March 2009

Mal's Musings

Malcolm A Traill

Updated 15/3/2009






Elsewhere[1], there are presented the measurable changes associated with cancers or their markers from two patients – one with lung cancer (patient LC) and one with prostate cancer (patient PC). Since these presentations, some additional points of interest have been examined, and are now presented:


Aspects for examination :

a)      The serum Lithium levels through the Lithium Rx cycle

b)      The Lithium Rx protocol and renal function

c)      The Lymphocyte subtypes, stimulation and suppression with the Lithium Rx

d)      Short-term Lithium Rx effects on urinary Calcium and urate excretion

e)      NK Cell maturation

f)        NKT (CD56/CD16)+ cells and their significance

g)   Li Rx on stem cells


a)      Serum Lithium over the complete Lithium Rx cycle (as defined[2]). The full cycle is shown in Figure 1 (together with other components – see later).



Figure 1: This is a composite graph derived from points measured during different Lithium Rx cycles. The serum Lithium levels are shown (red, square), with the laboratory day-to-day Standard Deviation (SD).  The stepwise rises from the two boluses on the treatment day is followed by the gradual decline to below the (psychiatric) therapeutically relevant level. Note that the straight lines do not accurately represent the changing serum levels. Above the Lithium graph is the serum creatinine (blue; triangle), with the +SD only shown. Spot urinary Calcium:creatinine, Urate:creatinine and Urate:Calcium ratios show probable cycle-related fluctuations (see later text).  


As before, the points are taken from various cycles. The low point has been estimated more accurately now, and the Standard Deviation for the analytical method used throughout, is shown. At best, the graph can only provide a general guide, showing the type of pattern that may be expected, but with some differences expected in other individuals. Note that the lines joining the points do not reflect, with any accuracy, the actual serum values. The serum levels will follow non-linear relationships.


b)      Renal Function: Lithium is known to be nephrotoxic and can, if the serum values are allowed to be held too high for too long, produce irreversible kidney damage with defects in renal function[3]. This is typically assessed by measuring the serum creatinine level. Creatinine, which is a waste product of muscle metabolism, is excreted by the kidneys. A decline in kidney reserve (function) results in a rising serum creatinine. Since the Lithium Rx is based upon the animal observations that the alternate day regime had effects different from the effects seen with a daily administration of Lithium[4], and the impression is that, in humans, the alternate regime also seems to be more effective for certain condition (eg rheumatoid arthritic, cancer), there could be concern that the Lithium Rx protocol could have an unexpectedly more damaging effect upon the kidneys. This possibility was examined on a short-term and medium-term basis.

                             i           The short-term creatinine values found at different points through a Lithium Rx cycle are shown in Figure 1. There might be a slight but seemingly reversible rise in the serum creatinine during the stages when the serum Lithium was in what is usually considered to be the psychiatric therapeutic range (0.4 - 1.0 mmol/L)[5].

                           ii            The long-term creatinine values measured at the low points of the Lithium Rx cycles at various times are shown in Figure 2 :




Figure 2: Baseline (red; triangle) and subsequent serum creatinine changes after the commencement of Lithium Rx (blue; square). The Baseline test was performed by a laboratory not used later. The slight rise from baseline may be due to the different laboratory involved or the low level of Lithium still present at the lowest points in the subsequent cycles. The rise following the prostatectomy surgery may be a consequence of surgical and anaesthetic-related factors. Thereafter, the level was stable, and still within the error limits of the baseline measurement.


Unfortunately, the baseline serum creatinine level was measured by a laboratory that was not used thereafter. This makes subsequent comparisons difficult. There may be a slight deterioration of the renal function shown after the surgery (open prostatectomy). Thereafter, the creatinine levels are unchanged. Even with the different laboratories, the change in creatinine levels from the reading of laboratory A would come within the 1 Standard Deviation ranges for the day-to-day testing (not having regard for the patient day-to-day variation), meaning that there is no clear change, given that the low level of Lithium in all the specimens for laboratory B may have produced a small impairment of the temporary type (see i. above). Whilst there is the possibility that the slight rise after the surgery could be Lithium Rx related, it is much more likely to be due to the operation and associated factors (eg blood loss, hypotension, hypoxia, aminoglycoside antibiotics and anaesthetic agents) upon the kidneys, together with the inevitable mild degree of age-related nephrosclerosis. Thereafter, the creatinine levels were stable. From all of this, the tentative conclusion is that, given peak levels of Lithium below 0.8 mmol/L, a patient with essentially normal renal function for his age and within the time frame of seven months of continuous Lithium Rx, there is currently no clear evidence that there has occurred any permanent progressive renal impairment due to the Lithium Rx. Outside those qualifying features, further studies would be required for assurance. 


