- Email: [email protected]
HIV impact on acute morbidity and pelvic tumor control following radiotherapy for cervical cancer
Gynecologic Oncology 100 (2006) 405 – 411 www.elsevier.com/locate/ygyno
HIV impact on acute morbidity and pelvic tumor control following radiotherapy for cervical cancerB Peter Gichangi a,*, Job Bwayo b, Benson Estambale b, Khama Rogo c, Eliud Njuguna d, Shadrack Ojwang e, Marleen Temmerman f a
Department of Human Anatomy and Obstetrics and Gynecology, University of Nairobi, P.O. Box 2631KNH 00202, Nairobi, Kenya b Department of Medical Microbiology, University of Nairobi, Nairobi, Kenya c Nairobi Oncology Center, Nairobi, Kenya d Radiotherapy Unit, Kenyatta National Hospital, Nairobi, Kenya e Department of Obstetrics and Gynecology, University of Nairobi, Nairobi, Kenya f International Center for Reproductive Health, Ghent University, Ghent, Belgium Received 13 April 2005 Available online 4 November 2005
Abstract Objective. To determine the impact of HIV infection on acute morbidity and pelvic tumor control following external beam radiotherapy (EBRT) for cervical cancer. Method. 218 patients receiving EBRT who also had HIV testing after informed consent was obtained were evaluated. Acute treatment toxicity was documented weekly during treatment and 1 month post-EBRT. Pelvic tumor control was documented at 4 and 7 months post-EBRT. Clinicians were blinded for HIV results. Results. About 20% of the patients were HIV-positive. Overall, 53.4% of the patients had radiation-related acute toxicity (grade 3 – 4). HIV infection was associated with a 7-fold higher risk of multisystem toxicity: skin, gastrointestinal tract (GIT) and genitourinary tract (GUT) systems. It was also an independent risk factor for treatment interruptions (adjusted relative risk 2.2). About 19% of the patients had residual tumor at 4 and 7 months post-EBRT. HIV infection was independently and significantly associated with 6-fold higher risk of residual tumor post-EBRT. The hazard ratio of having residual tumor after initial EBRT was 3.1-times larger for HIV-positive than for HIV-negative patients (P = 0.014). Conclusion. HIV is associated with increased risk of multisystem radiation-related toxicity; treatment interruptions and pelvic failure (residual tumor) following EBRT. HIV infection is an adverse prognostic factor for outcome of cervical cancer treatment. D 2005 Elsevier Inc. All rights reserved. Keywords: HIV; Cervical cancer; Radiotherapy; Outcome
Introduction Treatment of cervical cancer even in optimal conditions is associated with loco-regional failure. Pelvic failure rates of between 10 and 74% have been reported depending on the clinical stage [1,2]. Reasons for failure include tumor-related factors such as large volume of disease, FIGO clinical stage and treatment-related factors such as the inability to deliver adequate radiation doses, treatment prolongation and age of the i
This study was supported by Flemish Interuniversity Council (VLIR). * Corresponding author. E-mail addresses: [email protected], [email protected] (P. Gichangi). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.10.006
patient [1,3 – 5]. The overall treatment time has an impact on acute/late morbidity and survival. Petereit et al. [5] showed a decrease of 0.7%/day of local control and 0.6%/day of survival for each additional day of treatment beyond 55 days in cervical cancer patients. Erridge et al. [6] showed no significant adverse effect in survival if cervical cancer treatment was less than 7 weeks but patients treated over a short period had increased incidence of late morbidity. One of the common causes of treatment prolongation is comorbidity and acute toxicity. HIV infection is now a common comorbid infection in patients with invasive cervical cancer in the African setting. HIV infection in advanced stage is associated with certain opportunistic infections/conditions such as severe diarrhea, anemia, nonspecific skin dermatosis and genital tract infec-
406
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
tions. Chronic diarrhea of HIV infection may worsen in the setting of radiation proctitis thereby contributing to treatment interruption. Acute GIT toxicity, especially diarrhea, is significantly associated with an increased risk of late rectal injury [7,8], a phenomenon termed consequential late effect [8– 10]. HIV infection is associated with more aggressive invasive cervical cancer, and higher treatment failures and recurrences following standard treatment [11 – 16]. Studies on anal cancer, an HPV-related malignancy with increased frequency in HIV infected patients show that HIV infected patients have poorer tolerance to treatment and shorter time (1.4 years) to cancer-related death [17,18]. It is ill understood how HIV infection confers such an adverse effect on treatment outcome of cervical cancer. It is however apparent that chronic and opportunistic infections associated with HIV infection may adversely affect radiation treatment by causing treatment interruptions or acute toxicity and thereby adverse outcome. HIV infected cervical cancer patients untreated for HIV have a short survival of about 10 months [19], which is too short to evaluate the late morbidity associated with radiotherapy treatment. This study was therefore undertaken to evaluate the impact of HIV infection on acute morbidity related to EBRT and pelvic tumor control after EBRT in an African setting where both conditions are prevalent. Materials and methods Patient population Between January 2000 and April 2002, 208 consecutive patients who were treated with EBRT at Kenyatta National Hospital (KNH) radiotherapy unit and who consented to have HIV testing, were included in the study. Recruited patients received pre- and posttest HIV counseling. Pretreatment evaluation for all patients consisted of history, complete physical examination, examination
under anesthetic for clinical staging according to the International Federation of Gynecology and Obstetrics 1995 (FIGO) system, hemoglobin determination, chest X-ray and Karnofsky score determination. The tumors were classified as Stage 1 (16), II (98), III (78) and IV (16). All tumors were histologically verified. Squamous tumors constituted 87.0%, adenocarcinoma 5.3%, adenosquamous carcinoma 2.4% and anaplastic carcinoma 4.8%. One patient (0.5%) had cervical sarcoma (Table 1). Stage I patients either received adjuvant radiotherapy after radical hysterectomy (6), or had vault recurrence after surgery (4), or were not suitable for surgery for various reasons (6). Only 4 patients in stage IIA were receiving adjuvant radiotherapy after radical surgery. All other patients were treatment naı¨ve.
Radiotherapy treatment A total of 208 patients were invited to participate during treatment. All the patients reported in this study received EBRT using a Cobalt 60 (Siemens or Theratron T280) machine, via parallel-opposed anterior and posterior fields (AP/ PA). The field sizes were adopted depending on the FIGO clinical stage of the disease. Patients in stage I – IIA were treated with 15 15 cm portals (at patient’s surface) and 18 15 cm for stage IIB, III and IV. Most patients (96.5%) received a dose of 40 – 50 Gy to point A. Fractionation was 1.8 – 2.0 Gy tumor dose daily, 5 fractions per week for 5 weeks with 2 days rest from treatment during the weekend. Point A was defined as 2 cm above the external os and 2 cm lateral to the uterine canal. About 10% (20/208) of the patients received high dose fractions, 3.6 – 4.0 Gy as emergency radiotherapy to stop vaginal bleeding and thereafter, they were continued on regular fractions. After EBRT, patients were referred to neighboring countries for brachytherapy. None of the patient reported in this study received brachytherapy due to nonavailability. A treatment gap of 6 or more days was recorded as treatment interruption and the reason for the interruption was enquired. This duration was based on the findings of Wilson et al.’s study that showed that cancer of the cervix has short doubling time with Tpots ranging from 3 to 5 days [20]. Treatment interruptions were categorized as related to social reasons, toxicity, severe anemia, treatment machine maintenance, intercurrent morbidity and others. Once EBRT was initiated, patients were evaluated weekly and 1 month after completing EBRT for radiation-related acute toxicity in the GIT, skin and GUT systems. Acute reactions were reported using Radiation Therapy Oncology Group (RTOG) grading [21]. In this study, pelvic tumor control was documented by physical examination [5,22] and verified histologically. The
Table 1 Patient characteristics by HIV status (N = 208) Variable
All patients n/N (%)
HIV-positive n/N (%)
HIV-negative n/N (%)
Age (years)
47.6 T 12.5
38.1 T 8.6
50.0 T 12.1
10 (5.8) 60 (35.1) 101 (59.1)
0 15 (48.4) 16 (51.6)
10 (7.4) 45 (33.1) 81 (59.6)
16 (7.7) 98 (47.1) 78 (37.5) 16 (7.7) 164/208 (78.8)
1 (2.4) 26 (63.4) 13 (31.7) 1 (2.4) 31/41 (75.6)
15 (9.0) 72 (43.1) 65 (38.9) 15 (9.0) (133/167 (79.6)
0.6 – 2.8
0.571
181 (87) 11 (5.3) 5 (2.4) 1 (0.5) 10 (4.8) 119/159 (74.8) 87/205 (42.4)
33 (80.5) 1 (2.4) 0 1 (2.4) 6 (14.6) 27/33 (81.8) 24/38 (63.2)
148 (88.6) 10 (6.0) 5 (3.0) 0 4 (2.4) 92/126 (73.0) 63/167 (37.7)
1.7 (0.6 – 4.4) 2.8 (1.4 – 5.9)
0.300 0.004
Pretreatment Karnofsky score 20% 50% 100% FIGO clinical stage Stage I Stage II Stage III Stage IV Stage IIB and above Histological cell type Squamous cell Adenocarcinoma Adenosquamous Sarcoma Anaplastic Poorly differentiated Hemoglobin 10g/dl
OR, Odds ratio; CI, Confidence interval.
