“Key to the Future”: British American Tobacco and Cigarette Smuggling in China

Background Cigarette smuggling is a major public health issue, stimulating increased tobacco consumption and undermining tobacco control measures. China is the ultimate prize among tobacco's emerging markets, and is also believed to have the world's largest cigarette smuggling problem. Previous work has demonstrated the complicity of British American Tobacco (BAT) in this illicit trade within Asia and the former Soviet Union. Methods and Findings This paper analyses internal documents of BAT available on site from the Guildford Depository and online from the BAT Document Archive. Documents dating from the early 1900s to 2003 were searched and indexed on a specially designed project database to enable the construction of an historical narrative. Document analysis incorporated several validation techniques within a hermeneutic process. This paper describes the huge scale of this illicit trade in China, amounting to billions of (United States) dollars in sales, and the key supply routes by which it has been conducted. It examines BAT's efforts to optimise earnings by restructuring operations, and controlling the supply chain and pricing of smuggled cigarettes. Conclusions Our research shows that smuggling has been strategically critical to BAT's ongoing efforts to penetrate the Chinese market, and to its overall goal to become the leading company within an increasingly global industry. These findings support the need for concerted efforts to strengthen global collaboration to combat cigarette smuggling.

Summary.-The effect of successive inocula of tumour cells given to rats at intervals of 1 to 10 days was examined.If W256 cells were injected on both occasions, the second inoculum failed to grow if given into the footpad as early as 1 day, or intra- venously as soon as 4 days, after the first administration.However, although a second inoculum failed to grow, it produced significant augmentation of the growth of the primary implant if given during its latent or growth phases.If the second inoculum contained cells from a fibrosarcoma unrelated to W256, its growth was effectively curtailed if the initial inoculum had preceded it by 24 h or more.However, secondary inocula of fibrosarcoma cells did not augment the growth of the primary W256 tumour.
ALTHOUGH tumour-bearing hosts mount immune responses to both autochthionous and transplanted tumours, these responses are inadequate to prevent progressive growth (Alexander and Hall, 1968;  Brunner et al., 1968; Hellstrom, Hellstrom  and Pierce, 1968; Klein et al., 1960;  Mikulska, Smith and Alexander, 1966).Whilst spontaneous regression of a well established primary tumour is infrequent, the host is often resistant both to second- ary implants and to the establishment of metastases from the primary tumour, provided that growth of this is still in progress (Bashford et al., 1908; Deckers  et al., 1973; Gershon and Kondo, 1971).Whilst the influence of a primary tumour on the establishment and growth of spontaneous or artificially induced second- ary tumours has been thoroughly documented, there have been few investigations of the influence of the second implant on the growth of the primary.Yet this effect may be particularly relevant to clinical observations in humans, where an appar- ent change in the growth pattern of the primary tumour could be due to the establishment of metastases.Indeed reports of animal studies suggest that secondary tumours may produce considerable effects on the growth of the primary (Cheshire,  1970; Dewys, 1972; Yuhas and Pazmifno,  1974).
This paper reports the influence of a second challenge with cells from the same or a different tumour on the growth of a non-lethal tumour which had been inoculated previously.The first tumour can be inferred to have induced an immunological response as, after its regression, hosts are resistant to a second implant.The influence of the second challenge on the growth of the first implant, and the extent of immunity of the host towards a second challenge, were examined following re- administration of tumour cells at different sites during either the growth or regression phases of the first implant.

MATERIALS AND METHODS
Rats.Sevento 9-week-old male and fe- male (PVG/c x DA) F1 hybrid rats were used.
In any individual experiment, rats of the same sex and age were used.
Tumours.-The Walker 256 carcinoma (W256) was obtained from Dr M. Cauchi, Monash University, Melbourne.W256 cells were cultured in 250ml Falcon plastic tissue- culture flasks in Medium F15 (Eagle's mini- mum essential medium, Gibco, U.S.A.) supplemented with 10% foetal calf serum and 100 u/ml Mycostatin.W256 cells were harves- ted in Puck's saline containing 0 025% tryp- sin and 0-01% versene.Tumour cells were washed twice in Hanks' balanced salt solution (HBSS) and their viability was established by the use of 0.1% trypan blue in saline.In order to avoid the preferential selection of certain cells under tissue culture conditions, new tumour-cell cultures were set up each month from a stock vial stored in liquid N2- The fibrosarcoma used arose spontaneously in a (Lewis x DA) F1 hybrid male rat and was adapted to the same tissue-culture conditions as the Walker 256 carcinoma.
Tumour inoculation.-Recipientswere an- aesthetized with ether.Footpad injections (0.1-0-2 ml) were given into the subcutaneous tissue in the middle of the foot, the needle being inserted just distal to the heel.The lateral tail vein was used for i.v.injections (1-3 ml).
Tumour growth measurement.-Footpadsize was measured on alternate days after tumour- cell inoculation.Anaesthetized rats were examined in random order using calipers (Dial Caliper, Mitutoyo, Japan) with an accuracy of 0 1 mm.On the first day after the injection of tumour cells in HBSS or HBSS alone, a small footpad swelling was observed.The swelling had, however, completely regressed by the 4th day in the saline-injected controls, the time at which the first experimental readings were taken.At the doses used here for i.v.injection, assessment of tumour growth was an all-or-none pnenomenon, based on the death of the host due to complete replace- ment of lung tissue by tumour, or his survival.