c)      Lymphocyte subtypes. Since the earlier report, a final lymphocyte subtype assessment was performed. The new graph is shown in Figure 3.




Figure 3: The lymphocyte and NK cell counts over time are shown. Both CD8 (blue) and CD16/56 (NK cell; green) show convincing rises after the initiation of the lithium Rx, only to have more convincing falls after the prostatectomy. The CD4 (red) and CD19 (B cells; lavender) show an initial fall, then a rise to match the CD8 and NK cells, and a less marked fall after the prostatectomy. All bar the CD8 levels continued to fall over the ensuing months.


 The cell counts at the key stages are shown in Table 1 :

                                                                                                               TABLE 1

                                                  Subtype feature       Baseline          Pre-surgery    Last test           Reference Range

CD4 “helper




650-2000 cells/μL

CD8 “suppressor




330-1310 cells/μL

CD19 “B cell




190-550   cells/μL

CD16/56 “NK cell




130-540   cells/μL

                                                                 Those cell types with counts below the lower limits of the quoted reference ranges are highlighted in blue


Significant points are :

                             i            As shown earlier, the Lithium Rx was associated with a gratifying rise in the counts of CD8 and NK cells in particular. The counts then fall appreciably after the prostatectomy (the removal of the cancer). From this, the conclusion follows that the stimulus factor(s) provided by the Lithium Rx needed some additional co-stimulatory factor(s) from the cancer for there to be an effective stimulation; remove the cancer, and the stimulus ceases.

                           ii            Continuation of the Lithium Rx post prostatectomy did not bring about a recovery; rather, there is a continuing (but slight) decline in the counts of all cell types with the exception of the CD8, which rose mildly.

                          iii            For all cell types, the final cell counts were less than those at the initial baseline.

                         iv            For the cell types CD19 and CD16/56 (in particular) there was a steady fall in counts, below the initial baseline levels and consistent with an immunosuppressant action.

                           v            Both CD4 and CD19 subtypes ended up with cell counts below the lower limits of the quoted reference ranges.

                         vi            The graph illustrates the enigmatic duplicity of Lithium treatments, as noted by Levine & Saltzman, namely, an immune stimulatory role [requiring cofactor(s)], and an immune suppressant role which (presumably) suppressed experimental allergic encephalomyelitis (EAE).

                        vii            Without knowledge of what the patient’s pattern pre-Lithium Rx had he not had a tumour, the possible immune suppressive action of the Lithium Rx cannot be assessed accurately. Laboratory experiments on animals may be required to clarify this aspect.



Upon examination of the graphs, there appear to be similarities between the CD8 and NK cell points and lines, and between those for the CD4 and CD19 cell types.

The relationship between the CD8 & NK counts can be expressed[6] :

CD8 = 124.9 + 1.43 x CD16 ……………. 1

with the astonishing correlation coefficient of 0.97254 ! This is consistent with the Lithium/tumour stimuli affecting both cell types virtually identically – presumably, this would have to be in lymph nodes, which are considered to be the likely sites for NK cell maturation[7]. Whilst the right and left obturator lymph nodes (20x8x8 mm and 25x8x8 mm respectively, deemed tumour-free, and fat, with probably smaller lymphoid foci) were removed at surgery, the possibility that the loss of these would result in the lymphocyte subtype changes, whilst possible, would seem most unlikely, given that there were probably many more nodes and lymphoid foci in the internal and external iliac groups that drained the pelvis and prostate.

The relationship between the CD4 and CD19 cells is not as clear-cut, being :

                        CD4 = 370 + 1.31 x CD19 ……………… 2

            with a correlation coefficient of 0.751761. So, both factor groups are probably involved, but less simply &/or directly.