OR (95% CI)
P value <0.001
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
407
Table 2 Treatment characteristics by HIV status (N = 208) Variable Radiation fractions used in (Gy) 1.6 1.8 2.0 2.5 3.0 Emergency radiation to stop vaginal bleeding Prescribed radiation dose (Gy) Radiation dose received (Gy) Treatment interruptions Treatment duration (days) >50 days of treatment Needed blood transfusion
All patients n/N (%)
HIV-positive n/N (%)
HIV-negative n/N (%)
2 (1.0) 40 (19.2) 160 (76.9) 4 (1.9) 2 (1.0) 20/208 (9.6) 49.4 T 2.2 46.8 T 7.2 69/208 (33.2) 42.7 T 11.9 38/208 (18.3) 51/208 (24.5)
1 (2.4) 9 (22.0) 3 (75.6) 0 0 6/41 (14.6) 49.3 T 2.3 46.8 T 7.2 20/41 (48.8) 42.7 T 11.9 7/41 (17.1) 14/41 (34.1)
1 (0.6) 31 (18.6) 129 (77.2) 4 (2.4) 2 (1.2) 14/167 (8.4) 49.6 T 1.4 46.7 T 7.7 49/167 (29.3) 42.5 T 12.0 31/167 (18.6) 37/167 (22.2)
RR (95% CI)
P value
1.9 (0.7 – 5.2)
0.224 0.526 0.961 0.018 0.926 0.825 0.110
2.3 (1.1 – 4.6) 1.1 (0.4 – 2.7) 1.8 (0.9 – 3.8)
RR, Relative risk; CI, Confidence interval. primary end points of this study were acute morbidity (irradiation related acute toxicity) at the end of EBRT and pelvic tumor control at 7 months from the last day of EBRT treatment. Documentation of acute morbidity and pelvic tumor control was done as a routine. The clinicians were blinded with regard to the HIV serostatus of the patients.
Statistical analysis With a 95% confidence interval, power of 80% and a ratio of HIV-positive to HIV-negative patients of 1:4, we needed to evaluate 24 HIV-positive and 96 HIV-negative patients at 7-month post-EBRT to detect a 2-fold difference in pelvic tumor control. Loss to follow-up was estimated at 50%, thus double the sample size was to be enrolled. Data were entered and analyzed using SPSS version 10. Pearson’s Chi square test or where appropriate Fisher’s exact tests were used in univariate analysis for categorical variables. For continuous variables, the independent sample t test was used depending on Levene’s test for assumed equality of variance at 5% level of significance. Multivariate analysis logistic regression model included variables significant at univariate analysis or those that may have a biological influence. Relative risk (RR) in univariate analysis and adjusted relative risk (ARR) on multivariate analysis were also computed. Cox proportional hazard models was used to estimate the hazard of pelvic tumor. The Kenyatta National Hospital Research and Ethics committee and University of Nairobi approved the study.
Results About 20% (41/208) of the patients enrolled were HIVpositive. All HIV-positive patients were asymptomatic for HIV infection. The mean age was 47.6 years with a range of 23 –80
years. HIV-positive women were significantly younger than HIV-negative patients (38 vs. 50 years, P < 0.001). Table 1 shows patient characteristics. More than half of the patients had Karnofsky performance score of 100% although data were missing for 17.8% (37/208) patients. Mean hemoglobin concentration at enrolment for the 208 patients was 10.9 g/dl with a range of 4 to 17 g/dl. HIV infected patients were 3 times more likely to have Hemoglobin < 10 g/dl as compared to HIVnegative patients (63.2% vs. 37.7%, OR 2.8) (Table 1). Table 2 shows treatment characteristics. Most of the patients received 2.0 Gy fraction per day. About 10.0% needed emergency radiation given as a dose of 3.6 to 4.0 Gy to a total dose of 10.8 to 12.0 Gy to stop vaginal bleeding. Mean dosage of EBRT prescribed was 49.4 Gy with a range of 40 to 60 Gy, while mean dose received was 46.8 Gy (range, 10 to 60 Gy). Mean duration of treatment was 42.7 days with a range of 28 to 90 days. There was no difference in the prescribed, received dosage or duration of treatment between HIV-positive and -negative patients. Thirty-three percent had treatment interruptions of 6 or more days. About 74.0% (51/69) of treatment interruption were due to severe (grade 4) irradiation acute toxicity, 20.3% (14/69) severe anemia (Hb < 7 g/dl) while three patients (4.3%) interrupted treatment due to social reasons and one (1.4%) lost to follow-up. About 49% of HIVpositive as compared to 29% of HIV-negative patients had treatment interruptions (RR 2.3). After controlling for hemo-
Table 3 Acute toxicity (grade 3 – 4) following external beam radiotherapy (EBRT): all patients and by HIV status (N = 208) Variable
All patients n/N (%)
HIV-positive n/N (%)
HIV-negative n/N (%)
RR (95% CI)
P value
Overall toxicity Skin toxicity Gastrointestinal toxicity Genitourinary toxicity Toxicity in: One system Two systems Three systems Total Toxicity in all 3 systems
111/208 70/208 83/208 16/208
(53.4) (33.7) (39.9) (7.7)
21/41 16/41 14/41 8/41
(51.2) (39.0) (34.1) (19.5)
90/167 54/167 69/167 8/167
(53.9) (32.3) (41.3) (4.8)
1.1 1.3 1.4 4.8
(0.6 – 2.2) (0.7 – 2.7) (0.7 – 2.8) (1.7 – 13.8)
0.759 0.417 0.401 0.002
61 41 9 111 9/111
(29.3) (19.7) (4.3) (100) (8.1)
9 7 5 21 5/21
(42.9) (33.3) (23.8) (100) (23.8)
52 34 4 90 4/90
(57.8) (37.8) (4.4) (100) (4.4)
6.7 (1.6 – 27.8)
0.003
RR, Relative risk; CI, Confidence interval.
408
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
Table 4 Determinants of loss to follow-up Variable
Patients on follow-up n/N = 117 (%)
Patients lost to follow-up n/N = 91 (%)
Age years HIV-positive Treatment interruptions Stage IIB+ Well-differentiated tumor
47.7 T 11.4 24/117 (20.5) 39/117 (33.3) 90/117 (76.9) 17/90 (18.9)
47.4 T 13.8 17/91 (18.7) 30/91 (33.0) 74/91 (81.3) 23/69 (33.3)
RR
95% CI
P value
1.1 1.0 1.3 2.1
0.6 – 2.2 0.6 – 1.8 0.7 – 2.6 1.0 – 4.4
0.853 0.742 0.956 0.441 0.037
RR, Relative risk; CI, Confidence interval.
