Hakimian, Roger MD; Fang, Hongbin PhD; Thomas, Leno MD; Edelman, Martin J. MD
Human immunodeficiency virus infection (HIV) and primary lung cancer are two major contemporary epidemics. It is estimated that in the year 2007, 213,380 patients will be diagnosed with primary lung cancer and 160,390 will die of the disease.1 Many carcinogens contribute to the development of lung cancer, but cigarette smoking is overwhelmingly the most important risk factor.
HIV infection and the acquired immunodeficiency syndrome (AIDS) increase the risk of opportunistic infections and also non-Hodgkin’s lymphoma and Kaposi sarcoma. These malignancies were later joined by invasive cervical cancer as AIDS-defining malignancies.2 Other cancers that occur excessively in HIV/AIDS patients include lung cancer, Hodgkin’s disease, soft-tissue malignancies, penile cancer, lip cancer, and testicular seminoma. Frish et al.3 reviewed cancer and HIV registries of 11 geographic areas of the United States and found that lung cancer is the most frequently diagnosed non-AIDS defining cancer (from 302,834 patients with AIDS, 808 were diagnosed with lung cancer between the year 1978 and 1996). Frish et al.3 also note that lung cancer relative risk increases with the progression of the disease from HIV status to the AIDS disease, which meets the criteria of association with immunosuppression. The standard risk ratio of lung cancer in HIV and AIDS patients diagnosed between 1990 and 1995, compared with the risk among the U.S. population, was 6.5.4
Most case reports and series of HIV/AIDS patients with lung cancer reported in the literature were published before the era of highly active antiretroviral therapy (HAART). This multidrug approach has revolutionized the care of AIDS patients since it entered general clinical practice in 1996.5
In 1984, Irwin et al.6 reported the first case of HIV and lung cancer. In 1989, the Italian cooperative group on AIDS-related tumor reported eight cases of lung cancer. The median age was 34 years; three patients could not be treated, and four patients died of progressive disease while receiving treatment. Five of the eight patients had a limited disease at diagnosis.7 Since then, additional series have been published8–13 (Table 1). Though most of the patients were relatively young at lung cancer diagnosis, few received treatments that were generally considered effective for lung cancer, and the prognosis was dismal (reported median survival was between 1 and 5 months). The largest series was from the Italian cooperative group on AIDS and tumor; it included 36 patients diagnosed with HIV and lung cancer between the years 1986 and 1998. The median age was 38 years compared with 53 years for the control group (102 HIV-negative patients with age <60 years). Three patients had only received HAART therapy. Thirty-one patients had non-small cell lung cancer (NSCLC) and five patients had small cell lung cancer (SCLC). The median survival of the total group was 5 months versus 10 months for the control group (p = 0.0001).13
One review of HIV patients with lung cancer diagnosed after the HAART era was presented by Bower et al.14; from 8640 HIV-seropositive patients diagnosed since 1986, 11 patients had lung cancers (10 NSCLC and one SCLC). Nine of those patients developed lung cancer in the post-HAART era (1997–2001), and six patients received HAART therapy before developing lung cancer (median, 2 years). All patients had stage IIIB or stage IV disease. The overall survival was 4 months. The author noted that the incidence of HIV-related lung cancer increased from 0.8 per 100,000 patient-years in the pre-HAART era, which is similar to the age- and sex-matched rate for the general population, to 6.7 per 100,000 patient-years in the post-HAART era.15
We retrospectively reviewed lung cancer patients with an onset of disease after 1996 who had concomitant infection with HIV, which might have been diagnosed either before or simultaneously with lung cancer.
MATERIALS AND METHODS
The tumor registry at University of Maryland Medical System was searched for patients with a diagnosis of primary lung cancer who also had been diagnosed with HIV/AIDS between 1996 and January 2003. We also searched the medical records at an affiliate institution, Baltimore Veterans Affairs Medical Center, for the years 2000 to 2003. Paper and electronic charts were reviewed with special attention to demographic data, HIV risk factors, smoking habits, comorbidities, dates of diagnosis for HIV and lung cancer, CD4 counts, cancer histology, performance status (PS), staging, treatment modalities, and survival. Viral load data were not available for much of this period and were not captured. For patients who were lost to follow-up, dates of death were collected by using the Social Security Web site and/or calling the next of kin. Survival was calculated from the date of diagnosis of lung carcinoma according to the product limit method of Kaplan and Meier.16 Survival information of patients who were lost to follow-up was censored. This study was approved by the institutional review board of the University of Maryland.