            Clearly, the behaviour of the lymphocyte subtypes in cancer patients receiving Lithium Rx would seem worthy of further study to confirm and extend understanding of the immune mechanisms.


d)      Purine and calcium metabolism. Spot urinary measurements of creatinine, urate/uric acid and Calcium seem to be sensitive, cheap and non-invasive indicators of tumour responses (not necessarily specific – “happenings”). These can be of value particularly when the tumour has no tumour marker. Even if there is a tumour marker, the marker is usually a glycoprotein formed in the endoplasmic reticulum, transported through the Golgi apparatus, and then exported from the cell. As such, the responses shown by glycoprotein tumour markers can show fluctuations in response to therapy difficult to interpret in the short term; typically, following an episode of tumour destruction, a flood of tumour marker from the damaged tumour cells reaches the bloodstream and causes a rise in the marker level, later reversing into a fall as the damaged tumour produces less and the glycoprotein is catabolized. With the unusual effects of the Lithium Rx witnessed upon different conditions, the urinary Calcium and urate, in relation to the urinary creatinine (as the reference chemical), were measured at various times during the Lithium Rx cycle. These samples were collected whilst within different cycles post prostatectomy. No result has been duplicated or confirmed – they are presented for mental stimulation and perhaps for further examination. The results are included within the Figure 1. The noteworthy feature is the drop in the Calcium:creatinine and Urate:creatinine ratios at about the Lithium peak. This is followed by a fall of the Urate:Calcium ratio the next day. Reliable explanation for these troughs is not easy to provide at present. Accordingly, when using such a spot urine measure for cellular damage and evidence of some clinical effect (“happening”), the spot urine collection would best be taken at the lowest Lithium level.


e)      NK cell maturation. Each lymphocyte subtype study produced what is known as a Scatter-plot, being a graphical representation of the numbers of cells being decorated by (usually) antibodies, which sorts them into clusters. The peripheral blood NK cells are typically mainly CD56dull/dim, CD16+. Views on the relationships with CD56bright have varied (see later), but in the last few years, the prevailing evidence points to those CD56dim cells being the mature forms and those CD56bright cells being less mature forms, as found in lymph glands[8]. The reagent used by the laboratory[9] involved with the NK testing is reported to contain a mixture of anti-CD16 and anti-CD56 antibodies.


f)        Natural Killer Cells (NK) have no T Cell Receptor (TCR; CD3) and have CD16 &/or CD56 surface receptors. CD56 can be further subdivided into the CD56bright (~5-10%, with CD16dim/-) and CD56dim (~90-95%, with CD16high) cells[10] which were previously regarded as having differing origins, having differing functions, with the CD56bright able to transform without division to the CD56dim upon activation, as by contact with the FGFR1 fibroblast receptor[11]: the CD56bright being considered primarily responsible for most immuno-regulatory cytokine (eg IFN-γ) responses, the CD56dim more with effector cytotoxicity functions with granzymes and perforin. The final step in maturation is marked by the acquisition of killer Ig-like receptor (KIR).

      PC's Scatterplots are presented below :

                                                                                                    FIGURE 4


 Fig. 4a Baseline                 Fig. 4b After start of Rx      Fig. 4c Preoperative            Fig. 4d Post-Op.                 Fig. 4e Final Post-Op.


Figure 4 a-e: These Scatter-plots show (upper row) the intensity of CD56/CD16 decoration  (vertical) against CD3 decoration (horizontal). The lower row show CD56/CD16 (vertical) against CD19 (B cells). They are for routine clinical use, and one has still not been supplied after 3 months of repeated asking. One may wonder about the possibility of different settings etc. The inks used were different for Fig. 4c and different colours represent computer-generated differentiation, with NKT orange in a, b,  & e, but green in c, with no clear colour distinction for those in rectangles C2 and C4. The grey indefinite features across the centre of the plots are fold-lines in the paper.  Interest can be concentrated on the scattered points within rectangle C1 if each upper panel, and H1 of the lower panels. With the mixed antibody NK-decorating combination in the upper row, there can be little more to say other than that there seem to be changes that may worth pursuing - the (vertical) scatter was wide (a), it then became slightly tighter and the mode moved lower (b), becoming tighter still (c), then it  moves higher, with a more open lower portion (d),  with a slight widening (e).