globin, HIV infected patients were still more likely to have treatment interruptions (ARR 2.2, 95% CI 1.1 –4.5, P = 0.039). Table 3 shows acute toxicity following EBRT. Overall, 53.4% (111/208) patients had irradiation-related acute toxicity, grade 3 –4. About 4.0% had acute toxicity involving the skin, GIT and GUT systems. Most of the acute reactions were involving the gastrointestinal system (39.9%). HIV infected patients had a 5-fold higher relative risk of acute toxicity involving GUT system (RR 4.8). Among patients with radiation-related toxicity, 23.8% of HIV-positive patients as compared to 4.4% of HIV-negative patients had the three systems (skin, GIT and GUT) involved (RR 6.7). Of the 208 patients evaluated during treatment, 56.3% were reviewed at 4 and 7 months. Table 4 shows comparisons between the patients with continuous follow and those lost to follow-up. Patients on follow-up were similar to those lost to follow-up in mean age, HIV seropositivity, FIGO clinical stage, treatment interruptions and treatment duration. However, patients lost to follow-up were 2 times more likely to have well-differentiated tumors than those on follow-up ( P = 0.037). Of the 117 patients on follow-up, 18.8% (22/117) clinically had residual tumor at 4 to 7 months after EBRT. Table 5 shows correlates of residual tumors. Forty-one percent of the patients with residual tumor as compared to 16% without residual tumor were HIV-positive (RR 3.7). Patients with treatment duration more than 50 days were 5 times more likely to have residual tumor as compared to those whose treatment duration was less than 50 days (RR 4.8). Total duration of treatment was about 10 days longer for patients with residual tumor (51.8 vs. 41.7 days, P =
0.001). On multivariate logistic regression analysis including FIGO clinical stage, HIV infection, histologic differentiation, need for blood transfusion before or during treatment, pretreatment hemoglobin, radiation dose received and treatment duration, HIV infection was significantly and independently associated with a 6-fold increased risk of residual tumor at 4 – 7 months post-EBRT (ARR 6.2). Duration of treatment was also a significant risk factor for residual tumor (P = 0.016). The hazard of having residual pelvic tumor after initial EBRT was 3.1 times larger for HIV-positive than for HIV-negative patients, P = 0.014. Discussion Radiation therapy is a very effective treatment for patients with carcinoma of the uterine cervix. It is the most common treatment modality for cervical cancer in Kenya where the majority of patients present with advanced disease, FIGO stage IIB and above are not suitable for surgery [22 – 24]. It is also an alternative to surgery for selected patients with similar outcome in survival and tumor control [25]. Several factors influence radiation therapy-related morbidity. These include dose of irradiation, quality of the intracavitary insertions, type of applicator used, proportion of external beam or brachytherapy dose delivered, combination therapy and treatment duration [1,2,26,27]. Host-related factors such as age of the patient, comorbidity states such as pelvic inflammatory disease and diabetes mellitus are also important [1,2,26,27]. Overall, 53% of the patients in our study had irradiationrelated acute toxicity, either skin, GIT or GUT systems. There
Table 5 Determinants of residual tumor at 7 months (N = 117)
HIV infection Stage IIB+ Poorly differentiated Needed blood transfusion Hemoglobin (g/dl) Radiation dose received (Gy) Treatment days >50 days of treatment
Residual tumor N = 22 (%)
No residual tumor N = 95 (%)
Univariate analysis RR (95% CI)
P value
ARR (95% CI)
P value
9/22 (40.9) 20/22 (90.9) 14/17 (82.4) 6/22 (27.3) 11.3 T 2.0 46.2 T 7.7 51.8 T 19.0 10/24 (42)
15/95 (15.8) 71/95 (73.7) 59/73 (80.8) 28/95 (29.5) 11.3 T 2.2 47.7 T 6.0 41.7 T 10.7 12/93 (12.9)
3.7 3.61 1.1 1.1
0.009 0.084 0.885 0.838 0.922 0.323 0.001 <0.001
6.2 3.6 1.0 1.1
0.018 0.152 0.962 0.898 0.568 0.119 0.016
(1.3 – 10.2) (0.8 – 16.4) (0.3 – 4.4) (0.4 – 3.1)
4.4 (1.8 – 13.3)
SD, Standard deviation, RR, Relative risk, CI, Confidence interval ARR, Adjusted RR. * Not included in logistic model because of colinearity with average duration of treatment.
Multivariate analysis
(1.4 – 28.1) (0.6 – 20.1) (0.2 – 5.1) (0.2 – 5.3)
*
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
are few clinical and animal based studies on early morbidity in relation to radiation therapy [8,28 –30]. The rate of early complication in our study is not comparable to other studies. In our study, patients received EBRT only, while in other studies EBRT, brachytherapy and chemotherapy were used [8,22,30]. Pedersen et al. [30] reported on 442 cervical cancer patients, 77% of whom had early complications, mostly in the GIT, which is similar to our findings. In Sood et al.’s [27] study, 4% of 24 patients treated with EBRT and brachytherapy had grade 3 –4, acute toxicity as compared to 80% of 25 patients who in addition got chemotherapy. The risk of radiation-related acute toxicity involving skin, GIT and GUT combined was 7-fold higher in HIV infected patients than in uninfected patients. There was no increased GIT toxicity in HIV infected patients which is reassuring. Chronic diarrhea is a significant morbidity in advanced HIV infection, which could theoretically worsen in the setting of irradiation proctitis. Diarrhea is a common manifestation of acute injury of the small and large intestinal mucosa during pelvic radiation. The lack of increased morbidity in GIT among HIV infected cervical cancer patients could be related to the relative immunocompetence in our patients. The majorities of the HIV-positive patients were asymptomatic for HIV infection and tested positive for the first time during this study. Maiman et al. [14] reported that most of the cervical cancer HIV-positive patients in Brooklyn, USA, were asymptomatic of HIV and were identified by routine screening. Frutcher et al. [13] suggested that the relative immunocompetence of HIV infected cervical cancer patients indicates that they acquired HIV infection after the initiation of the neoplastic process rather than developing neoplasia as a result of immunodeficiency. What has consistently been documented is the lower immunosuppression (CD4 cell count) in cervical cancer patients as compared to patients with other AIDS defining cancers [19,31]. Data from similar population of ICC patients showed relative immunocompetence with HIV infected cervical cancer patients at an average CD4 cell count of 535 cells/mm3 which is well above the 200 cell/mm3 CD4 cell count when major GIT opportunistic infections set in [32]. Hoffman et al. [17] reported markedly increased morbidity if pretreatment CD4 cell count was < 200 cells/mm3 for anal cancer patients following standard therapy. About 19% of our patients had pelvic failure (residual tumor) at 4– 7 months. HIV infection was associated with a 6fold higher risk of pelvic failure. Thirty-eight percent of HIV infected patients as compared to 12% of non-infected patients had pelvic failure. Other studies have reported a 5-year pelvic failure rate of 10 –74% depending on the clinical stage [1,2]. There are two possible explanations of the association of pelvic failure with HIV infection. HIV infection is associated with anemia [33,34]. Chronic and transient hypoxia makes tumors radioresistant and is an adverse prognostic factor for both local control and survival outcome [35 – 37]. Therefore, HIV infection could be a proxy indicator of anemia. Indeed, HIV was significantly and independently associated with anemia in our study. However, even after controlling for anemia, HIV
409
infected patients still had more pelvic failure than uninfected patients. Another hypothesis is that HIV infection may alter cervical cancer cell kinetics resulting in more radioresistant tumor. HIV infected patients had more treatment interruptions than noninfected patients. Although we did not show an association between treatment interruptions and residual tumor in our study, the overall treatment time was less than 50 days. Significant loco-regional failure due to treatment prolongation is often seen if the treatment period is beyond 50 days [4,30]. Treatment duration of longer than 50 days in our study was associated with a 5-fold higher risk of residual tumor. It is possible that the treatment interruptions associated with HIV infection could have resulted in pelvic failure. Treatment prolongation contributes to loss of loco-regional control by allowing clonogenic repopulation [5]. It is uncertain at the moment whether HIV infection is associated with different cell kinetics or radioresistant tumors. Like in head and neck tumors, time to failure in cervical carcinoma is often within 2 years. This may be a reflection of rapid growth kinetics [5]. Maiman [38] reported that HIV infected cervical cancer patients died of complications due to cancer and not to HIV on average of 10 months despite appropriate treatment. Kim et al. [18] reported that median time to cancer-related death of anal cancer patients HIV-positive was 1.4 years as compared to 5.3 years for HIVnegative patients. This suggests that HPV – HIV-related disease is aggressive, which could be a manifestation of rapid growth kinetics although there is no laboratory data to support this hypothesis. However, in anal and cervical cancer, HIV infection is associated with microsatellite instability and loss of heterozygosity (LOH), which is probably a molecular explanation of the aggressiveness of HPV –HIV-related cancer [39,40]. Recurrent LOH at D2S123 has been demonstrated in HIV-positive patients with radioresistant anal squamous-cell carcinoma [39]. Certain limitations need to be considered in interpretation of data from this study. Although we had sufficient power, the absolute number of patients included is small, thus making the study less robust. These findings need to be replicated in more extensive studies. Documentation of the parameters reported is prone to observer variation though all the documentation for this study was done by one person (PG). As observed by Sismond et al. [41] regarding late morbidity, literature on the subject shows large variations in reported complications and definition of complications. The other limitation was the use of clinical documentation of pelvic failure, which is subjective. More objective measures such as pre- and posttreatment computerized pelvic tomogram (CT-scan) or magnetic nuclear resonance imaging (MRI) [42,43] could not be utilized due to cost limitations. The third limitation is that the patient lost to follow-up had well-differentiated tumors. This could have resulted in overestimation of pelvic failure since well-differentiated tumors have a better outcome than poorly differentiated tumors. Lastly, data from this study relate only to EBRT, and only the patients who were followed up, thus it may not be generalized to other populations receiving more than EBRT for cervical cancer.
