SURVIVAL IN PATIENTS WITH NON-SMALL CELL LUNG CANCER …journal.usm.my/journal/MJMS-13-2-024.pdf · Key words: non-small cell lung cancer, survival, cancer-specific treatment, Malaysia

24

ORIGINAL ARTICLE

SURVIVAL IN PATIENTS WITH NON-SMALL CELL LUNG CANCERWHO OPTED OUT OF CANCER-SPECIFIC THERAPY

Li-Cher Loh, Ru-Yu Tan, Li-Yen Chan, Selvaratnam Govindaraju*, Kananathan Ratnavelu*, ShaliniKumar**, Sree Raman***, Pillai Vijayasingham***, Tamizi Thayaparan***

IMU Lung Research, International Medical University, Clinical School, Seremban; *Nilai CancerInstitute, Nilai; **Department of Pathology, Seremban Hospital; ***Department of Medicine, Seremban

Hospital, Seremban, Malaysia

In Malaysia, many patients opted out of cancer-specific treatment for variousreasons. This study was undertaken to investigate the survival rate of patientswith stages I to III non-small cell lung cancer (NSCLC) who opted out of treatment,compared with those who accepted treatment. Case records of 119 patientsdiagnosed with NSCLC between 1996 and 2003 in two urban-based hospitals wereretrospectively examined. Survival status was ascertained from follow-up medicalclinic records or telephone contact with patients or their next-of-kin. Median (25-75% IQR) survival rate for 79 patients who accepted and 22 patients who optedout of treatment, were 8.6 (16.0-3.7) and 2.2 (3.5-0.8) months respectively [log rankp< 0.001, Kaplan-Meier survival analysis]. Except for proportionately more patientswith large cell carcinoma who declined treatment, there was no significantdifference between the two groups in relation with age, gender, ethnicity, tumourstage, and time delays between symptom onset and treatment or decision-to-treat.We concluded that there was a small but significant survival benefit in acceptingcancer-specific treatment. The findings imply that there is no effective alternativetherapy to cancer-specific treatment in improving survival. However, overallprognosis for patients with NSCLC remains dismal.

Key words : non-small cell lung cancer, survival, cancer-specific treatment, Malaysia

Introduction

The prognosis of lung cancer remains poor,with overall five-year survival figures varyingbetween 5 and 10% worldwide (1). Unlike othersolid tumours where survival rates have improvedin the order of 60-90% with modern treatment, themanagement and prognosis of lung cancer haschanged very little over the past 20 years (2).Malaysia produced its long-overdue first nationalcancer registry in 2003 and in its report, lung canceris the commonest cancer in males and the fifthcommonest in females (3).

In view of this, there is merit in consideringwhat may influence and be responsible for the poorprognosis of lung cancer in Malaysia. Doctors herefrequently encounter patients who have decided toopt out of cancer-specific treatment, be it surgery,

chemotherapy or radiotherapy, for various reasons.It is also recognized that many patients turn toalternative treatment, usually in the form oftraditional medicine, although there is little data ontheir efficacy or survival benefit.

In order to investigate the impact on survivalin patients who specifically opted out of cancer-specific treatment, we compared the survival ratesbetween patients with Stage 1 to III NSCLC whoaccepted and those who opted out of cancer-specifictreatment in two urban-based hospitals (a stategovernment general hospital and a private oncologyhospital) in Malaysia. We also attempted to identifyany patient clinico-demographic variables, includingthe time interval between onset of symptoms andtreatment or decision-to-treat, which could possiblyinfluence survival rates.

Submitted-10.10.2004, Accepted-24.11.2005

Malaysian Journal of Medical Sciences, Vol. 13, No. 2, July 2006 (24-29)

25

Patients & Methods

Data collectionUsing a structured data collection form,

relevant information was retrospectively collectedfrom the medical records of patients with confirmedNSCLC between 1 January 1996 and 1 April 2004in Seremban General Hospital and Nilai CancerHospital. Patients, with their NSCLC histologytypes, were first identified from the pathologydatabase of the Department of Pathology, SerembanHospital and the Cancer Register, Nilai CancerInstitute. Their medical records were then retrievedfor perusal. Data on tumour stage, date of onset offirst symptoms as stated by the patient, date of firsthospital consultation, date of diagnosis, date of

treatment (or decision-to-treat if date of treatmentwas not available or not applicable), the type oftreatment offered, and whether treatment wasaccepted or declined, were obtained from thesepatients’ medical records. Patients whose recordsindicated that lung involvement was metastasis, andpatients who opted out of treatment but died withintwo weeks from the date treatment was offered, wereexcluded from the study. The latter was intended toexclude patients who were probably better classifiedas Stage IV disease because of the rapid progressionof disease. The protocol of the study was approvedby the local university Research & Ethics Committeeand carried out in accordance to therecommendations of the Helsinki Declaration of1975.