These observations of change are non-specific but expected. The changes probably indicate changing affinities of the cells to the antibodies. Using separate and specific antibodies on specimens taken from patients with and without tumours, and from those who receive the Lithium Rx protocol, clearer evidence as to what occurs within the NK cell subset would be expected.

Examining the plots above, the main bulk of points corresponding with the NK cells (in rectangle C1) forms a cluster between vertical scale readings 1-1.05 x 101 and 100 for plots a, b, d, & e (lower for plot c, which was processed/printed at a different time). Based upon the information above, there is the likelihood that the CD56bright (vertical scale >1.05 x 101) will be relatively uncontaminated by CD16 readings. At the baseline testing, the number, whilst present seems quite small (++). At the first post Lithium Rx test, they have almost disappeared (±). At the pre-operative test there may be a mild increase  (+), and again at the first post operative test, there appears a maintenance of numbers (+), but at the last test they have almost disappeared again (±). This pattern, in the context of rising NK cell numbers in the peripheral blood, would be consistent an activation-induced depletion of CD56bright cells in a shift to CD56dim cells, with a later reactionary increase until inflammation has died down.

There is not very much that can be deduced from this, apart from the existence of changes that may be worthwhile studying in more detail and with more specific antibodies, using other cases. The general trends of the NK CD56bright cell numbers seemed to follow the inverse of the other main lymphocyte subtypes with the Lithium Rx initially, then increasing preoperatively, with a later fall.

Within the main cluster that presumably represents the bulk of CD56 and CD16 cells, there are subtle changes: throughout, the cluster appears unimodal, yet the baseline test shows a wide vertical scatter which, on subsequent tests, narrows and the mode generally shifts downwards slightly. These changes might reflect a stimulation-induced decrease of the CD56 antigen, although the contribution of the anti-CD16 in the mixed antigen assay is unknown.





Figure 4. Graphical representation of the cell counts per μL for the major lymphocyte subtypes and NK cells against time from baseline to well past the operation.  All cell types show a noticeable increase associated with the Lithium Rx pre-operatively.  The level for the NKT cells shows a slower fall, consistent with operation-associated effects. The appreciable rise in all counts immediately pre-operatively may be consistent with some additional stimulus to proliferation caused by release of one or more factors from the cancer cells, which were probably starting to develop degenerative changes at or about that point.


(g) NKT cells                                                                               NKT details :

                                                                           Values related to the graph above, with NKT being (CD56/CD16)+ CD3+ :

                                                                           Baseline              Lithium Rx 1st        Pre-operative      1st Post-operative  2nd Post-op.

Total Lymph.

1430 cells/μL

1510 cells/μL

Rise of 105.6%

1850 cells/μL

Rise of 129.4%

1140 cells/μL

Fall to 79.7%

1230 cells/μL

Fall to 86.0%

Total CD3

991 cells/μL

1079 cells/μL

Rise of 108.9%

1352 cells/μL

Rise of 136.3%

838 cells/μL

Fall to 84.6%

882 cells/μL

Fall to 89.0%

NKT% of CD3



Rise of 1.09%


Rise of 304.5%


Rise of 204.5%


Fall to 45.5%

NKT cells/μL

26.8 cells/μL

32.4 cells/μL

Rise of 120.9%

90.6 cells/μL

Rise of 338.1%

75.4 cells/μL

Rise of  281.3%

18.5 cells/μL

Fall to 69.0%

NKT% of Total Lymph.



Rise of 115%


Rise of 262.0%


Rise of  353.5%


Fall to 80.2%

(Rises and falls are all related to the baseline readings.)


PC’s NKT ([CD16/CD56]+ CD3+)  counts were 18.5 – 90.6 cells/μL, and showed an impressive rise before the operation, with a level that carried over into the first postoperative testing, not dropping rapidly like all the others. Tests for anti-nuclear factor and double-stranded DNA binding were non-reactive in the first post-operative sample. The TSH on the same date was within the reference range. Sadly, tests of plasma proteins, other than albumin, were not preformed. Based upon the histological appearance of the cellular degeneration shown by the tumour, the estimate was that it probably became (microscopically) apparent some weeks before the prostatectomy. That would be close to the time of the pre-operative blood sample, with the peak values of all lymphocyte subtypes and NK cells. Further study will be required to test the possibility that the appreciable rises in the lymphocyte subtype counts ties in with the tumour degeneration – with the latter possibly being a stimulus to cell proliferation/mobilization.