410
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
In conclusion, this study has shown that HIV infection is associated with more extensive irradiation-related acute toxicity, treatment interruptions and pelvic failure to control cervical cancer following EBRT. Our study and others on anal and cervical cancer suggest that HIV infection is an adverse prognostic factor. Our results need to be replicated in more rigorous extensive studies. Studies to determine survival outcome of ICC HIV-positive patients are needed. These studies will assist to design appropriate treatment strategies for invasive cervical cancer, HIV-positive patients, as these patients are likely to increase in numbers as HIV pandemic continues growing. Acknowledgments We thank VLIR (Flemish Interuniversity Council), Belgium for financial support, the Director of Kenyatta National hospital for allowing access to patients, staff in radiotherapy unit and laboratory staff in Department of Medical Microbiology, University of Nairobi. Special thanks go to Regina Kilonzo, and Jayne Mbithi, the study staff and S. Vansteelandt, from the Department of Applied Mathematics and Informatics, University of Ghent, Belgium, for their contribution to data analysis. References [1] Lanciano RM, Pajak TF, Martz K, Hanks GE. The influence of treatment time on outcome for squamous cell cancer of the uterine cervix treated with radiation: a patterns-of-care study. Int J Radiat Oncol Biol Phys 1993;25:391 – 7. [2] Perez CA, Grigsby PW, Camel HM, et al. Irradiation alone or combined with surgery in stage IB, IIA, and IIB carcinoma of uterine cervix: update of a nonrandomized comparison. Int J Radiat Oncol Biol Phys 1995;31:703 – 16. [3] Fyles A, Keane TJ, Barton M, Simm J. The effect of treatment duration in the local control of cervix cancer. Radiother Oncol 1992;25:273 – 9. [4] Girinsky T, Rey A, Roche B, Haie C, Gerbaulet A, Randrianarivello H, et al. Overall treatment in advanced cervical carcinomas: a critical parameter in treatment outcome. Int J Radiat Oncol Bio Phys 1993;27:1051 – 6. [5] Petereit DG, Sarkaria JN, Chappell R, Flowee JF, Hartmann TJ, Kinsella TJ, et al. The adverse effect of treatment prolongation in cervical carcinoma. Int J Radiat Oncol Biol Phys 1995;32:1301 – 7. [6] Erridge SC, Kerr GR, Downing D, Duncan W, Price A. The effect of overall treatment time on the survival and toxicity of the radical radiotherapy for cervical carcinoma. Radiother Oncol 2002;63:59 – 66. [7] Schultheiss TE, Lee WR, Hunt MA, et al. Late GI and GU complication in the treatment of prostate cancer. Int J Radiat Oncol Biol Phys 1997;37:3 – 11. [8] Wang CJ, Leung SW, Chen HC, et al. The correlation of acute toxicity and late rectal injury in radiotherapy for cervical carcinoma: evidence suggestive of consequential late effect. Int J Radiat Oncol Biol Phys 1998;40:85 – 91. [9] Peters LJ, Ang KK, Thames HD. Accelerated fractionations in the radiation treatment of head and neck cancer. Acta Oncol 1988;27:185 – 94. [10] Peters KJ, Ang KK. The role of altered fractionation in head and neck cancers. Semin Radiat Oncol 1992;2:180 – 94. [11] Buehler JW, Ward JW. A new definition for AIDS surveillance. Ann Int Med 1993;118:390 – 2. [12] Gemignani M, Maiman M, Fruchter RG, Arrastia CD, Gibbon D, Ellison T. CD4 lymphocytes in women with invasive and preinvasive cervical neoplasia. Gynecol Oncol 1995;59:364 – 9. [13] Frutcher RG, Maiman M, Sedlis A, Bartley L, Camilien L, Arrastia CD. Multiple recurrences of cervical intraepithelial neoplasia in women with the human immunodeficiency virus. Obstet Gynecol 1996;87:338 – 44.
[14] Maiman M, Fruchter RG, Guy L, et al. Human immunodeficiency virus infection and invasive carcinoma. Cancer 1993;71:402 – 6. [15] Matorras R, Ariceta JM, Rementeria A, et al. Human immunodeficiency virus-induced immunosuppression: a risk factor for human papillomavirus infection. Am J Obstet Gynecol 1991;164:42 – 4. [16] Wright TC, Koulos J, Schnoll F, et al. Cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus: outcome after loop electrosurgical excision. Gynecol Oncol 1994;55:253 – 8. [17] Hoffman R, Welton ML, Klencke B, Weinberg V, Krieg R. The significance of pretreatment CD4 count on the outcome and treatment tolerance of HIV-positive patients with anal cancer. Int J Radiat Oncol Biol Phys 1999;44:127 – 31. [18] Kim JH, Sarani B, Orkin BA, Young HA, White J, Tannebaum I, et al. HIV-positive patients with anal carcinoma have poorer treatment tolerance and outcome than HIV-negative patients. Dis Colon Rectum 2001;44: 1496 – 502. [19] Maiman M, Fruchter MG, Clark M, Arrastia CD, Mathews R, Gates E. Cervical cancer as an AIDS-defining illness. Obstet Gynecol 1997; 89:76 – 80. [20] Wilson GD, McNally NJ, Disches S, Saunders MI, Desrochers C, Lewis AA, et al. Measurement of cell kinetics in human tumors and flow cytometry. Br J Cancer 1988:423 – 31. [21] Cox JD, Stetz J, Pajak TF. Toxicity criteria of the radiation therapy oncology group (RTOG) and the European organization for research and treatment of cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31: 1341 – 6. [22] Sedlis A, Bundy BN, Rotman MZ, Lentz SS, Muderspach LI, Zaino RJ. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: a gynecology oncology group study. Gynecol Oncol 1999;73:177 – 83. [23] Gichangi P, De Vuyst H, Estambale B, Rogo K, Bwayo J, Temmerman M. HIV and cervical cancer in Kenya. Int J Gynecol Obstet 2002;76:55 – 63. [24] K.O. Rogo, Human Papillomavirus and Human Immunodeficiency virus infection in relation to cervical cancer: studies and observations of basic cliQ nical and epidemiological aspects of cancer of the cervix with special refeQ rences to Kenya, Africa. PhD thesis, University of Umea, Sweden; 1990. [25] Landoni F, Maneo A, Colombo A, et al. Randomized study of radical surgery versus radiotherapy for stage IB – IIB cervical cancer. Lancet 1997;350:535 – 40. [26] Roeske JC, Mundt AJ, Halpern H, et al. Late rectal sequelae following definitive radiation therapy for carcinoma of the uterine cervix: a dosimetric analysis. Int J Radiat Oncol Biol Phys 1997;37:351 – 8. [27] Sood BM, Gorla G, Gupta S, Garg M, Deore S, Runowicz CD, et al. Two fractions of high-dose-rate brachytherapy in the management of cervix cancer: clinical experience with and without chemotherapy. Int J Radiat Oncol Biol Phys 2002;53:702 – 6. [28] Langberg CW, Sauer T, Reitan JB, Hauer-Jensen M. Tolerance of rat small intestine to localized single dose and fractionated irradiation. Acta Oncol 1992;31:781 – 7. [29] Hauer-Jensen M, Poulakos L, Osborne JW. Intestinal complications following accelerated fractionated x-irradiation. An experimental study in the rat. Acta Oncol 1990;29:229 – 34. [30] Pedersen D, Bentzen SM, Overgaard J. Early and late radiotherapeutic morbidity in 442 consecutive patients with locally advanced carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1994;29:941 – 52. [31] Klevens RM, Fleming PL, Mays MA, et al. Characteristics of women with AIDS and invasive cervical cancer. Obstet Gynecol 1996;88:269 – 73. [32] Gichangi PB, Bwayo J, Estambale B, De Vuyst H, Ojwang S, Rogo K, et al. Impact of HIV infection on invasive cervical cancer in Kenyan women. AIDS 2003;17:1963 – 8. [33] Coyle TE. Hematologic complications of human immunodeficiency virus infection and the acquired immunodeficiency syndrome. Med Clin North Am 1997;81:449 – 70. [34] Sullivan PS, Hanson DL, Chu SY, Jones JL, Ward JW. Epidemiology of anemia in human immunodeficiency virus (HIV)-infected persons: results from the multistate adult and adolescent spectrum of HIV disease surveillance project. Blood 1998;91:301 – 8.
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411 [35] Fyles M, Milosevic D, Hedley M, Pintilie W, Levin L, Manchul L, et al. Tumor hypoxia has independent predictor impact only in patients with node-negative cervix cancer. J Clin Oncol 2002;20:680 – 7. [36] Fyles AW, Milosevic M, Wong R, et al. Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 1998;48:149 – 56. [37] Knoncke TH, Weitmann HD, Feldmann HJ, et al. Intratumoral pO2 measurements as predictive assay in the treatment of carcinoma of the uterine cervix. Radiother Oncol 1999;53:99 – 104. [38] Maiman M. Management of cervical neoplasia in Human Immunodeficiency Virus-infected women. Monogr Natl Cancer Inst 1998;23:43 – 9. [39] Gervaz P, Efron J, Poza AF, et al. Loss of heterozygosity and HIV infection in patients with anal squamous-cell carcinoma. Dis Colon Rectum 2001;44:1503 – 5.
411
[40] Wistuba II, Syed S, Behrens C, Duong M, Milchgrub S, Muller CY, et al. Comparison of molecular changes in cervical intraepithelial neoplasia in HIV-positive and HIV-indeterminate subjects. Gynecol Oncol 1999;74: 519 – 26. [41] Sismondi P, Sinistrero G, Zola P, Volpe T, Ferraris R, Castelli GL, et al. Complications of uterine cervix carcinoma treatment: the problem of a uniform classification. Radiother Oncol 1989;14:9 – 17. [42] Mayr NA, Taoka T, Yuh WT, et al. Method and timing of tumor volume measurement for outcome prediction in cervical cancer using magnetic resonance imaging. Int J Radiat Oncol Biol Phys 2002;52:14 – 22. [43] Mayr NA, Yuh WTC, Zheng J, Ehrhardt JC, Sorosky JI, Magnotta VA, et al. Tumor size evaluated by pelvic examination compared with 3-D MR quantitative analysis in the prediction of outcome for cervical cancer. Int J Radiat Oncol Biol Phys 1997;39:395 – 404.