From January 1, 1996 to December 31, 2003, 2042 patients were seen with lung cancer at University of Maryland; 29 patients (1.4%) had the diagnosis of HIV/AIDS infection. From January 1, 2000 to December 31, 2003, 387 patients with lung cancer were seen at the Veterans Affairs Medical Center in Baltimore; five (1.3%) had HIV/AIDS. Patients’ demographic characteristics are described in Table 2.
Sixty-eight percent of HIV-positive patients with lung cancer were male. All patients were African Americans except for one Caucasian. In comparison, 63% and 30% of all lung cancer patients were male and African American, respectively. For patients in the registry with available information, 86% (19 of 22 patients) had previous history of intravenous drug abuse. All patients were smokers, with an average of 25 packs of cigarettes per year (range, 6–50 pack-years).
Thirty patients (88%) were diagnosed with NSCLC, and four patients (12%) had SCLC. All of the patients with SCLC had extensive-stage disease. In the NSCLC group, three patients had stage I to stage IIIA disease, and 27 patients had stage IIIB or stage IV disease. Of the total group, 18 patients (53%) had a PS of 0 or 1 on the Zubrod scale,17 11 patients (32%) had a PS of 2, and five patients (15%) had a PS of at least 3.
Twenty-seven patients were diagnosed with HIV or AIDS at least 6 months before the lung cancer diagnosis was made. For seven patients (20%), the diagnosis of HIV infection was made concurrently with the diagnosis of lung cancer. Of the 91 African American patients younger the 60 years diagnosed at the University of Maryland with advanced-stage NSCLC during the same period, 21 (23%) were HIV positive. The characteristics of the patients’ HIV infection and therapy are described in Table 3.
Treatment and Outcome
The treatment and outcome of these patients is summarized in Table 4. Eighteen patients (60%) were treated with HAART at some point before lung cancer diagnosis. Duration of therapy could not be assessed for all of the patients because some had received care at multiple institutions. All SCLC patients received chemotherapy composed of cisplatin and etoposide. Median survival was 7.2 months (5–11 months).
In the NSCLC group with stage IIIB or stage IV disease, seven patients were too ill to be treated (median survival was 2 months). Fourteen patients received chemotherapy (51%); 10 of these patients tolerated a two-drug regimen that included a platinum agent. Five patients were able to tolerate two or more lines of chemotherapy. Three patients were treated with concurrent chemotherapy, and four patients received a palliative dose of radiation therapy to a metastatic site. The median survival of patients with stage IIIB or stage IV NSCLC was 5.2 months (range, 1–16). For this group, patients with absolute CD4 of at least 200 at diagnosis (n = 15) had a median survival of 11.5 months compared with 3.4 months for patients with CD4 counts below 200 (p = 0.12), Figure 1. Similarly, there was a numeric superiority for HAART versus no HAART (6.7 versus 2.8 months) in terms of median survival, although this did not achieve statistical significance (p = 0.55), Figure 2. For the entire group, length of time from HIV diagnosis to lung cancer diagnosis had a significant (p < 0.001) positive influence on survival.
This is one of the largest single-institution experiences with HIV-positive lung cancer patients in the post-HAART era. The number of HIV patients diagnosed with lung cancer seems to be increasing. From the tumor registry at the University of Maryland, we found only four lung cancer patients with a diagnosis of HIV in the period of 1990 to 1996, and 29 patients in the period of 1996 to 2003 (1.4% of lung cancer patients). This increase in the incidence is probably not related to changes in the incidence of HIV or lung cancer but, rather, to the new therapeutic advances for HIV. The age-adjusted death rate from AIDS declined by 48% from 1996 to 1997.18 The natural history of HIV infection has changed because of improvements to immune function and prolonged survival.19,20 This decline in HIV mortality is related to many factors other than a possible decrease in the disease incidence, including improved prophylaxis for opportunistic infections and more aggressive viral therapy, which is represented by the application of HAART in clinical practice.