                                                         Correlations : -                                                   

                                                                            NKT (CD16/56)+ to:       CD4                        CD8                        CD19                       CD16/56

Overall figures





Preoperative data






The response to Lithium Rx + tumour of the NKT cells most closely followed the CD19 (B cells) overall. Preoperatively, the NKT levels most closely followed the CD4 T cell subtype. 


Brief Review of NKT cells

T cells can express NK cell receptors, and some may be induced to do so, especially after activation[12]. Most NKT cells in humans are CD4+, CD8+ or double negative and many have an invariant TCR receptor (iNKT[13]), Vα14i (mice) or Vα24i (humans)[14]. Peripheral blood relative NKT (CD3+ CD56+) counts may be raised in subacute-chronic inflammatory conditions, such as pulmonary tuberculosis with NKT cells ~6.8% of gated lymphocytes, compared to ~3.5% in negative controls, and a fall after treatment to ~3.4%, at which time the circulating IFN-γ was significantly elevated[15]. Intravenous infection of mice with Mycobacterium bovis (Bacillus Calmette Guérin; BCG)[16] resulted in an elevated peripheral blood count of iNKT cells, maximum at 8 day, but still appreciable at 30 day. The conclusions were that these cells governed the number and extent of granuloma formation, possibly by modulating TNF-α, thereby preventing excessive and damaging inflammatory responses. However, in HIV sufferers with late disease, CD3+ CD16/56+ NKT cells were not significantly altered in numbers, but the numbers did correlate weakly with plasma HIV RNA. NKT cells were increased in Salmonella bacteraemia sufferers and in those (total of groups ?) who died, a higher proportion of NKT cells producing IFN-γ[17]

CD3+ CD8+ CD56+ cells comprise some 4-30% of the human CD8+ population and direct cytolytic activity is linked to these cells. Their percentage in an individual can vary by x3 in a few days[18]. The NKT cells defined by (CD16/CD56)+ CD3+ CD8+ found in broncho-alveolar lavages are appreciably increased in numbers in patients with hypersensitivity lung disease. The level is unrelated to the state of disease activity[19], and NK cells did not show such changes. The invariant NKT subtype was poorly represented, raising doubts about its clinical value. (There was no tested correlation with peripheral blood levels.) The typical range of the specifically defined subtype Vα24+ Vb11+ iNKT cells in human peripheral blood was ~0.1 – 10 cells/μL[20] (estimated from their Figure 1 A & B), or, for the more heterogeneous group of NKT (CD56+ CD3+), 0.5-6% of total lymphocytes[21].  The iNKT comprise 0.01-1% of total blood lymphocytes or, ~½ median NKT (CD56+ CD3+). In mice, they can be activated and be destroyed within hours, taking ~3 day to recover[22]. They are typically decreased in patients with autoimmune conditions (eg SLE[23]) involving tissue lymphocyte infiltration and destruction, and tend to show, as numbers (%) an inverse relationship to total serum IgG. There is one notable exception, the uveitis of Behçet’s syndrome, in which the CD3+ CD56+ NKT cell counts were raised above those from other causes of uveitis and unrelated to disease activity[24]. An early report of iNKT in insulin-dependent diabetes mellitus patients and relatives, indicated that there were lower numbers of iNKT cells in sufferers and relatives, and that there was a tendency for stimulated iNKT cells to produce IFN-γ only, without IL-4, showing Th1 behaviour[25]. NKT activity, as well as numbers, may be decreased in SLE. Most NKT are concentrated in the thymus, then the liver and spleen (low level) by homing, with very low levels in the lymph nodes. (Murine NK1.1+ T cells are at low levels in the spleen.) CD4- iNKT are reduced in MS when in remission[26]. The murine model of hepatitis by Con A involves NKT cells releasing FasL, with the NKT cells and hepatocytes reacting to FasL/Fas[27].