HIV impact on acute morbidity and pelvic tumor control following radiotherapy for cervical cancerB Peter Gichangi a,*, Job Bwayo b, Benson Estambale b, Khama Rogo c, Eliud Njuguna d, Shadrack Ojwang e, Marleen Temmerman f a
Department of Human Anatomy and Obstetrics and Gynecology, University of Nairobi, P.O. Box 2631KNH 00202, Nairobi, Kenya b Department of Medical Microbiology, University of Nairobi, Nairobi, Kenya c Nairobi Oncology Center, Nairobi, Kenya d Radiotherapy Unit, Kenyatta National Hospital, Nairobi, Kenya e Department of Obstetrics and Gynecology, University of Nairobi, Nairobi, Kenya f International Center for Reproductive Health, Ghent University, Ghent, Belgium Received 13 April 2005 Available online 4 November 2005
Abstract Objective. To determine the impact of HIV infection on acute morbidity and pelvic tumor control following external beam radiotherapy (EBRT) for cervical cancer. Method. 218 patients receiving EBRT who also had HIV testing after informed consent was obtained were evaluated. Acute treatment toxicity was documented weekly during treatment and 1 month post-EBRT. Pelvic tumor control was documented at 4 and 7 months post-EBRT. Clinicians were blinded for HIV results. Results. About 20% of the patients were HIV-positive. Overall, 53.4% of the patients had radiation-related acute toxicity (grade 3 – 4). HIV infection was associated with a 7-fold higher risk of multisystem toxicity: skin, gastrointestinal tract (GIT) and genitourinary tract (GUT) systems. It was also an independent risk factor for treatment interruptions (adjusted relative risk 2.2). About 19% of the patients had residual tumor at 4 and 7 months post-EBRT. HIV infection was independently and significantly associated with 6-fold higher risk of residual tumor post-EBRT. The hazard ratio of having residual tumor after initial EBRT was 3.1-times larger for HIV-positive than for HIV-negative patients (P = 0.014). Conclusion. HIV is associated with increased risk of multisystem radiation-related toxicity; treatment interruptions and pelvic failure (residual tumor) following EBRT. HIV infection is an adverse prognostic factor for outcome of cervical cancer treatment. D 2005 Elsevier Inc. All rights reserved. Keywords: HIV; Cervical cancer; Radiotherapy; Outcome
Introduction Treatment of cervical cancer even in optimal conditions is associated with loco-regional failure. Pelvic failure rates of between 10 and 74% have been reported depending on the clinical stage [1,2]. Reasons for failure include tumor-related factors such as large volume of disease, FIGO clinical stage and treatment-related factors such as the inability to deliver adequate radiation doses, treatment prolongation and age of the i
This study was supported by Flemish Interuniversity Council (VLIR). * Corresponding author. E-mail addresses: [email protected], [email protected] (P. Gichangi). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.10.006
patient [1,3 – 5]. The overall treatment time has an impact on acute/late morbidity and survival. Petereit et al. [5] showed a decrease of 0.7%/day of local control and 0.6%/day of survival for each additional day of treatment beyond 55 days in cervical cancer patients. Erridge et al. [6] showed no significant adverse effect in survival if cervical cancer treatment was less than 7 weeks but patients treated over a short period had increased incidence of late morbidity. One of the common causes of treatment prolongation is comorbidity and acute toxicity. HIV infection is now a common comorbid infection in patients with invasive cervical cancer in the African setting. HIV infection in advanced stage is associated with certain opportunistic infections/conditions such as severe diarrhea, anemia, nonspecific skin dermatosis and genital tract infec-
406
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
tions. Chronic diarrhea of HIV infection may worsen in the setting of radiation proctitis thereby contributing to treatment interruption. Acute GIT toxicity, especially diarrhea, is significantly associated with an increased risk of late rectal injury [7,8], a phenomenon termed consequential late effect [8– 10]. HIV infection is associated with more aggressive invasive cervical cancer, and higher treatment failures and recurrences following standard treatment [11 – 16]. Studies on anal cancer, an HPV-related malignancy with increased frequency in HIV infected patients show that HIV infected patients have poorer tolerance to treatment and shorter time (1.4 years) to cancer-related death [17,18]. It is ill understood how HIV infection confers such an adverse effect on treatment outcome of cervical cancer. It is however apparent that chronic and opportunistic infections associated with HIV infection may adversely affect radiation treatment by causing treatment interruptions or acute toxicity and thereby adverse outcome. HIV infected cervical cancer patients untreated for HIV have a short survival of about 10 months [19], which is too short to evaluate the late morbidity associated with radiotherapy treatment. This study was therefore undertaken to evaluate the impact of HIV infection on acute morbidity related to EBRT and pelvic tumor control after EBRT in an African setting where both conditions are prevalent. Materials and methods Patient population Between January 2000 and April 2002, 208 consecutive patients who were treated with EBRT at Kenyatta National Hospital (KNH) radiotherapy unit and who consented to have HIV testing, were included in the study. Recruited patients received pre- and posttest HIV counseling. Pretreatment evaluation for all patients consisted of history, complete physical examination, examination
under anesthetic for clinical staging according to the International Federation of Gynecology and Obstetrics 1995 (FIGO) system, hemoglobin determination, chest X-ray and Karnofsky score determination. The tumors were classified as Stage 1 (16), II (98), III (78) and IV (16). All tumors were histologically verified. Squamous tumors constituted 87.0%, adenocarcinoma 5.3%, adenosquamous carcinoma 2.4% and anaplastic carcinoma 4.8%. One patient (0.5%) had cervical sarcoma (Table 1). Stage I patients either received adjuvant radiotherapy after radical hysterectomy (6), or had vault recurrence after surgery (4), or were not suitable for surgery for various reasons (6). Only 4 patients in stage IIA were receiving adjuvant radiotherapy after radical surgery. All other patients were treatment naı¨ve.
Radiotherapy treatment A total of 208 patients were invited to participate during treatment. All the patients reported in this study received EBRT using a Cobalt 60 (Siemens or Theratron T280) machine, via parallel-opposed anterior and posterior fields (AP/ PA). The field sizes were adopted depending on the FIGO clinical stage of the disease. Patients in stage I – IIA were treated with 15 15 cm portals (at patient’s surface) and 18 15 cm for stage IIB, III and IV. Most patients (96.5%) received a dose of 40 – 50 Gy to point A. Fractionation was 1.8 – 2.0 Gy tumor dose daily, 5 fractions per week for 5 weeks with 2 days rest from treatment during the weekend. Point A was defined as 2 cm above the external os and 2 cm lateral to the uterine canal. About 10% (20/208) of the patients received high dose fractions, 3.6 – 4.0 Gy as emergency radiotherapy to stop vaginal bleeding and thereafter, they were continued on regular fractions. After EBRT, patients were referred to neighboring countries for brachytherapy. None of the patient reported in this study received brachytherapy due to nonavailability. A treatment gap of 6 or more days was recorded as treatment interruption and the reason for the interruption was enquired. This duration was based on the findings of Wilson et al.’s study that showed that cancer of the cervix has short doubling time with Tpots ranging from 3 to 5 days [20]. Treatment interruptions were categorized as related to social reasons, toxicity, severe anemia, treatment machine maintenance, intercurrent morbidity and others. Once EBRT was initiated, patients were evaluated weekly and 1 month after completing EBRT for radiation-related acute toxicity in the GIT, skin and GUT systems. Acute reactions were reported using Radiation Therapy Oncology Group (RTOG) grading [21]. In this study, pelvic tumor control was documented by physical examination [5,22] and verified histologically. The
Table 1 Patient characteristics by HIV status (N = 208) Variable
All patients n/N (%)
HIV-positive n/N (%)
HIV-negative n/N (%)
Age (years)
47.6 T 12.5
38.1 T 8.6
50.0 T 12.1
10 (5.8) 60 (35.1) 101 (59.1)
0 15 (48.4) 16 (51.6)
10 (7.4) 45 (33.1) 81 (59.6)
16 (7.7) 98 (47.1) 78 (37.5) 16 (7.7) 164/208 (78.8)
1 (2.4) 26 (63.4) 13 (31.7) 1 (2.4) 31/41 (75.6)
15 (9.0) 72 (43.1) 65 (38.9) 15 (9.0) (133/167 (79.6)
0.6 – 2.8
0.571
181 (87) 11 (5.3) 5 (2.4) 1 (0.5) 10 (4.8) 119/159 (74.8) 87/205 (42.4)
33 (80.5) 1 (2.4) 0 1 (2.4) 6 (14.6) 27/33 (81.8) 24/38 (63.2)
148 (88.6) 10 (6.0) 5 (3.0) 0 4 (2.4) 92/126 (73.0) 63/167 (37.7)
1.7 (0.6 – 4.4) 2.8 (1.4 – 5.9)
0.300 0.004
Pretreatment Karnofsky score 20% 50% 100% FIGO clinical stage Stage I Stage II Stage III Stage IV Stage IIB and above Histological cell type Squamous cell Adenocarcinoma Adenosquamous Sarcoma Anaplastic Poorly differentiated Hemoglobin 10g/dl
OR, Odds ratio; CI, Confidence interval.