There are several interesting epidemiologic factors in the current report. The young age at lung cancer diagnoses (44 years) is similar to that observed by others and is consistent with the age affected by HIV. It may also be related to an early starting age of smoking and chronic dysfunction of the immune system.3 Another important observation is the high incidence of intravenous drug abuse as a risk factor for HIV (86%). Serraino et al.21 have previously described a 6.2-fold higher incidence of lung cancer in HIV patients with history of intravenous drug abuse compared with patients with history of homosexual activity. It is likely that those who are intravenous drug abusers are also tobacco users. A recent epidemiologic report from another institution serving the same metropolitan area as the University of Maryland found that there was a very high rate of smoking (97%) in the HIV-positive population. They did not observe the strong preponderance of African American males seen in our study.22
The median survival (8 months) for all patients seems to represent an improvement compared with the survival of patients in the pre-HAART era (median survival, 1–5 months). The low incidence of cases at our institution before 1996 precluded a survival comparison based on pre– versus post–HAART-era cases. In advanced stages of NSCLC (stages IIIB and IV), patients with higher CD4 counts tended to have a better survival then those with lower counts (11.4 versus 9 months, respectively). The length of time from HIV infection to lung cancer diagnosis had a positive influence on prognosis; this observation is related to the fact that patient who had both diseases diagnosed in a shorter interval of time is the one who had more advanced diseases that were ignored by the patient because of lifestyle or health care access difficulties.
There are several possible mechanisms by which HAART therapy could improve survival. Better immune function could potentially result in a less aggressive disease. Clearly, fewer opportunistic infections, preservation of performance, and nutritional status allowed patients to tolerate more active therapy than in the past. In addition, advances made in supportive care, opportunistic infection prophylaxis, and the use of new-generation chemotherapeutic agents have played a major role in the management of this disease since 1996.
Patients with stage III/IV NSCLC were analyzed as a group because this constituted a fairly homogenous cohort and because the approach and outcomes of these patients with standard treatment have been well described. The median survival of 5.2 months for patients with advanced-stage disease is comparable with that reported from our institution for patients (n = 20, MST = 5.8 months) with a PS of 2 or 3 treated on a recent trial and is superior to the outcomes reported for untreated patients in prior series.23 Nevertheless, this outcome is clearly inferior to the results from standard two-drug platinum-based chemotherapy at our institution in patients (n = 33, MST = 9.3 months) with a PS of 0 or 1.24
The absence of a statistically significant difference in survival detected between those patients with preserved CD4 counts (>200) and those with CD4 <200, and also between HAART-treated and non–HAART-treated patients, is likely to be the result of the small number of patients evaluated. In addition, the number of patients was too small to perform a multivariate analysis.
The high proportion of African American patients deserves comment. The University of Maryland and the Baltimore Veterans Affairs Medical Center are located in Baltimore, MD, a city with a predominantly African American population. Nevertheless, the tumor registry at the University of Maryland indicates that approximately 30% of all lung cancer patients are African American. Therefore, it does not seem that this demographic fact alone is sufficient to account for the high prevalence of lung cancer among African Americans. One possible explanation is that the HIV-positive population in Baltimore is cared for by the University of Maryland infectious disease service and that referral patterns could account for this finding.
Lung cancer patients with HIV infection have been encountered more frequently since the advancement of HAART, especially in the minority population.
HIV infection needs to be ruled out in every young patient with lung cancer, even in the presence of obvious risk factors such as smoking.
The major finding of this retrospective series is that patients with advanced-stage NSCLC have a survival time that approaches that of the HIV-negative lung cancer patient. Though not achieving statistical significance, it seems that those patients with preserved CD4 counts and/or on HAART have improved outcomes compared with those with more compromised immune status. Therefore, those individuals presenting with good performance status, an absence of severe opportunistic infection, and preserved nutritional status should be considered for active therapy of lung cancer.