They are favoured by CD1d double positive thymocytes, together with a special cytokine mix. Both NK and iNKT share the requirement for T-bet[28] in their development and peripheral homeostasis, but with cell line differences; the immature iNKT favouring IL-4 over IFN-γ and lacking CD122, the receptor for IL-15, which controls numbers. T-bet[29] and GATA-3[30] (which may act earlier) are necessary for maturation. The mature cells express inhibitory receptors of the Ly49 family.

The Vα14 invariant receptor tends to respond to glycosphingo-ceramides (or similar).

The lack of iNKT results in a reduced ability to clear some bacteria[31].  Mice infected with live Borrelia organisms (Lyme disease) showed an activation of the Vα14i NKT at 7 day, which had largely abated by 14 day[32]. The iNKT cells contributed to bacterial clearance, but were not a feature in the tissue inflammation. They may play a part, together with Th17 cells, in the modulation by gut flora upon EAE induction in mice[33], and may suppress induced uveitis by an INF-γ related inhibition of Th1 and Th17 adaptive response[34]. iNKT are attracted, together with neutrophils, to bacterial-induced acute inflammation in mice, in a process that is not antigen-driven, and which is independent of the Vα24+ receptor[35]. iNKT numbers in periodontitis and gingivitis also tended to be close to, or higher than the levels in the subjects’ peripheral blood[36]. A lipopolysaccharide sensitizing dose in mice causes macrophages to release IL-12, which then promptly triggers iNKT cells to release IFN-γ as a priming step in the Schwartzman shock reaction[37].

Activated iNKT cells can inhibit (as by IL-10 production[38]) or accentuate inflammation by rapid (~ hour) cytokine production, the polarization being determined by the ligand and other factors, not well understood. They can stimulate NK cells to produce IFNγ[39], which can be cancericidal (and in the presence of Lithium, this effect may be augmented).

Advanced human prostate cancer patients showed appreciable depression of Vα24+ Vb11+ NKT cell numbers, with partial recovery when in remission[40]. In vitro stimulation revealed reduced proliferation, IL-4 production but reduced IFN-γ production (restored by IL-12) and a very low IFN-γ/IL-4 ratio. Overall there was a shift to a Th2 (immune suppression) role.

Both normal controls and cancer patients in groups tended to show decreasing peripheral blood counts of iNKT. Except for a few notable exceptions, in vitro α-galactosylceramide (α-GalCer, a CD1d-associated CD3 activator for NKT cells) stimulation of iNKT cells from cancer suffers was reduced. The exceptions provided expansions that exceeded those in the control group, with no clear explanation. Radiation therapy reduced blood counts but chemotherapy showed no clear pattern[41].  

Human cancer patients generally tend to have low Vα24+ Vb11+ NKT subset counts in the peripheral blood. When α-GalCer was administered intravenously, there was a rapid fall in the count, and variable cytokine release. As a potential treatment for cancer, it was disappointing[42].

But when α-GalCer-stimulated monocytes were administered intravenously into human cancer patients in an initial dose and booster regime[43], some patients (eg KS204 & KS303) showed a distinct and prolonged rise of the Vα24-Vb11 NKT subset, lasting over 28 day. This was associated with an increase in cytotoxicity, as assessed in one patient with a rise (KS204) and one with a lesser cell-count rise (KS202). (In patient KS202, there was an NK cell count rise after the booster dose, lasting some 11 day.)

Examination of the published[44] mean counts for different cells derived from the testing of the blood of groups of sufferers of Trypanosoma cruzi at different stages of disease progression produced some interesting correlations:


NKT subtype and Treg association in T.  cruzi


                                                                                                 Percentages relating to total CD3 

                                              Phenotype criteria                       NI                    IND                  CARD            DIG    

Treg, CD4+ CD25high





Total NKT CD3+ CD16+/– CD56+/–





CD3+ CD16+ CD56





CD3+ CD16 CD56+





CD3+ CD16+ CD56+





                                                              NI = normal individual, IND = indeterminate stage of disease

                                                              CARD = Cardiac enlargement, DIG = digestive system enlargement


                                                                                               Correlation coefficients

Treg to total NKT


Treg to CD16+ CD56


Treg to CD16 CD56+


Treg to CD16+ CD56+


 As derived from these group contexts, the CD3+ CD16 CD56+ may be a surrogate estimate of Treg numbers, with the total NKT cell numbers a fair second.