OR (95% CI)
P value <0.001
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
407
Table 2 Treatment characteristics by HIV status (N = 208) Variable Radiation fractions used in (Gy) 1.6 1.8 2.0 2.5 3.0 Emergency radiation to stop vaginal bleeding Prescribed radiation dose (Gy) Radiation dose received (Gy) Treatment interruptions Treatment duration (days) >50 days of treatment Needed blood transfusion
All patients n/N (%)
HIV-positive n/N (%)
HIV-negative n/N (%)
2 (1.0) 40 (19.2) 160 (76.9) 4 (1.9) 2 (1.0) 20/208 (9.6) 49.4 T 2.2 46.8 T 7.2 69/208 (33.2) 42.7 T 11.9 38/208 (18.3) 51/208 (24.5)
1 (2.4) 9 (22.0) 3 (75.6) 0 0 6/41 (14.6) 49.3 T 2.3 46.8 T 7.2 20/41 (48.8) 42.7 T 11.9 7/41 (17.1) 14/41 (34.1)
1 (0.6) 31 (18.6) 129 (77.2) 4 (2.4) 2 (1.2) 14/167 (8.4) 49.6 T 1.4 46.7 T 7.7 49/167 (29.3) 42.5 T 12.0 31/167 (18.6) 37/167 (22.2)
RR (95% CI)
P value
1.9 (0.7 – 5.2)
0.224 0.526 0.961 0.018 0.926 0.825 0.110
2.3 (1.1 – 4.6) 1.1 (0.4 – 2.7) 1.8 (0.9 – 3.8)
RR, Relative risk; CI, Confidence interval. primary end points of this study were acute morbidity (irradiation related acute toxicity) at the end of EBRT and pelvic tumor control at 7 months from the last day of EBRT treatment. Documentation of acute morbidity and pelvic tumor control was done as a routine. The clinicians were blinded with regard to the HIV serostatus of the patients.
Statistical analysis With a 95% confidence interval, power of 80% and a ratio of HIV-positive to HIV-negative patients of 1:4, we needed to evaluate 24 HIV-positive and 96 HIV-negative patients at 7-month post-EBRT to detect a 2-fold difference in pelvic tumor control. Loss to follow-up was estimated at 50%, thus double the sample size was to be enrolled. Data were entered and analyzed using SPSS version 10. Pearson’s Chi square test or where appropriate Fisher’s exact tests were used in univariate analysis for categorical variables. For continuous variables, the independent sample t test was used depending on Levene’s test for assumed equality of variance at 5% level of significance. Multivariate analysis logistic regression model included variables significant at univariate analysis or those that may have a biological influence. Relative risk (RR) in univariate analysis and adjusted relative risk (ARR) on multivariate analysis were also computed. Cox proportional hazard models was used to estimate the hazard of pelvic tumor. The Kenyatta National Hospital Research and Ethics committee and University of Nairobi approved the study.
Results About 20% (41/208) of the patients enrolled were HIVpositive. All HIV-positive patients were asymptomatic for HIV infection. The mean age was 47.6 years with a range of 23 –80
years. HIV-positive women were significantly younger than HIV-negative patients (38 vs. 50 years, P < 0.001). Table 1 shows patient characteristics. More than half of the patients had Karnofsky performance score of 100% although data were missing for 17.8% (37/208) patients. Mean hemoglobin concentration at enrolment for the 208 patients was 10.9 g/dl with a range of 4 to 17 g/dl. HIV infected patients were 3 times more likely to have Hemoglobin < 10 g/dl as compared to HIVnegative patients (63.2% vs. 37.7%, OR 2.8) (Table 1). Table 2 shows treatment characteristics. Most of the patients received 2.0 Gy fraction per day. About 10.0% needed emergency radiation given as a dose of 3.6 to 4.0 Gy to a total dose of 10.8 to 12.0 Gy to stop vaginal bleeding. Mean dosage of EBRT prescribed was 49.4 Gy with a range of 40 to 60 Gy, while mean dose received was 46.8 Gy (range, 10 to 60 Gy). Mean duration of treatment was 42.7 days with a range of 28 to 90 days. There was no difference in the prescribed, received dosage or duration of treatment between HIV-positive and -negative patients. Thirty-three percent had treatment interruptions of 6 or more days. About 74.0% (51/69) of treatment interruption were due to severe (grade 4) irradiation acute toxicity, 20.3% (14/69) severe anemia (Hb < 7 g/dl) while three patients (4.3%) interrupted treatment due to social reasons and one (1.4%) lost to follow-up. About 49% of HIVpositive as compared to 29% of HIV-negative patients had treatment interruptions (RR 2.3). After controlling for hemo-
Table 3 Acute toxicity (grade 3 – 4) following external beam radiotherapy (EBRT): all patients and by HIV status (N = 208) Variable
All patients n/N (%)
HIV-positive n/N (%)
HIV-negative n/N (%)
RR (95% CI)
P value
Overall toxicity Skin toxicity Gastrointestinal toxicity Genitourinary toxicity Toxicity in: One system Two systems Three systems Total Toxicity in all 3 systems
111/208 70/208 83/208 16/208
(53.4) (33.7) (39.9) (7.7)
21/41 16/41 14/41 8/41
(51.2) (39.0) (34.1) (19.5)
90/167 54/167 69/167 8/167
(53.9) (32.3) (41.3) (4.8)
1.1 1.3 1.4 4.8
(0.6 – 2.2) (0.7 – 2.7) (0.7 – 2.8) (1.7 – 13.8)
0.759 0.417 0.401 0.002
61 41 9 111 9/111
(29.3) (19.7) (4.3) (100) (8.1)
9 7 5 21 5/21
(42.9) (33.3) (23.8) (100) (23.8)
52 34 4 90 4/90
(57.8) (37.8) (4.4) (100) (4.4)
6.7 (1.6 – 27.8)
0.003
RR, Relative risk; CI, Confidence interval.
408
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
Table 4 Determinants of loss to follow-up Variable
Patients on follow-up n/N = 117 (%)
Patients lost to follow-up n/N = 91 (%)
Age years HIV-positive Treatment interruptions Stage IIB+ Well-differentiated tumor
47.7 T 11.4 24/117 (20.5) 39/117 (33.3) 90/117 (76.9) 17/90 (18.9)
47.4 T 13.8 17/91 (18.7) 30/91 (33.0) 74/91 (81.3) 23/69 (33.3)
RR
95% CI
P value
1.1 1.0 1.3 2.1
0.6 – 2.2 0.6 – 1.8 0.7 – 2.6 1.0 – 4.4
0.853 0.742 0.956 0.441 0.037
RR, Relative risk; CI, Confidence interval.
globin, HIV infected patients were still more likely to have treatment interruptions (ARR 2.2, 95% CI 1.1 –4.5, P = 0.039). Table 3 shows acute toxicity following EBRT. Overall, 53.4% (111/208) patients had irradiation-related acute toxicity, grade 3 –4. About 4.0% had acute toxicity involving the skin, GIT and GUT systems. Most of the acute reactions were involving the gastrointestinal system (39.9%). HIV infected patients had a 5-fold higher relative risk of acute toxicity involving GUT system (RR 4.8). Among patients with radiation-related toxicity, 23.8% of HIV-positive patients as compared to 4.4% of HIV-negative patients had the three systems (skin, GIT and GUT) involved (RR 6.7). Of the 208 patients evaluated during treatment, 56.3% were reviewed at 4 and 7 months. Table 4 shows comparisons between the patients with continuous follow and those lost to follow-up. Patients on follow-up were similar to those lost to follow-up in mean age, HIV seropositivity, FIGO clinical stage, treatment interruptions and treatment duration. However, patients lost to follow-up were 2 times more likely to have well-differentiated tumors than those on follow-up ( P = 0.037). Of the 117 patients on follow-up, 18.8% (22/117) clinically had residual tumor at 4 to 7 months after EBRT. Table 5 shows correlates of residual tumors. Forty-one percent of the patients with residual tumor as compared to 16% without residual tumor were HIV-positive (RR 3.7). Patients with treatment duration more than 50 days were 5 times more likely to have residual tumor as compared to those whose treatment duration was less than 50 days (RR 4.8). Total duration of treatment was about 10 days longer for patients with residual tumor (51.8 vs. 41.7 days, P =
0.001). On multivariate logistic regression analysis including FIGO clinical stage, HIV infection, histologic differentiation, need for blood transfusion before or during treatment, pretreatment hemoglobin, radiation dose received and treatment duration, HIV infection was significantly and independently associated with a 6-fold increased risk of residual tumor at 4 – 7 months post-EBRT (ARR 6.2). Duration of treatment was also a significant risk factor for residual tumor (P = 0.016). The hazard of having residual pelvic tumor after initial EBRT was 3.1 times larger for HIV-positive than for HIV-negative patients, P = 0.014. Discussion Radiation therapy is a very effective treatment for patients with carcinoma of the uterine cervix. It is the most common treatment modality for cervical cancer in Kenya where the majority of patients present with advanced disease, FIGO stage IIB and above are not suitable for surgery [22 – 24]. It is also an alternative to surgery for selected patients with similar outcome in survival and tumor control [25]. Several factors influence radiation therapy-related morbidity. These include dose of irradiation, quality of the intracavitary insertions, type of applicator used, proportion of external beam or brachytherapy dose delivered, combination therapy and treatment duration [1,2,26,27]. Host-related factors such as age of the patient, comorbidity states such as pelvic inflammatory disease and diabetes mellitus are also important [1,2,26,27]. Overall, 53% of the patients in our study had irradiationrelated acute toxicity, either skin, GIT or GUT systems. There
Table 5 Determinants of residual tumor at 7 months (N = 117)
HIV infection Stage IIB+ Poorly differentiated Needed blood transfusion Hemoglobin (g/dl) Radiation dose received (Gy) Treatment days >50 days of treatment
Residual tumor N = 22 (%)
No residual tumor N = 95 (%)
Univariate analysis RR (95% CI)
P value
ARR (95% CI)
P value
9/22 (40.9) 20/22 (90.9) 14/17 (82.4) 6/22 (27.3) 11.3 T 2.0 46.2 T 7.7 51.8 T 19.0 10/24 (42)
15/95 (15.8) 71/95 (73.7) 59/73 (80.8) 28/95 (29.5) 11.3 T 2.2 47.7 T 6.0 41.7 T 10.7 12/93 (12.9)
3.7 3.61 1.1 1.1
0.009 0.084 0.885 0.838 0.922 0.323 0.001 <0.001
6.2 3.6 1.0 1.1
0.018 0.152 0.962 0.898 0.568 0.119 0.016
(1.3 – 10.2) (0.8 – 16.4) (0.3 – 4.4) (0.4 – 3.1)
4.4 (1.8 – 13.3)
SD, Standard deviation, RR, Relative risk, CI, Confidence interval ARR, Adjusted RR. * Not included in logistic model because of colinearity with average duration of treatment.