1. Jemal A, Siegel B, Ward E, Muriay T, Xu J, Thun MJ. Cancer statistics, 2005. CA Cancer J Clin 2007;57:43–66.
2. AIDS definition update by CDC. Am Fam Physician 1993;47:265.
3. Frish M, Biggar RJ, Engels EA, et al. Association of cancer with AIDS-related immunosuppression in adults. JAMA 2001;285:1736–1745.
4. Parker MS, Leveno DM, Campbell TJ, Worrell JA, Carozza SE. AIDS-related bronchogenic carcinoma: fact or fiction? Chest 1998;113:154–161.
5. Flepp M, Schiffer V, Weber R, Hirschel B. Modern anti-HIV therapy. Swiss Med Wkly 2001;13:207–213.
6. Irwin LE, Begandy MK, Moore TM. Adenosquamous carcinoma of the lung and the acquired immunodeficiency syndrome. Ann Intern Med 1984;100:158.
7. Monfardini S, Vaccher E, Pizzocaro G, et al. Unusual malignant tumours in 49 patients with HIV infection. AIDS 1989;3:449–452.
8. Moser RJ III, Tenholder MF, Ridenour R. Oat-cell carcinoma in transfusion-associated acquired immunodeficiency syndrome. Ann Intern Med 1985;103:478.
9. Nusbaum NJ. Metastatic small-cell carcinom of the lung in a patient with AIDS. N Engl J Med 1985;312:1706.
10. Karp J, Profeta G, Marantz PR, Karpel JP. Lung cancer in patients with immunodeficiency syndrome. Chest 1993;103:410–413.
11. Flores MR, Sridhar KS, Thurer RJ, Saldana M, Raub WA Jr, Klimas NG. Lung cancer in patients with human immunodeficiency virus infection. Am J Clin Oncol 1995;18:59–66.
12. Alshafie MT, Donaldson B, Oluwole SF. Human immunodeficiency virus and lung cancer. Br J Surg 1997;84:1068–1071.
13. Tirelli U, Spina M, Sandri S, et al. Lung carcinoma in 36 patients with human immunodeficiency virus infection. The Italian Cooperative Group on AIDS and Tumors. Cancer 2000;88:563–569.
14. Braun MA, Killam DA, Remick SC, Ruckdeschel JC. Lung cancer in patients seropositive for human immunodeficiency virus. Radiology 1990;175:341–343.
15. Bower M Powles T, Nelson M, et al. HIV-related lung cancer in the era of highly active antiretroviral therapy. AIDS 2003;17:371–375.
16. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–481.
17. Zubrod CG, Schneiderman M, Frei IIIE et al. Appraisal of methods for the study of chemotherapy of cancer in man: comparative therapeutic trial of nitrogen mustard and triethylene thiophosphoramide. J Chron Dis 1960;11:7–33.
18. Hoyert DL, Kochanek KD, Murphy SL. Deaths: final data for 1997. Natl Vital Stat Rep 1999;47:1–104.
19. Navarro JT, Ribera JM, Oriol A, et al. Influence of highly active anti-retroviral therapy on response to treatment and survival in patients with acquired immunodeficiency syndrome-related non Hodgkin’s lymphoma treated with cyclophosphamide, hydroxydoxorubicin, vincristine and presnisone. Br J Haematol 2001;112:909–915.
20. Palella FJ Jr, Delaney KM, Moorman AC et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998;338:853–860.
21. Serraino D, Boschini A, Carrieri P, et al. Cancer risk among men with, or at risk of, HIV infection in southern Europe. AIDS 2000;14:553–559.
22. Engels EA, Brock MV, Chen J, Hooker CM, Gillison M, Moore RD. Elevated incidence of lung cancer among HIV-infected individuals. J Clin Oncol 2006;24:1383–1388.
23. Thomas L, Edelman MJ, Guo C, Doyle LA, Hausner P, Smith R. Phase II trial of gemcitabine/carboplatin followed by paclitaxel in patients with poor performance status PS = 2, 3 in advanced non-small cell lung cancer. Proc ASCO 2003:22.
24. Edelman MJ, Smith R, Kendall J, et al. Phase II trial of the novel retinoid, bexarotene, and gemcitabine/carboplatin in advanced non-small cell lung cancer. J Clin Oncol 2005;23:5774–5778.