RESULTS
The injection of W256 cells into the s.c.tissues of the footpad initated localized tumour growth.The consequences of injecting a variety of doses of viable W256 cells into the footpad of normal (PVG/c x DA) F1 hybrid rats are shown in Fig. 1.A dose of at least 5 x 107 cells was required to ensure a lethal outcome to tumour growth, following non-reversible metastatic spread to the popliteal lymph node, the leg muscles and finally the lungs.Following injection of non-lethal doses, the tumour growth pattern took the form of a latent period varying in length from 4 to 7 days followed by a growth phase of about 7-10 days, when tumour cells could be seen in the draining lymph node, and then by a regression phase with concurrent dis- appearance of tumour cells from the lymph node.Inoculation of 5 x 105 cells resulted in tumour growth in the footpads of 100% of normal F1 hybrid rats.As tumour growth rate varied with the batch of W256 cells, each experiment included a control group of rats which were inoculated in the foot- pad with W256 cells, but received no other treatment.Tumour growth in the footpads of the rats in the experimental groups was assessed in comparison with tumour growth in these control rats.For this reason no mean values could be calculated using data from experiments done on different days, and in the following para- graphs the results of one characteristic test are given for each experimental protocol.

L
The effects of a second challenge with the same tumour (PVG/c x DA) F1 hybrid rats were injected in the left footpad with 5 x105 viable W256 cells.At various times after this injection, a similar number of W256 cells was administered into the contralateral footpad.The times chosen for the second injection, namely 1, 7, 14 and 30 days, fell within the latent, growth, regression and post-regression phases of the first implant respectively.The results of one characteristic experiment are sum- marized in Fig. 2. In no case did the second inoculum of W256 cells give rise to tu- mours in the footpads of previously chal- lenged rats.However, the second inoculum, if given during the latent or growth phases of the first implant, effected a significant augmentation of the growth of the first implant in comparison with tumour growth in the footpads of rats in the control group.If the second challenge was deferred until the regression phase of the first implant, this tumour continued to decrease in size as in control rats, until the footpad regained its normal thickness.The preceding experiment did not indicate whether the antigenic stimulus provided by the second challenge was sufficient of itself to produce the observed effects, or whether activity was required on the part of the tumour cells in the second inoculum.To clarify this point, the experiments were repeated with the substitution of lethally irradiated (10,000 rad) W256 cells in the re-challenge inoculum.A dose of 5 x 105 W256 cells was used again, and the tumour-bearing hosts were chal- lenged with irradiated cells either 1, 2, 5, 6, or 7 days after the first inoculation.In contrast to the experience with a second inoculum of viable cells, no change was observed in the growth of the first implant at any time.
It was inferred from the preceding 25 experiments that, whereas re-challenge with tumour antigen alone was insufficient 56 to modify the growth of the tumour im- al planted first, the injection of a non-lethal vXe dose of viable tumour cells into the   3.-The effect of a second i.v.inoculum of W256 cells on the growth of a primary inoculum in the footpad.Six groups, each of 5 (PVG/c x DA) F1 hybrid rats, were in- jected in the footpad with 5 x 105 viable W256 cells.At various intervals after the initial injection, inocula of 5 x 106 W256 cells were given i.v.As insufficient W256 cells were available from tissue culture to inject all groups of rats with the same batch, two batches, A and B, each with its own control group, were used.(A) A*-Control group.No. W256 cells injected i.v.* * W256 cells injected i.v.1 day after footpad inoculation.All recipients died.Mean survival time 19-6 i 1-3 days.