Lymphocyte Homeostasis. PC’s graphs show a generalized proliferation of T, NKT and NK cells associated with the lithium Rx, with the response falling away dramatically post prostatectomy for all but the NKT subtypes. The NKT subset may have had its count partially maintained post-operatively as a reaction to subacute/chronic inflammation associated with the surgery (eg inflammation around sutures, the drain tube track etc.) in association with the Lithium Rx.

In that Lithium can stimulate the production of TNF-α RNA transcription and TNF-α release, part of the response (at least) may be attributed to TNF-α acting

through the TNF-R2 receptor[45]. In mice, activated CD8+ cells are selected into the liver and undergo apoptosis under the influence of TNF-α[46], thereby

controlling the cell counts in the peripheral blood. Given that the main lymphocyte subtypes, in particular CD8+ and NK cells, have shown high degrees of correlation throughout the Lithium Rx study, there is no basis to support a liver elimination pathway of CD8+ cells and such a route does not appear to contribute appreciably to CD8+ homeostatic control in this study. The involvement of TNF-α is of particular interest, because Lithium is reported to stimulate a rapid and prolonged TNF-α release from human cultured monocytes (active at 3 mmol/L, inactive at 1 mmol/L)[47], and augment its cytotoxicity[48]. Again, in the pre-operative phase, the lithium Rx was associated with an overall, appreciable, peripheral cell count rise of all subtypes of lymphocytes and the CD16+ CD56+ NK cells.

The only common feature in the setting of the peripheral blood cell counts for the generation of all types is the bone marrow, meaning that the responses obtained here are consistent with one or more tumour (humoral ?) factor(s) coming from the prostate cancer plus the Lithium Rx protocol affecting the basic bone marrow proliferation setting. A very simplistic diagram illustrating the cell derivations is provided below.  Postoperatively, the counts for all cell types fall ultimately, despite the continuing Lithium Rx, consistent with the loss of some cancer-derived stimulus. The pre-operative peak for all cell types may reflect an additional stimulus boost from the early degenerative changes later identified in the surgical specimen.

                                                                                     FIGURE 5























                                            NK cells         B cells            NKT cells     CD8 T cells   CD4 T cells  



(h)    Lithium Rx - No side effect identified. There have been no clinical or laboratory features to suggest a failure of normal stem cells – no haematological, gastrointestinal, skin or mucosal abnormality, and no appreciable weight loss. Experimentally, blocking the Wnt pathway also seemed to have no appreciable effect on stem cells[49].


  Click to E-mail me with concerns about  CANCER, RHEUMATOID ARTHRITIS,
                                                                                    LUPUS ERYTHEMATOSUS,

                                                                                    MULTIPLE SCLEROSIS,
                                                                                    SARCOIDOSIS, HEPATITIS C, etc

                                                                                    INVESTIGATIONS, DIAGNOSIS & TREATMENT                      





Malcolm Adams Traill

Copyright © MA Traill March 2009


[1] Chapters “Prostate Cancer & Lithium – a Therapeutic Renaissance ?” and “Your Cancer & Lithium Treatment.”

[2] Ibid

[3] McCann SM Daly J et al. Ulster Med J. 2008; 77(2):102-105

[4] Levine A & Salzman A. Immunopharmacology 1991; 22:207-214.

[5] Martindale. 33rd edition, 2002. Editor SC Sweetman. Pharmaceutical Press, Great Britain. p. 296.

[6] Microsoft Windows XP 2003. Excel formulae/processing

[7] Caligiuri MA. Blood 2008; 112:461-469

[8] Cooper MA Fehniger TA et al. Blood 2001; 97(10):3146-51

[9] Austin Pathology, Heidelberg, Victoria, Australia, which was responsible for the slow, piecemeal and incomplete delivery of documents with folds through important plots, varying inks, handwritten highlights and different analysis settings.

