, Schimke RT: Transient hy-poxia enhances the frequency ofdihydrofolate reductase gene amplification in chinese hamster ovary cells. Proc Natl Acad Sci USA 83:59785982, 1986

9

37. Schimke RT: Gene amplification, drug resistance, and cancer. Cancer Res 44:1735-1742, 1984 38. Young SD, Marshall RS, Hill RP: Hypoxia induces DNA overrcplication and enhances metastatic potential of murine tumor cells. Proc Natl Acad Sci USA 85:9533-9537, 1988 39. McCord JM: Oxygen-derived free radicals in post-ischemic tissue injury. N EnglJ Med 312:159-163, 1985 40. Russel RC, Roth AC, KucanJO, et al: Reperfusion injury and oxygen free radicals, a review.J Reconstr Microsurg 5:79-87, 1989 41. Malins DC, Holmes EH, Polissar NL, et al: The etiology of breast cancer. Cancer 71:3036-3043, 1993 42. Schlenger K, H6ckel M, Mitze M, et al: Tumor vascularity--A novel prognostic factor in advanced cervical carcinoma. Gynecol Oncol (in press) 43. Awwad H, El Naggar M, Mocktar N, et al: Intercapillary distance measurements as an indicator of hypoxia in carcinoma of the cervix uteri. Int Radiat Oncol Biol Phys 12:13291333, 1986 44. Sirack~ E, SirackyJ, Pappalov~ N, ct al: Vasctilarization and radiocurability in cancer of the uterine cervix. A retrospective study. Neoplasma 29:183-189, 1982 45. H6ckel M, Knapstein PG, Kutzner J: A novel combined operative and radiotherapeutic treatment approach for recurrent gynecologic malignant lesions infiltrating the pelvic wall. Snrg Gynecol Obstet 173:297-302, 1991 46. H6ckel M, Knapstein PG: The combined operative and radiotherapeutic treatment (CORT) of recurrent tumors infiltrating the pelvic wall: First experience with 18 patients. Gynecol Onco146:20-28, 1992 47. Suit HD, Westgate SJ: Impact of improved local control on survival. IntJ Radiat Oncol Biol Phys 12:453-458, 1986