Multivariate analysis
(1.4 – 28.1) (0.6 – 20.1) (0.2 – 5.1) (0.2 – 5.3)
*
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
are few clinical and animal based studies on early morbidity in relation to radiation therapy [8,28 –30]. The rate of early complication in our study is not comparable to other studies. In our study, patients received EBRT only, while in other studies EBRT, brachytherapy and chemotherapy were used [8,22,30]. Pedersen et al. [30] reported on 442 cervical cancer patients, 77% of whom had early complications, mostly in the GIT, which is similar to our findings. In Sood et al.’s [27] study, 4% of 24 patients treated with EBRT and brachytherapy had grade 3 –4, acute toxicity as compared to 80% of 25 patients who in addition got chemotherapy. The risk of radiation-related acute toxicity involving skin, GIT and GUT combined was 7-fold higher in HIV infected patients than in uninfected patients. There was no increased GIT toxicity in HIV infected patients which is reassuring. Chronic diarrhea is a significant morbidity in advanced HIV infection, which could theoretically worsen in the setting of irradiation proctitis. Diarrhea is a common manifestation of acute injury of the small and large intestinal mucosa during pelvic radiation. The lack of increased morbidity in GIT among HIV infected cervical cancer patients could be related to the relative immunocompetence in our patients. The majorities of the HIV-positive patients were asymptomatic for HIV infection and tested positive for the first time during this study. Maiman et al. [14] reported that most of the cervical cancer HIV-positive patients in Brooklyn, USA, were asymptomatic of HIV and were identified by routine screening. Frutcher et al. [13] suggested that the relative immunocompetence of HIV infected cervical cancer patients indicates that they acquired HIV infection after the initiation of the neoplastic process rather than developing neoplasia as a result of immunodeficiency. What has consistently been documented is the lower immunosuppression (CD4 cell count) in cervical cancer patients as compared to patients with other AIDS defining cancers [19,31]. Data from similar population of ICC patients showed relative immunocompetence with HIV infected cervical cancer patients at an average CD4 cell count of 535 cells/mm3 which is well above the 200 cell/mm3 CD4 cell count when major GIT opportunistic infections set in [32]. Hoffman et al. [17] reported markedly increased morbidity if pretreatment CD4 cell count was < 200 cells/mm3 for anal cancer patients following standard therapy. About 19% of our patients had pelvic failure (residual tumor) at 4– 7 months. HIV infection was associated with a 6fold higher risk of pelvic failure. Thirty-eight percent of HIV infected patients as compared to 12% of non-infected patients had pelvic failure. Other studies have reported a 5-year pelvic failure rate of 10 –74% depending on the clinical stage [1,2]. There are two possible explanations of the association of pelvic failure with HIV infection. HIV infection is associated with anemia [33,34]. Chronic and transient hypoxia makes tumors radioresistant and is an adverse prognostic factor for both local control and survival outcome [35 – 37]. Therefore, HIV infection could be a proxy indicator of anemia. Indeed, HIV was significantly and independently associated with anemia in our study. However, even after controlling for anemia, HIV
409
infected patients still had more pelvic failure than uninfected patients. Another hypothesis is that HIV infection may alter cervical cancer cell kinetics resulting in more radioresistant tumor. HIV infected patients had more treatment interruptions than noninfected patients. Although we did not show an association between treatment interruptions and residual tumor in our study, the overall treatment time was less than 50 days. Significant loco-regional failure due to treatment prolongation is often seen if the treatment period is beyond 50 days [4,30]. Treatment duration of longer than 50 days in our study was associated with a 5-fold higher risk of residual tumor. It is possible that the treatment interruptions associated with HIV infection could have resulted in pelvic failure. Treatment prolongation contributes to loss of loco-regional control by allowing clonogenic repopulation [5]. It is uncertain at the moment whether HIV infection is associated with different cell kinetics or radioresistant tumors. Like in head and neck tumors, time to failure in cervical carcinoma is often within 2 years. This may be a reflection of rapid growth kinetics [5]. Maiman [38] reported that HIV infected cervical cancer patients died of complications due to cancer and not to HIV on average of 10 months despite appropriate treatment. Kim et al. [18] reported that median time to cancer-related death of anal cancer patients HIV-positive was 1.4 years as compared to 5.3 years for HIVnegative patients. This suggests that HPV – HIV-related disease is aggressive, which could be a manifestation of rapid growth kinetics although there is no laboratory data to support this hypothesis. However, in anal and cervical cancer, HIV infection is associated with microsatellite instability and loss of heterozygosity (LOH), which is probably a molecular explanation of the aggressiveness of HPV –HIV-related cancer [39,40]. Recurrent LOH at D2S123 has been demonstrated in HIV-positive patients with radioresistant anal squamous-cell carcinoma [39]. Certain limitations need to be considered in interpretation of data from this study. Although we had sufficient power, the absolute number of patients included is small, thus making the study less robust. These findings need to be replicated in more extensive studies. Documentation of the parameters reported is prone to observer variation though all the documentation for this study was done by one person (PG). As observed by Sismond et al. [41] regarding late morbidity, literature on the subject shows large variations in reported complications and definition of complications. The other limitation was the use of clinical documentation of pelvic failure, which is subjective. More objective measures such as pre- and posttreatment computerized pelvic tomogram (CT-scan) or magnetic nuclear resonance imaging (MRI) [42,43] could not be utilized due to cost limitations. The third limitation is that the patient lost to follow-up had well-differentiated tumors. This could have resulted in overestimation of pelvic failure since well-differentiated tumors have a better outcome than poorly differentiated tumors. Lastly, data from this study relate only to EBRT, and only the patients who were followed up, thus it may not be generalized to other populations receiving more than EBRT for cervical cancer.