SURVIVAL IN PATIENTS WITH NON-SMALL CELL LUNG CANCER OPTED OUT OF CANCER-SPECIFIC THERAPY

Table 1 : Patients (n=119) who accepted and opted out of cancer-specific treatment in relationto their demographic characteristics, tumour stage and treatment received

Patients with KNUCKLES recommended for cancer-specific therapy

Variables

All patients, nSourceSeremban HospitalNilai Cancer InstituteGenderMaleFemaleAgeMean yrs (95% CI)< 55 yrs55-70 yrs> 70 yrsEthnicityMalayChineseIndianCigarette smokerCurrent or pastNeverUnknownTumour stageStage I and IIStage IIITumour typeAdenocarcinomaSquamous cellLarge cellTreatment receivedSurgeryChemotherapy ±radiotherapyNeoadjuvant therapy

119

39.5 (47)60.5 (72)

65.5 (78)34.5 (41)

61 (58-63)26.9 (32)48.7 (58)24.4 (29)

20.2 (24)67.2 (80)12.6 (15)

58.0 (69)32.8 (39) 9.2 (11)

6.7 (8) 93.3 (111)

29.4 (35)46.2 (55)6.7 (8)

2.5 (3)75.6 (90)

3.4 (4)

97

30.9 (30)69.1 (67)

63.9 (62)36.1 (35)

60 (57-62)29.9 (29)48.5 (47)21.6 (21)

19.6 (19)67.0 (65)13.4 (13)

55.7 (54)35.1 (34) 9.3 (9)

7.2 (7) 92.8 (90)

30.9 (30)50.5 (49)4.1 (4)

3.1 (3)92.8 (90)

4.1 (4)

22

77.3 (17)22.7 (5)

72.7 (16)27.3 (6)

65 (60-70)13.6 (3)50.0 (11)36.4 (8)

22.7 (5)68.2 (15) 9.1 (9.1)

68.2 (15)22.7 (5)

9.1 (9.1)

4.5 (1) 95.5 (21)

22.7 (5)27.3 (6)18.2 (4)

00

0

-

-<0.001

-0.432

0.069--

0.186

--

0.837

--

0.518

-0.651

--

0.030

--

-

Whole group Accepted Opted out p1

26

Tumour stageFor the purpose of data collection, tumour

stage was categorized into those whose cancers wereamenable to resection, those with unresectable,locally advanced cancers that were amenable to non-surgical cancer-specific therapy i.e. chemotherapy,radiotherapy or both, with the intention ofprolonging life and those with terminal disease,where only palliative treatment was recommended.Broadly, they were consistent with Tumour NodeMetastasis (TNM) Stage I to II, III and IVrespectively. The reason for this approach wasbecause many records in Seremban Hospital did notstate the TNM classification. Only patients withStage 1 to III disease were accepted into the study.

Survival statusSurvival status was ascertained from records

of follow-up visits in medical outpatient clinics, andif necessary, by direct contact with patients or next-of-kin by telephone. Patients whose survival statuscould not be confirmed were excluded from thestudy.

Accepting vs. opting out of treatmentThis information was based on medical

records and if necessary, verified by telephonecontact with patients or their next-of-kin. Patientswhere this information could not be verified wereexcluded from the study. According to our studyprotocol, no attempt was made to elucidate thereason for opting out of treatment. This was due tothe lack of such information from medical recordsand the possible sensitive nature of direct enquiry.

Data analysisDescriptive analyses were used to

characterize all patients. Differences between thosewho accepted and those who opted out of treatmentwere tested using Chi Square or unpaired t tests.The cumulative and median survival rate wasmeasured using Kaplan-Meier survival analysis withlog-rank test for the detection of difference insurvival between the two groups of patients. Thetime at risk was accumulated from the date oftreatment or decision-to-treat, until death or, in thosealive, until 1 April 2004. Median delay (with 25%to 75% interquartiles, IQR) was calculated for eachgroup, and Mann-Whitney test was used for pair-

Li-Cher Loh, Ru-Yu Tan et. al

Figure 1 : Kaplan-Meier survival curve in patients who accepted and opted out fromcancer-specific treatment. Median (25-75% IQR) survival rates for patientswho accepted and opted out of cancer-specific treatment were 8.6 (16.0-3.7) and 2.2 (3.5-0.8) months respectively.