[10] Romagnari C Juelke K et al. J. Immunol. 2007; 178:4947-4955

[11] Chan A Hong D-L et al. J. Immunol. 2007; 179:89-94

[12] Wingender G & Kronenberg M. Am. J. Physiol. Gastrointest. Liver Physiol. 2008; 294:1-8

[13] Kronenberg M & Rudensky A. Nature 2005; 435:598-604

[14] Kronenberg M. Annu. Rev. Immunol. 2005; 26:877-900 

[15] Barcelos W Martins-Filho OA et al. Microbiol. Immunol. 2006; 50(8):597-605

[16] Dieli F Taniguchi M et al. J. Immunol. 2003; 171(4):1961-8

[17] Jason J Buchanan I et al. J. Infect. Dis. 2000; 182(2):474-81

[18] Pittet MJ Speiser DE et al. J. Immunol. 2000; 164:1148-52

[19] Korosec P Osolnik  K et al. Chest 2007; 132:1291-7

[20] Van der Vliet HJ von Blomberg BM et al. Clin. Immunol. 2001; 100(2):144-8

[21] Green MR Kennel AS et al. Arthritis Rheum. 2007; 56(1):303-10

[22] Takeda K Hayakawa Y et al. Proc. Natl. Acad. Sci. USA. 2000; 97(10):5498-503

[23] Wither J Y-C Cai et al. Arthrit. Res. Ther. 2008; 10:R108 

[24] Yato H Matsumoto Y. Br. J. Ophthalmol. 1999; 83:1386-8

[25] Wilson SB Kent SC et al. Nature 1998; 391:177-81. Nature 1999; 399:84

[26] Araki M Kondo T et al. Int. Immunol. 2003; 15(2):279-88

[27] Takeda K Hayakawa Y et al. Proc. Natl. Acad. Sci. USA. 2000; 97(10):5498-503

[28] Townsend MJ Weinmann AS et al. Immunity 2004; 20(4):477-94

[29] Pai SY Truitt ML et al. Proc. Natl. Acad. Sci. USA. 2004; 101(7):1993-8

[30] Townsend ML Weinmann AS et al. Proc. Natl. Acad. Sci. USA. 2004; 20(4):477-94

[31] Tupin E Kinjo Y et al. Nat. Rev. Microbiol. 2007; 5:405-17

[32] Tupin E Benhnia MR et al. Proc. Natl. Acad. Sci USA. 2008; 105(50):19863-68

[33] Yokote H Miyake S et al. Am. J. Pathol. 2008; 173(6):1714-23

[34] Grajewski RE Hansen AM et al. J. Immunol. 2008; 181(7):4791-97

[35] Mempel M Ronet C et al. J. Immunol. 2002; 168:365-71

[36] Yamazaki K Ohsawa Y et al. Am. J. Pathol. 2001; 158(4):1391-98

[37] Dieli F Sireci G et al.J. Exp. Med. 2000; 192(11):1645-51

[38] Sonoda K-H Faunce DE et al. J. Immunol. 2001; 166(1):42-50

[39] Hammond KJ & Godfrey DI. Tissue Antigens, 2002; 59(5):353-63

[40] Tahir SM Cheng O et al. J. Immunol. 2001; 167(7):4046-50

[41] Crough T Purdie DM et al. Br. J. Cancer 2004; 91(11):1880-6

[42] Giaccone G Punt CJ et al. Clin. Cancer Res. 2002; 8(12):3702-9

[43] Nieda M Okai M et al. Blood 2004; 103(2):383-9.

[44] Vitelli-Avelar DM Sathler-Avelar R et al. Scandinavian J. Immunol. 2005; 62:297-308

[45] Tartaglia LA Weber RF et al. Proc. Natl. Acad. Sci. USA. 1991; 88:9292-96

[46] Murray DA & Crispe IN. J. Immunol. 2004; 173(4): 2402-9

[47] Kleinerman ES Knowles RD et al. J. Leukoc. Biol. 1989; 46:484-92

[48] Beyaert R Vanhaesebroeck B et al. Proc. Natl. Acad. Sci USA 1989; 86:9494-8

[49] DeAlmeida VI Miao L et al. Cancer Res. 2007; 67(11):5371-79