O
O W256 cells injected i.v. 3 days after footpad inoculation.One rat sur- vived indefinitely whilst the other 4 died after a mean survival time of 48-5 ± 2-1 days.The mean survival time of rats not injected in the footpad with W256 cells before an i.v.injection was 16*4± 0 4 days.(B) A Control group.* W256 cells injected i.v. 4 days after footpad inoculation.All recipients survived and there was no significant augmentation of the growth of the footpad tumour.marginal protective effect against i.v.injected tumour cells could be demonstra- ted.Tumour-bearing hosts which had been re-challenged i.v. on the 2nd and 3rd days after challenge in the footpad surviv- ed significantly longer than normal rats injected i.v. with tumour cells alone, although only one out of 10 rats survived indefinitely.Tumour growth in the foot- pads of rats which had been re-challenged i.v.during the 3 days after the initial footpad inoculation was significantly aug- mented and progressed until death.However, i.v.injections of 5 x 106 W256 cells given on the 4th or later days after injec- tion of tumour cells into the footpad failed to influence the growth of the initial tumour, and did not kill the recipient.The effects of a second challenge with a different tumour To determine the specificity of the effects observed in the preceding experi- ments, the effect of substituting cells from an unrelated tumour in the second inoculum was examined.The additional tumour used was a fibrosarcoma which had arisen spontaneously in a (Lewis x DA) F1 hlybrid rat, but which also grew well after injection into the footpad of normal (PVG/c x DA) F1 hybrid rats.In the first experi- ments 5 x 105 viable fibrosarcoma cells were inoculated into the footpad of (PVG/c x DA) F1 hybrid rats 1, 2, 3, 5 or 7 days after 5 x 105 W256 cells had been injected into the contralateral footpad.Unless fibrosarcoma cells were adminis- tered within one day of the W256 cells challenge in the contralateral footpad, the second inoculum did not give rise to a visible tumour.There was no modification of the growth of the W256 tumour in these rats at any time, in comparison with tumour growth in the footpads of the control group.
In other experiments, in which a lethal dose of fibrosarcoma cells (106) was injec- ted i.v. at various times after inoculation of W256 cells into the footpad, the growth of the latter tumour was unaffected.In these rats, injection of W256 cells into the footpad afforded complete protection with- in one day against the i.v.administration of fibrosarcoma cells.

DISCUSSION
The mutual influence exerted on the growth of each other by two inocula of tumour cells which had been administered to the same host has been examined.It was found that following the injection of W256 cells into one footpad, the growth of a second, similar inoculum in the contralateral footpad was prevented, even if this was administered within 24 h of the first.If the second challenge was administered by the i.v.route and contained sufficient W256 cells to be lethal if given to a normal animal, systemic tumours developed unless an interval of at least 4 days had elapsed since the initial footpad challenge.As regards the influence of the second inoculum on the progress of the first, significant augmentation of the growth of the initial tumour was observed to follow re-challenge with viable tumour cells provided that these were administered within 7 days, if in the footpad, or within 3 days if i.v.The injection of an un- related tumour failed to influence growth of the initial inoculum of W256 cells in the footpad, although non-specific immunity induced by W256 inoculation prevented growth of the unrelated tumour if injection of this was deferred for more than one day.
The discrepancy between the fates of the secondary inocula of W256 cells adminis- tered via different routes may indicate that after i.v.injection tumour cells were established more rapidly in the lungs than in the footpad with cells injected s.c.In this way, they may have evaded the developing immune response.However, in the absence of information on the response to a range of i.v.doses of tumour cells, it is not feasible to clarify this point.Similarly, no significance can be attributed to the apparently earlier onset of concur- rent immunity directed against the fibrosarcoma in comparison with the W256 cells administered i.v.without the results of injection of a range of doses of the two tumours.
That cross-reactivity should be demon- strable between two tumours of such different morphology and origin indicates that the concomitant immunity observed was very non-specifically based.In con- trast, the augmentation of the growth of a pre-existing tumour as a result of a second exposure of the host to tumour cells was characterized by a greater degree of specificity.Apart from the requirement for identity with the cells of the first inocu- lum, it was necessary for the tumour cells in the second inoculation to be viable for them to modify the growth of the primary tumour.This requirement for viability may reflect a necessity for tumour cell proliferation in, or emigration from, the footpad if the host's immune response is to be influenced.
There are several possible ways whereby the secondary tumour inocula could have influenced the course of the initial tu- mour.Augmentation of the growth of a footpad tumour in a rat bearing pulmonary tumour as a result of i.v.injection of cells is most likely to have been consequent upon general debility of the host.The increased size attained by footpad tumours in rats which had received a further inoculum of tumour cells in the contra- lateral paw, may have resulted from enhancement, as has been well documented in experiments in which tumour cells were injected before tumour transplantation (Kaliss et al., 1953).Alternatively, the selective recruitment of host lymphocytes with the appropriate reactivity away from the primary tumour may have interfered with the host's immune response to it (Ford and Atkins, 1971; Sprent, Miller and  Mitchell, 1971).The limitation of the augmentation to tumour re-challenge with- in a week of the primary inoculum would favour this interpretation.
FIG. 1. Changes in footpad thickness follow- ing the s.c.injection of various numbers of W256 cells.Normal (PVG/c x DA) F1 hybrid rats were injected in one hind footpad with the indi- cated number of viable W256 cells.Each group consisted of 5 recipients.0* 0 105 cells; A A 106 cells; 0-O 107 cells; * * 5 x 107 cells.Following this injection of tumour cells all recipients died.Mean survival time was 18-6 + 3-3 days.
on both the growth of o the first tumour in the footpad and the er survival of the host are summarized in Fig. irt 3. During the first 3 days after footpad challenge with W256 cells, no more than a