27

way comparisons of delay. All computation wasmade using the statistical package SPSS version 11.5for Windows (Chicago, Illinois, USA). In all cases,the significance was defined at the 5% level and two-tailed.

Results

Of the 142 patients identified with thediagnosis of Stage 1 to III NSCLC, 21 patients wereexcluded due to incompleteness of data or missingmedical records. Two were excluded due to deathwithin two weeks from the date of treatment offer.The remaining 119 (83.8%) patients constituted thefinal analyzable sample for this study.

Of these patients, 22 (18.4%) patients optedout of cancer-specific treatment (1 recommended forsurgery; 21 for chemotherapy +/- radiotherapy).Most of them were from Seremban Hospital(77.3%), male (72.7%), of Chinese origin (68.2%),and were either current or past cigarette smokers(68.2%). Half of them were between 55 and 70 yearsof age. The majority of the tumour histological typewas adenocarcinoma (46.2%). Significantly greaterproportion of patients in Seremban Hospital optedout of treatment, compared with those in Nilai

Cancer Institute (p<0.001). The histological typesbetween the two groups were also significantlydifferent in that there were proportionately morepatients with large cell carcinoma in the group thatopted out of treatment. Otherwise, there were nosignificant differences in the patient demographiccharacteristics and tumour stage between the twogroups (Table 1). Mean age of patients who optedout of treatment was higher than those who accepted(65 vs. 60 yrs). This was, however, not statisticallysignificant (p=0.069).

Median (25-75% IQR) survival rates forpatients who accepted and those who opted out ofcancer-specific treatment were 8.6 (16.0-3.7) and2.2 (3.5-0.8) months, respectively. The differencewas statistically significant (log rank test p<0.001)(Figure 1). In patients who opted out of cancer-specific treatment, the median (25- 75% IQR) timeinterval between onset of first symptoms and firsthospital consultation, between first hospitalconsultation and confirmation of diagnosis, andbetween diagnosis and treatment or decision-to-treat,were 3 (1-6) months, 15 (6-25) days, 15 (8-30) days,respectively (Table 2). These time intervals were notsignificantly different from those who acceptedcancer-specific treatment.

SURVIVAL IN PATIENTS WITH NON-SMALL CELL LUNG CANCER OPTED OUT OF CANCER-SPECIFIC THERAPY

Table 2 : Patients (n=119) who accepted and opted out of cancer-specific treatment in relation totime interval from onset of symptoms until treatment or decision-to-treat

VariablesInlerval between symptomonset and first hospitalconsultationMedian (25-75% IQR) months< 1 month1 to 3 months> 3 months

Interval between first hospitalconsultation and confirmationof diagnosisMedian (25-75% IQR) days< 30 days≥ 30 days

Interval between diagnosis and treatment or decision-to-trearMedian (25-75% IQR) days< 30 days≥ 30 days

Whole group Accepted

Patients with NSCLC recommended forcancer-specific therapy

Opter out p1

2 (1-5)18.5 (20)52.9 (56)29.6 (32)

14 (6-38)72.3 (86)27.7 (33)

15 (6-30)72.3 (86)27.7 (33)

2 (1-4)20.7 (18)52.9 (46)26.4 (23)

12 (6-42)69.1 (67)30.9 (30)

12 (2-29)71.1 (69)28.9 (28)

3 (1-6)9.5 (2)

47.6 (10)42.9 (9)

15 (6-25)86.4 (19)13.6 (3)

15 (8-30)77.3 (17)22.7 (5)

0.082--

0.251

0.945-

0.102

0.324-

0.561

28

Discussion

We have shown that there was a clear survivalbenefit for patients who accepted cancer-specifictreatment. Except for tumour histology, we couldnot identify any association with the patients’demographic characteristics, tumour stage, and timeintervals from onset of symptoms to treatment ordecision-to-treat, between patients who accepted andwho opted out of treatment.