410
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411
In conclusion, this study has shown that HIV infection is associated with more extensive irradiation-related acute toxicity, treatment interruptions and pelvic failure to control cervical cancer following EBRT. Our study and others on anal and cervical cancer suggest that HIV infection is an adverse prognostic factor. Our results need to be replicated in more rigorous extensive studies. Studies to determine survival outcome of ICC HIV-positive patients are needed. These studies will assist to design appropriate treatment strategies for invasive cervical cancer, HIV-positive patients, as these patients are likely to increase in numbers as HIV pandemic continues growing. Acknowledgments We thank VLIR (Flemish Interuniversity Council), Belgium for financial support, the Director of Kenyatta National hospital for allowing access to patients, staff in radiotherapy unit and laboratory staff in Department of Medical Microbiology, University of Nairobi. Special thanks go to Regina Kilonzo, and Jayne Mbithi, the study staff and S. Vansteelandt, from the Department of Applied Mathematics and Informatics, University of Ghent, Belgium, for their contribution to data analysis. References [1] Lanciano RM, Pajak TF, Martz K, Hanks GE. The influence of treatment time on outcome for squamous cell cancer of the uterine cervix treated with radiation: a patterns-of-care study. Int J Radiat Oncol Biol Phys 1993;25:391 – 7. [2] Perez CA, Grigsby PW, Camel HM, et al. Irradiation alone or combined with surgery in stage IB, IIA, and IIB carcinoma of uterine cervix: update of a nonrandomized comparison. Int J Radiat Oncol Biol Phys 1995;31:703 – 16. [3] Fyles A, Keane TJ, Barton M, Simm J. The effect of treatment duration in the local control of cervix cancer. Radiother Oncol 1992;25:273 – 9. [4] Girinsky T, Rey A, Roche B, Haie C, Gerbaulet A, Randrianarivello H, et al. Overall treatment in advanced cervical carcinomas: a critical parameter in treatment outcome. Int J Radiat Oncol Bio Phys 1993;27:1051 – 6. [5] Petereit DG, Sarkaria JN, Chappell R, Flowee JF, Hartmann TJ, Kinsella TJ, et al. The adverse effect of treatment prolongation in cervical carcinoma. Int J Radiat Oncol Biol Phys 1995;32:1301 – 7. [6] Erridge SC, Kerr GR, Downing D, Duncan W, Price A. The effect of overall treatment time on the survival and toxicity of the radical radiotherapy for cervical carcinoma. Radiother Oncol 2002;63:59 – 66. [7] Schultheiss TE, Lee WR, Hunt MA, et al. Late GI and GU complication in the treatment of prostate cancer. Int J Radiat Oncol Biol Phys 1997;37:3 – 11. [8] Wang CJ, Leung SW, Chen HC, et al. The correlation of acute toxicity and late rectal injury in radiotherapy for cervical carcinoma: evidence suggestive of consequential late effect. Int J Radiat Oncol Biol Phys 1998;40:85 – 91. [9] Peters LJ, Ang KK, Thames HD. Accelerated fractionations in the radiation treatment of head and neck cancer. Acta Oncol 1988;27:185 – 94. [10] Peters KJ, Ang KK. The role of altered fractionation in head and neck cancers. Semin Radiat Oncol 1992;2:180 – 94. [11] Buehler JW, Ward JW. A new definition for AIDS surveillance. Ann Int Med 1993;118:390 – 2. [12] Gemignani M, Maiman M, Fruchter RG, Arrastia CD, Gibbon D, Ellison T. CD4 lymphocytes in women with invasive and preinvasive cervical neoplasia. Gynecol Oncol 1995;59:364 – 9. [13] Frutcher RG, Maiman M, Sedlis A, Bartley L, Camilien L, Arrastia CD. Multiple recurrences of cervical intraepithelial neoplasia in women with the human immunodeficiency virus. Obstet Gynecol 1996;87:338 – 44.
[14] Maiman M, Fruchter RG, Guy L, et al. Human immunodeficiency virus infection and invasive carcinoma. Cancer 1993;71:402 – 6. [15] Matorras R, Ariceta JM, Rementeria A, et al. Human immunodeficiency virus-induced immunosuppression: a risk factor for human papillomavirus infection. Am J Obstet Gynecol 1991;164:42 – 4. [16] Wright TC, Koulos J, Schnoll F, et al. Cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus: outcome after loop electrosurgical excision. Gynecol Oncol 1994;55:253 – 8. [17] Hoffman R, Welton ML, Klencke B, Weinberg V, Krieg R. The significance of pretreatment CD4 count on the outcome and treatment tolerance of HIV-positive patients with anal cancer. Int J Radiat Oncol Biol Phys 1999;44:127 – 31. [18] Kim JH, Sarani B, Orkin BA, Young HA, White J, Tannebaum I, et al. HIV-positive patients with anal carcinoma have poorer treatment tolerance and outcome than HIV-negative patients. Dis Colon Rectum 2001;44: 1496 – 502. [19] Maiman M, Fruchter MG, Clark M, Arrastia CD, Mathews R, Gates E. Cervical cancer as an AIDS-defining illness. Obstet Gynecol 1997; 89:76 – 80. [20] Wilson GD, McNally NJ, Disches S, Saunders MI, Desrochers C, Lewis AA, et al. Measurement of cell kinetics in human tumors and flow cytometry. Br J Cancer 1988:423 – 31. [21] Cox JD, Stetz J, Pajak TF. Toxicity criteria of the radiation therapy oncology group (RTOG) and the European organization for research and treatment of cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31: 1341 – 6. [22] Sedlis A, Bundy BN, Rotman MZ, Lentz SS, Muderspach LI, Zaino RJ. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: a gynecology oncology group study. Gynecol Oncol 1999;73:177 – 83. [23] Gichangi P, De Vuyst H, Estambale B, Rogo K, Bwayo J, Temmerman M. HIV and cervical cancer in Kenya. Int J Gynecol Obstet 2002;76:55 – 63. [24] K.O. Rogo, Human Papillomavirus and Human Immunodeficiency virus infection in relation to cervical cancer: studies and observations of basic cliQ nical and epidemiological aspects of cancer of the cervix with special refeQ rences to Kenya, Africa. PhD thesis, University of Umea, Sweden; 1990. [25] Landoni F, Maneo A, Colombo A, et al. Randomized study of radical surgery versus radiotherapy for stage IB – IIB cervical cancer. Lancet 1997;350:535 – 40. [26] Roeske JC, Mundt AJ, Halpern H, et al. Late rectal sequelae following definitive radiation therapy for carcinoma of the uterine cervix: a dosimetric analysis. Int J Radiat Oncol Biol Phys 1997;37:351 – 8. [27] Sood BM, Gorla G, Gupta S, Garg M, Deore S, Runowicz CD, et al. Two fractions of high-dose-rate brachytherapy in the management of cervix cancer: clinical experience with and without chemotherapy. Int J Radiat Oncol Biol Phys 2002;53:702 – 6. [28] Langberg CW, Sauer T, Reitan JB, Hauer-Jensen M. Tolerance of rat small intestine to localized single dose and fractionated irradiation. Acta Oncol 1992;31:781 – 7. [29] Hauer-Jensen M, Poulakos L, Osborne JW. Intestinal complications following accelerated fractionated x-irradiation. An experimental study in the rat. Acta Oncol 1990;29:229 – 34. [30] Pedersen D, Bentzen SM, Overgaard J. Early and late radiotherapeutic morbidity in 442 consecutive patients with locally advanced carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1994;29:941 – 52. [31] Klevens RM, Fleming PL, Mays MA, et al. Characteristics of women with AIDS and invasive cervical cancer. Obstet Gynecol 1996;88:269 – 73. [32] Gichangi PB, Bwayo J, Estambale B, De Vuyst H, Ojwang S, Rogo K, et al. Impact of HIV infection on invasive cervical cancer in Kenyan women. AIDS 2003;17:1963 – 8. [33] Coyle TE. Hematologic complications of human immunodeficiency virus infection and the acquired immunodeficiency syndrome. Med Clin North Am 1997;81:449 – 70. [34] Sullivan PS, Hanson DL, Chu SY, Jones JL, Ward JW. Epidemiology of anemia in human immunodeficiency virus (HIV)-infected persons: results from the multistate adult and adolescent spectrum of HIV disease surveillance project. Blood 1998;91:301 – 8.
P. Gichangi et al. / Gynecologic Oncology 100 (2006) 405 – 411 [35] Fyles M, Milosevic D, Hedley M, Pintilie W, Levin L, Manchul L, et al. Tumor hypoxia has independent predictor impact only in patients with node-negative cervix cancer. J Clin Oncol 2002;20:680 – 7. [36] Fyles AW, Milosevic M, Wong R, et al. Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 1998;48:149 – 56. [37] Knoncke TH, Weitmann HD, Feldmann HJ, et al. Intratumoral pO2 measurements as predictive assay in the treatment of carcinoma of the uterine cervix. Radiother Oncol 1999;53:99 – 104. [38] Maiman M. Management of cervical neoplasia in Human Immunodeficiency Virus-infected women. Monogr Natl Cancer Inst 1998;23:43 – 9. [39] Gervaz P, Efron J, Poza AF, et al. Loss of heterozygosity and HIV infection in patients with anal squamous-cell carcinoma. Dis Colon Rectum 2001;44:1503 – 5.
411
[40] Wistuba II, Syed S, Behrens C, Duong M, Milchgrub S, Muller CY, et al. Comparison of molecular changes in cervical intraepithelial neoplasia in HIV-positive and HIV-indeterminate subjects. Gynecol Oncol 1999;74: 519 – 26. [41] Sismondi P, Sinistrero G, Zola P, Volpe T, Ferraris R, Castelli GL, et al. Complications of uterine cervix carcinoma treatment: the problem of a uniform classification. Radiother Oncol 1989;14:9 – 17. [42] Mayr NA, Taoka T, Yuh WT, et al. Method and timing of tumor volume measurement for outcome prediction in cervical cancer using magnetic resonance imaging. Int J Radiat Oncol Biol Phys 2002;52:14 – 22. [43] Mayr NA, Yuh WTC, Zheng J, Ehrhardt JC, Sorosky JI, Magnotta VA, et al. Tumor size evaluated by pelvic examination compared with 3-D MR quantitative analysis in the prediction of outcome for cervical cancer. Int J Radiat Oncol Biol Phys 1997;39:395 – 404.