Although our study showed what is alreadyintuitively known, i.e. cancer-specific treatmentconveys survival benefit, albeit small especiallywhen complete resection of tumour is not possible(1), the purpose of our study was also to examinethe category of patients who specifically opted outof modern-day cancer-specific treatment in theMalaysian setting. As far as possible, thisinformation was verified during follow-up visits orduring telephone contact with patients or their next-of-kin. Unfortunately, the retrospective nature of thestudy and the possible sensitive aspect of directenquiries prevented our study from probing into thereasons for the decline.

It is widely known that patients declinecancer-specific treatment for various reasons.Studies addressing reasons for this are sparse,perhaps due to the complexity in studies of suchnature. A study in Russia, published in 1980 byEfimov et al (4), looked at the reasons in 180 patientswho refused surgical treatment for lung cancer. Theyshowed that 45% of patients declined treatment forfear of surgery, or due to unawareness of its potentialbenefit, or opted for home remedies instead. Another35% did not believe in the cure offered by surgery.It is very possible that these same reasons are stillrelevant today in Malaysia and worldwide.

The psychosocial impact of lung cancer onpatients and their families is well recognized andhas been extensively studied (5 - 7). Patients withlung cancer experience stigmatization, blame andshame (5), and frequently manifest psychiatricsymptoms such as insomnia, poor concentration anddisinterest (6). Family members including spousesalso play an important role, and it has been shownthat they tend to view patients’ functioning morenegatively than the patients themselves (7). Thesesocio-psychological factors play an important rolein the patients’ decision making of whether to acceptor decline treatment.

Seeking home remedies (4) or traditionaltreatment as a cause for opting out of treatment maybe of particular relevance in Malaysia. Doctors often

encounter patients who prefer alternative treatment,usually in the form of traditional medicine. Onereason for this may be the fear of the toxicity ofchemotherapy or radiotherapy, as experienced byother cancer patients. Another reason may also becultural, reflecting a highly established tradition ofmedical pluralism in Malaysia where doctors,sinsehs and bomohs are readily available (8 - 10).Patients move freely between the modern andtraditional medicinal system, or use both systemssimultaneously (9). Confidence in traditionalmedicine has resulted in patients of self-dischargingthemselves from hospital against medical advice (11)or not complying to treatment (12). While there isemerging evidence that many traditional herbalmedicines contain anti-cancer properties (13, 14),our data implies that used alone (15), there is noeffective alternative therapy when compared tomodern-day cancer-specific treatment.

We do not think that financial considerationwas an important cause for declining treatment inour study, since the majority of patients who optedout of treatment were from Seremban Hospital wheretreatment is available at low cost due to governmentsubsidy. Finally, it is possible that religious sentimentmight play a role in the patients’ decision as mostreligions here consider death as the will of God.

With the exception of tumour histology, wedid not identify any factors that differentiatedbetween those who accepted and those who declinedtreatment. The reason for proportionately morepatients with large cell carcinoma opting out oftreatment is unclear. The small number of patientswith large cell carcinoma in both groups (4 vs. 4)suggests that the statistical significance found islikely to be coincidental. The trend towards moreolder patients being in the group that opted out oftreatment suggests the possibility of age affectingthe decision whether to accept specific treatment.

Being retrospective in nature, our study seeksto reduce bias by excluding nearly 20% of the initialpatient sample, in whom information wereincomplete. Nevertheless, inaccuracies of doctors’records and patients’ recall could still introduce biasin the study. Nevertheless, we have no reason tobelieve that any misclassification in this respectwould significantly affect our findings.

While there have been several studies on lungcancer in Malaysia (16 - 18), none, except for oneon surgically treated patients (19) has looked intothe question of survival. Our study provides survivaldata in patients with Stage I to III NSCLC, with andwithout cancer-specific treatment, in the Malaysian

Li-Cher Loh, Ru-Yu Tan et. al

29

setting. This is timely in view of the recently releasedNational Cancer Registry (3). The overall prognosisof NSCLC remains dismal, with the median survivalrate for patients (primarily Stage III) amenable tocancer-specific treatment (primarily non-surgery) is8.6 months. Our findings show that there is a smallsurvival benefit of several months in persuadingpatients to accept cancer-specific treatment. It isunclear however whether this longer survival isassociated with improved quality of life. Moreresearch is required to address this, and also to studythe reasons for patients to opt out of modern-daycancer-specific treatment. It is possible that manyof the reasons for declining treatment areunjustifiable and irrational, and appropriateintervention, perhaps in terms of support andcounseling, can prevent this (20-21).

Acknowledgements

The authors wish to thank the support ofHospital Directors, Seremban Hospital and NilaiCancer Institute for this research, and the permissionof the Director-General, Ministry of Health Malaysiato publish the data from Seremban Hospital. Theresearch is supported by an internal research grantfrom the International Medical University.

Correspondence :

Dr Li-Cher Loh MBBCh (Ireland), MRCP (UK),MD (London)Department of Medicine,Clinical School, International Medical University,Jalan Rasah, Seremban 70300, Negeri Sembilan,MalaysiaTel: (+606) 767 7798 Fax: (+606) 767 7709E mail: [email protected]

References

1. Janssen-Heijnen ML, Gatta G, Forman D, CapocacciaR, Coebergh JW. Variation in survival of patients withlung cancer in Europe, 1985-1989. Eur J Cancer 1998;34 : 2191-196.

2. Northern and Yorkshire Cancer Registry andInformation Service (NYCRIS). A report on incidenceand management for the main sites of cancer 1999;Leeds, NYCRIS 2002: 1-56.

3. Lim GCC, Yahaya H, Lim TO. The first report of thenational cancer registry cancer incidence in Malaysia2002. National Cancer Registry, Ministry of HealthMalaysia 2002.

4. Efimov GA, Krasnov YuO, Chigirinskii LM. [Survivalof lung cancer patients who refused surgical treatment].Vopr Onkol 1980; 26: 79-84.

5. Chapple A, Ziebland S, McPherson A. Stigma, shame,and blame experienced by patients with lung cancer:qualitative study. BMJ 2004; 328:1470.

6. Ginsburg ML, Quirt C, Ginsburg AD, MacKillop WJ.Psychiatric illness and psychosocial concerns ofpatients with newly diagnosed lung cancer. CMAJ1995; 152: 701-8.

7. Clipp EC, George LK. Patients with cancer and theirspouse caregivers. Perceptions of the illnessexperience. Cancer 1992; 69: 1074-9.

8. Heggenhougen HK. Bomohs, doctors and sinsehs—medical pluralism in Malaysia. Soc Sci Med (MedAnthropol) 1980; 14B: 235-44.

9. Chen PC. Traditional and modern medicine inMalaysia. Am J Chin Med 1979; 7: 259-75.

10. Ooi GL. Chinese medicine in Malaysia and Singapore:the business of healing. Am J Chin Med 1993; 21: 197-212.

11. Eng LS. Cases discharged “A.O.R.” A study of 110cases in Kulim district hospital. Med J Malaya 1968;23: 289-94.

12. Roy RN. Problems of tuberculosis management inSabah. N Z Med J 1972; 76: 97-101.

13. Li YM, Ohno Y, Minatoguchi S, et al. Extracts fromthe roots of Lindera strychifolia induces apoptosis inlung cancer cells and prolongs survival of tumor-bearing mice. Am J Chin Med 2003; 31: 857-69.

14. Hsu YL, Kuo PL, Lin CC. The proliferative inhibitionand apoptotic mechanism of Saikosaponin D in humannon-small cell lung cancer A549 cells. Life Sci 2004;75: 1231-42.

15. Li JH. A study on treatment of lung cancer by combinedtherapy of traditional Chinese medicine andchemotherapy. Zhongguo Zhong Xi Yi Jie He Za Zhi1996; 16: 136-8.

16. Gopal P, Iyawoo K, Hooi Lai Ngoh, Parameswary V.Lung cancer: a review of 589 Malaysian patients. MedJ Malaysia 1988; 43: 288-96.

17. Menon MA, Saw HS. Lung cancer in Malaysia. Thorax1979; 34: 269-73.

18. Liam CK, Lim KH, Wong CM. Lung cancer in patientsyounger than 40 years in a multiracial Asian country.Respirology 2000; 5: 355-61.

19. Hooi LN. What are the clinical factors that affectquality of life in adult asthmatics? Med J Malaysia2003; 58: 506-15.

20. Marin I, Higgins R. (Relationship with the patient withbronchopulmonary cancer. Psychological, family andsocial aspects). Rev Mal Respir 1984; 1: 277-84.

21. Moore C. Making a difference in oncology nursing.Fla Nurse 1999; 47: 15.

SURVIVAL IN PATIENTS WITH NON-SMALL CELL LUNG CANCER OPTED OUT OF CANCER-SPECIFIC THERAPY